BuildLog.Net - Document your builds!


Buildlog Title: Blackhole 60W Laser

newest first oldest first
Builder: iGull
Member Since: 2010-12-06

Monday, March 26th 2012 - 8:12 PM

Steve/Dave
Many thanks for the comments/suggestions.

To be honest, the delay after finish should really be a software thing on the DSP. However, my exhaust system is actually an inline fan extractor system - sans timer. I can just get the timer for the unit and set it to 20s or so - then it will hang on for 20s after switch off (like in the bathroom :D ). By the time I wire up a 555 and relay or similar to do the job, it would cost more than getting it on ebay sadly - such is the way of things.

On the refrigeration front, I have a nice radiator that was plucked from a YAG laser system - unfortunately, the pump and PID controller were broken.
I have a hole cut to fit a 120mm fan in the rear of my cabinet and I'm sure that it would be fine with that. Some extended tests done by a friend with his laser system suggest that keeping control of the water temp keeps the laser cutting/engraving levels much more consistent and I think I'll go down that route. He uses an external Chinese style cooler (refrigerated rather than air cooled) - that is probably the best bet, but not in the spirit of opensource LOL

Cheers

Neil

add commentadd comment in the forum

reader comment Comment from: steves on Sunday, March 25th 2012 - 3:08 PM
I installed a SS relay to the exhaust fan today as the DSP can control that now - much nicer than having it on all the time (although it would be good to have it stay on for 20s or so after cutting to clear the last of the smoke before opening the door).


Greolt

I'm just starting to get back into electronics so I can't give you a full solution, but I think a simple timer circuit based on a 555 timer would give you the delay you're after. The DSP would signal the 555 which would signal the SS relay.

There are probably even easier ways, but that's the first that comes to my mind.

Steve
reader comment Comment from: daveczrn on Sunday, March 25th 2012 - 2:57 AM
i really like how the extruded aluminum shows with the covers, looks really unique.

As far as a cooler goes to cool the water. I took apart a mini refrigerator once that was designed for a car. had a cigarette lighter adapter plug on it. It had a Peltier cooler in it. Basically it had a heat sink on one side with a fan blowing on it and the other side mounted to the base of the mini fridge. the fridge side got really cold, and the other side with the heat sink got really hot. i think it would work exceptionally well at cooling off the water. only issue with it was it was very inefficient and kinda power hungry.

Saturday, March 24th 2012 - 8:13 PM

It's been a wee while since I did any updates on the buildlog, so I took a few pix today ...

P1010050.JPG
Electronics Drawer

P1010051.JPG
General View

P1010048.JPG
DSP LCD Mount


The main update recently has been the installation of the Lightobject 2012 DSP controller. I have been using the Lasersaur controller since I started the build, but I really needed to get this machine making some pocket money, so opted to purchase the DSP. I have to say, even though it's PC based (I'm all Mac really), the controller and LCD interface have been exceptionally good - it really runs well (ethernet and USB both). I'm having a few hiccups with the LaserCAD software (it's not Adobe Illustrator is it :-)) - the interface could really do with an overhaul and get up to the standard and features that other similar laser control software systems have (LaserWorks etc). They all look much the same, but there are definite differences in the operating quality.
I'm running LaserCAD using Virtualbox (free) on the Mac - it's running Windows XP in a solo window - works very well and integrates with the rest of the Mac OS without having a complete XP screen. I use Illustrator as the main graphics app, but have many different graphics and CAD apps on the Mac. Again, a few issues with importing, it doesn't like illustrator files, but the DXF output is a workaround that works - I believe there is an update to improve the illustrator importer (like I said, it falls behind other similar apps). Overall, I'm VERY pleased with it's operation - I can actually do raster engraving now - even producing a Mayan Calendar LOL.

I installed a SS relay to the exhaust fan today as the DSP can control that now - much nicer than having it on all the time (although it would be good to have it stay on for 20s or so after cutting to clear the last of the smoke before opening the door).

That's the laser cutter pretty much complete now - although I may install a refrigeration unit to keep the water at a constant for cutting/engraving consistency. I wonder if there is an open source refrigeration system ? I guess an air cooled system could be made to work in this country as it's never that hot in summer :D :D

Cheers

Neil

add commentadd comment in the forum

reader comment Comment from: educa on Tuesday, January 17th 2012 - 7:59 AM
Here is another (also handy) image which I just received from cncoletech.

By the way, the laser head comes with a nozzle matching the focus lengt you ordered, but if you order another lens they send the extra (other version) of the nozzle with it to match

Image

Tuesday, January 17th 2012 - 7:22 AM

Yes, that is all correct.

Useful graphic to have too.

Cheers

Neil

add commentadd comment in the forum

reader comment Comment from: educa on Monday, January 16th 2012 - 10:26 PM
Ok so now to be 100% sure, can you please tell me if what I say below is 100% correct?

Image


1) laser light enters the laser head at point A

2) At point B it bounces on the mirror and turns 90° downward

3) At point C it gets to the focussing lens where the light begins to colimate

4) At a distance of 50.8mm awat from the lens (I use a 50.8mm lens) there will be the focal distance and also the focal point to burn.

5) Point F is the tip of the nozzle and this tip lies between points C and D, where the focal point will always be the same distance away from this tip (downward)

6) If you use a gauge block then this block has the hight of the distance between F and D ?

7) So if I put a plate of for example 4 mm onto the table (which does not move) then all I actually have to do is put the gauge block on my plate , loosen screw E so the nozzle goes up and down. Position the laser head above my gauge block , lower the nozzle until it touches the gauge block and then tighten screw E again ? When I then remove the gauge block I will always have the correct focal distance (where focal point will be at the top of my material ?

8) Suppose I want to have focal point in middle of material (4mm thick material) , then I just use a gauge block which is 2mm less high ?



Is this all correct? If it is (which I hope) then this would be sooooo nice. I would not need moveable Z axis on the machine then.
reader comment Comment from: Greolt on Monday, January 16th 2012 - 9:55 PM
Yes I am with Neil on this one.

A fixed table would be a big saving in complexity and size of the machine.

I mostly built a moving table so that I can fit a rotary axis under the laser head.

30 plus mm of adjustment in the head itself would cover many different material thickness.

Greolt
reader comment Comment from: educa on Monday, January 16th 2012 - 9:48 PM
And if I don't use a moveable Z table, what distance should I then best leave open below my machine?

I could make the machine very with little height, but I suppose if I leave too little space below it, then the laser can reflect on the bottom and bounce back up ?
Or can I cover the ground of my machine with some kind of material which absorbs the light/energy somehow ?

If I can make the machine fixed height, then this would dramatically ease the way how I need to make it + reduce the height of the machine at all.


So any advice on how much space I oucht to leave on the bottom + what material I best place at the bottom ?

Kind regards,

Bart

Monday, January 16th 2012 - 9:36 PM

There's no reason at all why you can't have a fixed bed if your material is going to be 10mm or less - either a 50 or 63.5mm lens will be fine - the mechanical adjustment for both lenses will still be 30mm. If the nozzle lengths are different between lens focal lengths then you'd just need to have two gauges (or a 'duogauge [tm]' :lol: ) as the focus position will be different from the end of the nozzle.
Certainly makes the machine much more rigid - not to mention cheaper :D

Cheers

Neil

add commentadd comment in the forum

reader comment Comment from: educa on Monday, January 16th 2012 - 9:28 PM
Thanks Greot, but can I ask you (like I asked iGull) if I do need a moveable Z table of I only itend to cut/engrave on plate material with thickness < 10mm ?

I suppose if I could position my nozzle correctly and then mount the table fixed, then I will at least have about 10mm of playing distance with putting the nozzle up and down ? This could lead to a dramatically easier design for me since it would eliminate all the stuff to move z axis table.

but I'm not 100% sure if what I say here makes sense.
reader comment Comment from: Greolt on Monday, January 16th 2012 - 9:23 PM
My understanding is that the head from CNCOLETEC has different length air assist nozzles for different focal length lens.

I bought mine with a 63mm focal length lens and had wondered if I should try to purchase a nozzle and lens holder for 50mm lens.

Then it would be a very quick change from one focal length to the other as required by the cutting or engraving job.

Greolt
reader comment Comment from: educa on Monday, January 16th 2012 - 9:05 PM
Just to be clear,


Are you saying now that I actually don't need a moveable Z table which goes up and down?


I also ordered a coletech laser head, but with 50mm focal lens (the larger diameter stuff)


So if I put my z table FIXED, I would normally just set the focus with the nozzle which I then set to do UP and down ?

You say your focal distance is 11mm below the nozzle. What I don't understand then is where it will be for me ? I have the same nozzle, but lens is 50mm instead of 63.5, so I expect my focal point to be 63.5 - 50 = 13.5mm above yours, but thats 2.5mm INSIDE the nozzle ????


Or am I completely wrong in my thinking here ?



If I don't need to make a up and down going Z table, then I would not do it :) I only want to handle plate material normally

Monday, January 16th 2012 - 8:53 PM

educa wrote:iGull there is something I don't understand.

You say you need 11mm below the nozzle for the focal point.

But now I'm confused. What focal distance does your lens have? And is the lens mounted INSIDE the nozzle maybe and not inside the holder where the nozzle then slides in ?

I know the nozzle can be moved up and down and possibly fixed with a set screw, but why would we do that? Or is it for people without a moveable Z table so you can setup the right focal distance that way ?
Sorry if question sounds silly or dumb.

Bart


No, not silly at all, my lens has a focal length of 63.5mm - it's mounted in the bottom of the parallel section of the tube just where the yellow acrylic ring is sitting in the picture above. 63.5mm below that happens to be 11mm from the end of the nozzle. The 63.5mm lens gives a better depth of field (over a 50mm lens) at the expense of a slightly fatter beam (just under 0.2mm in my case). As with everything in life, it's all a compromise :D

As you say, my nozzle TUBE height is adjustable with a thumbscrew (the parallel part of the tube telescopes in and out of the main body) - the focal point still remains at 11mm below the nozzle of course - that way, I can accommodate different heights of material (within a range of about 30mm) without having to change the bed Z position. You don't actually need to change the bed position unless you have large variations in material thickness or a fixed position lens.



Cheers

Neil

add commentadd comment in the forum

reader comment Comment from: educa on Monday, January 16th 2012 - 8:09 PM
iGull there is something I don't understand.

You say you need 11mm below the nozzle for the focal point.

But now I'm confused. What focal distance does your lens have? And is the lens mounted INSIDE the nozzle maybe and not inside the holder where the nozzle then slides in ?



I know the nozzle can be moved up and down and possibly fixed with a set screw, but why would we do that? Or is it for people without a moveable Z table so you can setup the right focal distance that way ?


Sorry if question sounds silly or dumb.


Bart

Monday, January 16th 2012 - 2:43 PM

Happy New Year to everyone :D

Work continues apace on the build - although it's essentially in place and complete now - even plumbed to the outside world.

I made a conscious decision to forget about the Z axis stepper and controller and all the paraphenalia that goes with it - I made a simple acrylic knob that fits on one of the M8 domed nuts that sit on the Z 'leadscrews'. It doesn't get any simpler really. Because I have the coletech laser head, I usually just drop the nozzle on to an 11mm gauge block on top of the material, but if I need more adjustment, then I can use the knob. I haven't gone the whole hog and replaced all the leadscrews yet - but I think I will at some point this year.

I still haven't bought any table material yet - while I realise that honeycomb/grid is a good way to go, I'm not fully convinced yet. I've been doing a lot of acrylic cutting recently and notice that I get a lot of flare-ups which only happen when I have the grid under (I have a small piece of metal grid and a plastic one too). It doesn't happen when the area underneath is fully open using the aluminium angle (I'm using air assist - no gas). I also don't get any flash marks due to laser reflection. My old Step-Four cnc system uses the same idea - just a bunch of C section aluminium across the bed - it's simple, accurate, easily and quickly reconfigurable - it's also relatively cheap. The downside is when using thin material like japanese tissue, you really need support, so I guess it's horses for courses. I may see if there are other alternatives to the honeycomb/lighting grid - something easily obtainable here.
While I'm on about tables, we had a discussion about material for under the table (can't find the post) - well, certainly in my case, even though it's de-focussed, the laser makes one mess of the underlying bed - it's about 250mm away from the head and makes a great camp fire :lol: :lol: I have white formica covered ply as a base - it certainly burns badly if you are running a slow job with high power - as in 10mm acrylic - if you are only moving at 300mm/min. I'll maybe drop in a few aluminium crumb trays if I can find some that fit.

While using the laser in anger, I missed the mark a few times when laying down my material to cut and thought I'd like a wee red dot marker. I lashed up a surface silvered mirror and red laser to run parallel to the beam from the fixed to the first flying mirror - while it actually worked really well, it was really another one of those pita alignment jobs I could do without. The arrangement was really simple, but could have done with some fine adjustments in all axes rather than using shims. Rather than go down this route, I chose the easy solution :D
I already had power up to the head as I have a capacitive sensor fitted (for Z height adjustment) - so I just removed that (the heavy cable was also causing some issues with the energy chain anyway) and made up a small acrylic adjustable mount that fits on the nozzle. I guess you could modify it easily for the 2.x if you wanted. I've attached a pic. Nothing complex, just a bit of laser cut 6mm acrylic with two holes and a couple of grub screws - knuckle joint in the middle to adjust the angle - the holes are aligned, so there shouldn't be any off centre problems.

Red Dot Laser.JPG
Red Dot Laser


On the controller front, I'm still using the Lasersaur setup. It's really showing signs of promise now (speed issue removed) and I'm keen to use it as I like the toolchain (it's also ultra cheap and it runs on the Mac :lol: ). I have nearly finished my LaOS board. The Lasersaur still has a few issues yet which I'm sure will be fixed. However, rasterising hasn't reared it's head yet and I feel that may well be an issue - however, perhaps not - I'd really love to be proved wrong.

While working with the software, there was really no 'test card' reference that we could use - so we can all 'sing from the same hymn sheet' so to speak. To that end, I lashed up a quick test card which I've attached. The data is in order, it writes the left hand diamond as part of a closure check, this is followed by the inner square working to the outer then across to the circles.The squares and circles are 100mm - inner ones are halved each time. It then jumps to the backlash patterns - a set for X and a set for Y - it writes a rectangular 'C' with overlapping centre bars - any gap in the bars indicates backlash for that axis. It then writes the outer frame of 287x200mm (that's 5mm inside an A4 sheet of paper - US B4 [or is it B5?] is about the same size) and finishes with the right hand diamond closure pattern - if the diamonds don't match up, then probably lost steps. You can check orthogonality too, the diagonal values are written in the file.
The file is in SVG format from Adobe Illustrator CS5 on Mac - it should convert easily to any other format. I believe Inkscape (and maybe in reverse - Illustrator ;) ) has a scaling issue - the sizes are written on the file. Any technical improvements that could be made to the card will be gratefully accepted (apart from making it imperial that is :lol: :lol: :lol: )
BTW, only write the red data, green data is just info.

Laser Test Card v1.02.svg.zip
Laser Test Card v1.02
(5.54 KiB) Downloaded 1494 times


Cheers

Neil

add commentadd comment in the forum

Wednesday, January 4th 2012 - 9:53 AM

Hi Bart

I used 3M HTD belt for x& y - http://www.motionco.co.uk/timing-belts- ... 38_39.html
They're 9mm wide 3mm pitch.
The pulleys were - http://www.motionco.co.uk/timing-pulley ... 35_50.html
You need to drill through and drill and tap to suit your shaft - in my case 5mm for the x nema17 and 6mm for the y nema 23. I used 6mm shoulder screws for the free pulleys - the screws are designed for this job with accurately ground 6mm 'shoulders' (hence the name :D ) - mine were cheap from ebay - they fit the 6mm bearing pair on the free pulleys.
I used the PLM025AL pulleys with a PCD of 23.87mm.
Step size is therefore pi*PCD/steps_per_rev (using 16 step =16*200=3200)=(3.1415*23.87)/3200=0.023434317700371mm
Steps per mm = 1/0.023434317700371=42.672
That's 0.0009" per step and 1,083.8805 steps/" in imperial FWIW :D
(PCD = Pitch Circle Diameter, multiplied by pi to give the distance the 'gear' [pulley] travels in one revolution)

I used a nema 17 hi current on the x and a nema 23 dual shaft on y - I'd change the nema 17 for the same as y as the 23 has a lot more 'oomph' for little mass gain.

Hope that helps

Cheers

Neil

add commentadd comment in the forum

reader comment Comment from: educa on Tuesday, January 3rd 2012 - 9:11 PM
iGull, I see you use joined T5 belt from beltingonline.com for your Z-axis.

Can you please tell what belt type you use for your X axis? Also what steppers and what kind of pulley wheel (how many tooth) ?

Thanks,

Bart

Tuesday, December 13th 2011 - 1:44 PM

I was working on the machine last night after finishing some work on the power tray - I managed to knock the X axis and popped two of the vee wheel rims off again :oops: Fortunately, I had pre-ordered some 16mm bearings and new extra length M5 screws so I could modify the vee wheels if needed at a later date - so I just modified them all.

The modification requires a 16mm OD x 5mm ID x 5mm thick bearing (same as already in the wheel - I got mine from eBay UK - 10 GBP for 10 - you need 9), an M5 x 30mm screw (that's 6mm more than my original - yours may be different), a plain 1mm thick M5 washer.
Replace the old M5 screw with the new longer M5 screw + bearing + plain washer in that order - you are sandwiching the 1mm rim of the acetal vee between two bearings. The plain washer makes up the space between the two bearings. There's now no need to glue the bearings of course.

You may need to uncouple the steppers to ensure that the stage is free running and perhaps re-tension the offset.

It took me about 1/2hr to swap out the nine wheels - all works fine now.

You could of course just buy the new dual bearing wheels and new standoffs from Bart of course and fit those - but being a cheapskate ... :D :D

Here's a pic of the modification ...

P1000986.JPG
Original Vee Wheel Modification


No doubt I'll have to re-align the laser now :D

Cheers

Neil

add commentadd comment in the forum

reader comment Comment from: TLHarrell on Monday, December 12th 2011 - 3:31 AM
True, but I like my little flying head on my little laser. It seems fairly straightforward to me to allow the table to raise and lower to set focus.

I've also seen a pretty cool laser head on a 400 watt machine that had a small plexiglass eye shield and a vacuum exhaust. I might be thinking of replicating that too...
reader comment Comment from: mikegrundvig on Monday, December 12th 2011 - 1:28 AM
Sarrus linkages are a neat mechanism but it seems the better approach is to minimize the need to lift the table in the first place.
reader comment Comment from: TLHarrell on Monday, December 12th 2011 - 1:06 AM
I'm working through the concept for a 4'x2' laser table as well. For my Z-axis lift, I'm considering using a set of Sarrus linkages in the four corners. These would keep the table moving in only the Z-axis and would eliminate twist and tilt movements. It's a great way to constrain the axis. For the lift mechanism, I'm considering a two-ended stepper with a set of cams that will lift the table near the center on the top and bottom edges. Kind of like a scissor lift.

On my little 40w Chinese I don't like how the table floats around when I adjust up and down as the threaded rods are not perfect.
reader comment Comment from: mikegrundvig on Sunday, December 11th 2011 - 11:57 PM
iGull wrote:This is all assuming that you have an adjustable optic tube for the laser head of course
I'm building my own optics carriers for a variety of reasons so I'll just add this to the mix. This is also something common on the larger format machines.

iGull wrote:Ok on the gantry, I can see some indecision creeping in there - hold on to that thought
Sorry, not near enough doubt to really change anything. With the gantry-mounted-tube being so common on large lasers, I'm sticking with it. As best I can tell, when you get into the 4' stuff for industrial use the flying optics become less common.

-Mike

Sunday, December 11th 2011 - 10:24 PM

mikegrundvig wrote:Thanks, that table discussion was very helpful. I think I'm going to build the table exactly as you suggested. Probably using something like these to lock it into position.


I've seen those used in loads of applications - one of which was to lock ejector seats in position ! - I think they'll be perfect if installed correctly. If you want to get fancy, I remember a matching steel sleeve bearing that goes with them - don't ask me for a link 'though :D This is all assuming that you have an adjustable optic tube for the laser head of course !

Ok on the gantry, I can see some indecision creeping in there - hold on to that thought :D :D :D

Cheers

Neil

add commentadd comment in the forum

reader comment Comment from: mikegrundvig on Sunday, December 11th 2011 - 9:42 PM
Thanks, that table discussion was very helpful. I think I'm going to build the table exactly as you suggested. Probably using something like these to lock it into position.

iGull wrote:You definitely don't want my advice on a gantry system 'though - moving glass/gas/water/25Kv/sensor cables and air is a recipe for trouble - go for it 'though - no gain without pain as they say
Heh, I'm debating with myself pretty heavily on the gantry tube system. Too many of the large commercial lasers do it for me to disregard the design out of hand and it actually makes for a simpler overall structure as you don't have as many structural members critical in the alignment. Running the HV wire doesn't worry me, as I'm just going to use much higher grade than the machine calls for. My only real concern is running water to the tube as it will need to flex frequently. The weight of the gantry even with everything on it isn't much of a concern at all. I'm using 36v steppers with higher-end drivers than most here so I'm not at all worried about either the power or speed problems.

Thanks for sharing your experience!

-Mike

Sunday, December 11th 2011 - 8:02 PM

mikegrundvig wrote:I've been following your build with great interest for quite a while now as it's the most similar to my intended build on this forum. Care to elaborate on your table/focus comment in more detail? I'm currently debating how I want my table movement to work and so any advice from someone who has "been there, done that" would be greatly appreciated.


Hi Mike

Well, it's all just too easy to add 'gadgets & gizmos' to a system - it's much more difficult to stand back and add less ! Interlocks, things that switch fans on/off, water level and flow monitors, automatic fire extinguishers, autofocus systems and extra processors, blue flashing lights etc etc - the list is endless - some engineers and folks that like to build stuff have this serious over-engineering problem :D . My experience is that the more there is, the more there is to go bad at the wrong time. You need to stand back and ask yourself what you REALLY need and could you keep it simpler.

This leads us to the autofocus/z lift system. If you have a laser nozzle that is not adjustable in height, then it follows that you need a fully movable system to adjust the table height instead. I initially went for the four corners (I was just following the fashion here without giving it too much thought - I actually have a coletech focus tube !), leadscrew belt and stepper approach thinking that was what was really needed - which seems great when you are designing it, but in practice, isn't that wonderful. On the smaller bed size, the 'leadscrews' (read threaded rod, not real leadscrews :) ) you can just about get away with M8 or M6 size threaded rod and some loose nut system to take up the inevitable inaccuracies. You can even get belts that fit that size too ! The bed isn't that stable (even though the screws are in tension) and will vibrate with stepper motion/resonances (which telegraphs it's way to the laser cut of course). What you need to do then is perhaps add sliders bolted to each of the corners and the frame to give that stability - then you have the vertical alignment issues trying to stop the whole thing from jamming up. We haven't even started to align the bed height to the laser yet - depending on the lens you are using, then you really want the bed to be within no more than 0.25 to 0.5mm (and that is very roughly) across the whole table. While aligning the bed sounds trivial, when it gets to be 1.2 metres long, you can't just stretch across for a wee tweak of the leadscrew :lol: It's very difficult to find a supplier to get the belt size you want too (I had to get mine specially made after failing miserably to get a chain working). BTW, tweaking the leadscrews to get the bed the same height IS relatively easy - just don't touch anything when you are refitting the belt (or tightening the locking grubscrews - which is another story :D )

FWIW, my advice would be to have a simple system of four 20x40 vertical rails 'match drilled' (pillar drill all four while taped together), parallel and securely bolted to the frame. The match drilling would fit four sprung and lockable pins in the table frame - keep one as a permanent reference then adjust each vertical rail in turn until the whole is dead nuts on. Moving the table up and down is a simple matter of locking back each pin and sliding the whole table to the height required - it can't be at the wrong height because you match drilled them. It's important that the table be rigid too of course - I'd go with the 'corner cube' connectors again - they work very well. There are no electronics/water/high voltage/software involved to go wrong in the system :D

You definitely don't want my advice on a gantry system 'though - moving glass/gas/water/25Kv/sensor cables and air is a recipe for trouble - go for it 'though - no gain without pain as they say :D :D :D :D

On another note, I was reading a post somewhere (maybe here) about someone 3D'ing a laser air assist nozzle - presumably from ABS - it reminded me of an appropriate engineering term 'about as much use as a chocolate fireguard' :D :D :D :D

Cheers

Neil

add commentadd comment in the forum

reader comment Comment from: mikegrundvig on Sunday, December 11th 2011 - 4:36 PM
IGull wrote:I'm not sure that I would bother with the autofocus and Z lift again - it would be a whole lot easier and cheaper to just have a solid fixed bed with a range of positions the bed could be put in - I may yet retrofit this idea.
I've been following your build with great interest for quite a while now as it's the most similar to my intended build on this forum. Care to elaborate on your table/focus comment in more detail? I'm currently debating how I want my table movement to work and so any advice from someone who has "been there, done that" would be greatly appreciated.

-Mike

Friday, December 9th 2011 - 4:54 PM

TLHarrell wrote:I like your idea of including all the additional support equipment inside the case. It's looking good, even in yellow (not really my favorite color). Too bad the magnets didn't work out.

Thanks, the support stuff has to go somewhere and it's just as easy to build a taller cabinet as it is to build a separate table and have all the pipes and stuff lying all over the place. Actually blue or yellow were my first choices, they didn't have the polychrome blue - so ... :lol:
TLHarrell wrote:How's alignment and accuracy working out for you on the 4' axis? Any issues, or anything you would plan differently?

Apart from forgetting to align the four corners last time I did a realignment (I had replaced some screws I didn't have at the time) - and melting a hole in my workshop printer because of it - really no major problems at all. Typical laser alignment - could really have done with another body to help as the machine is so big (three eyes instead of just one :) ) - I ended up making a long cable with a pushbutton to fire the laser remotely while I adjusted the mirrors. Use of dymo labelwriter labels is great - self adhesive and you can print the exact size target on - saves a LOT of guesswork over till receipts and masking tape.
To be honest, there are only a couple of minor things I would change - the 'optical table' members are only 80mm apart I could have done with 90 as the right angle torx key doesn't quite fit. The middle body rail could do with being made of 20x40 rather than 20x20 - it would have been easier to fit the plenum floor. I'm not sure that I would bother with the autofocus and Z lift again - it would be a whole lot easier and cheaper to just have a solid fixed bed with a range of positions the bed could be put in - I may yet retrofit this idea.
The coletech laser head adjusts over about 30mm or so and it's easy to just use a focus gauge block and tighten the tube to it (like a lot of 'pro' machines do).
The Igus energy chain is not that great (it's the smallest with an 18mm bend radius) - it tends to curl up rather than smoothly traverse in X - I couldn't get the Kabelschlepp stuff (I have it on my CNC router and it's much better).
One point I would DEFINITELY recommend is using core screws over right angle brackets - it is MUCH more rigid (and hugely cheaper). Ok, on the downside, it is a pita to drill the holes to access the core screws and you have to keep aware when building otherwise you end up with the 'insufficient panel nuts' problem in the 2.x and have to strip it down to refit the member you forgot :lol: There's also no major adjustment, so you need to know exactly where it's going.

Mike wrote:Any details on the magnets? I'm curious to know where the problem was.


Yes, my original intention was to use neodymium magnets to hold all the panels on - same system as I have with the screws currently, but magnets in each of the tabs and mating magnets adhered to the back of the panels (I've used it on standard panels a few times). The problem was keeping the magnets attached to the panels - it must be the paint that is on the reverse of the aluply - every adhesive I used (I tried many - cyano/acrylic/gorilla/epoxy etc etc) eventually pulled off (no, there wasn't a polythene film attached :lol: ). By this time, I didn't fancy grinding away the paint at each point (120 stations) to get a clean aluminium surface, so I chickened out and used button screws with captive grip nuts in the acrylic tabs. Maybe at a later stage I might try something else - meanwhile, I have lots of more important things to do :-)

Cheers

Neil

add commentadd comment in the forum

reader comment Comment from: mikegrundvig on Friday, December 9th 2011 - 9:12 AM
Any details on the magnets? I'm curious to know where the problem was. Thanks!

-Mike
reader comment Comment from: TLHarrell on Thursday, December 8th 2011 - 10:51 PM
I like your idea of including all the additional support equipment inside the case. It's looking good, even in yellow (not really my favorite color). Too bad the magnets didn't work out.

How's alignment and accuracy working out for you on the 4' axis? Any issues, or anything you would plan differently?

Thursday, December 8th 2011 - 10:07 PM

The Emperor's new clothes :-)

Eventually got around to getting some panel material for the machine. I had intended getting 'Aluply', but when at the distributor, he offered another material which was a 'clone' as far as I could see - called "Adabond" - looks exactly the same, same thickness, same core, same colours as aluply - except that it is about half the price ! Of course, that's what I bought ! I had always intended getting yellow, but nearly changed to the nice poly blue - but it's just so common on machines these days LOL :D

Cut all the panels on the table saw with an 80T blade - cuts just fine - just a matter of rubbing off the swarf with a thumbnail. I set up a jig on the pillar drill to do all the holes - I had intended using magnets, but after some 'sticky' issues, moved over to stainless button capscrews and laser cut acrylic tabs that are a friction fit in the extrusion grooves - there are M4 panel nuts crimped to the tabs.

P1000973.JPG
Laser cut acrylic tabs


P1000971.JPG
Machine rear


The design of the machine was to include all the peripheral stuff inside - cooling, extraction and air assist. The space under the 'smoke plenum' was the intended area. I've fitted a 150mm (6") mixed flow (560m^3/hr) 'vent axia' brand inline fan unit to the base and added a purpose built inlet to the base of the smoke plenum (with a hat to stop all the cut bits getting sucked off :) )...

P1000969.JPG
Fan inlet


P1000972.JPG
Fan Inlet cover


The outlet of the fan is a hundred millimetres or so away from the workshop wall and is a straight outlet - no bends or restrictions to add pressure to the system. I'll give it a smoke stick check once I get the rest of the covers screwed on and post a video. It is VERY quiet - I don't have a sound level meter to hand, but the spec says 47dB(A) at 3m. I'm aware that the fan should be outside in case the hose bursts on the outlet side and into the workshop !

The electronics/power drawer is nearly finished, so should be able to get on with some serious controller testing now. I've left off the water chiller for now - it's all wired, but after a pile of cutting, the water was hardly warm (it's a 12 litre flat container).
I have a design for a flow meter that fits to the open end of the outlet pipe rather fitted than inline (bad memories of gummed up flow meters causing flow to stop moved me in this direction). It's cut from acrylic and like the pump is submersible - bit like one of those 'treasure chests' that you see in aquariums. If it fails, the flow water is not restricted.
Just a few magnets rotating against a reed switch - needs something to check that the switch is opening and closing - I'm using a picaxe - part of a 'utility processor' that handles the temperature and z motor, but any old micro will do the job - even an LED and a capacitor should work. Once I get the thing cut, I'll post up the design.

Cheers

Neil

add commentadd comment in the forum

Tuesday, November 22nd 2011 - 4:36 PM

Apologies Mike, I missed your question ...

mikegrundvig wrote:I'm very glad to hear it's working great. I'm going to take a partial assit on this one with my comment " Do you know you are getting the power you expect?" :)


Actually, I suppose the answer is no, but judging from what others can cut with 40W and that I can exceed that by a decent margin - single pass on 9mm MDF, 3mm liteply at 4000mm/min - I'm fairly happy that it is putting out a good bit more than 40W. I don't think the power output will be linear for cutting - for instance - 6mm at 40W will most likely need a lot more than 80W to cut 12mm at the same feedrate - I'd be interested to see if anyone has any figures on that.

Cheers

Neil

add commentadd comment in the forum

reader comment Comment from: mikegrundvig on Wednesday, November 16th 2011 - 8:55 PM
I'm very glad to hear it's working great. I'm going to take a partial assit on this one with my comment " Do you know you are getting the power you expect?" :)

-Mike
reader comment Comment from: TLHarrell on Wednesday, November 16th 2011 - 4:45 PM
MDF cuts great on my little laser! It's very homogenous, so cuts are always the same. I usually use 1/8". Edges do come out quite dark. I have also cut 1/4" MDF, but it requires higher power and more passes, and has significant charring. Pieces still come out with very tight tolerances though. I can maintain a tight slip fit on slots by offsetting the original part by .003" and cutting normally.

Rastering produces zero char. Vector cutting, even at very low power for marking (see rocket above), does produce dark lines. The hatching on the fins was rastered, the remaining details were vector cut at 2% power.

Wednesday, November 16th 2011 - 2:33 PM

The sweet smell of success (or burning 6mm ply in my case :D ).

I took some time this morning to monitor the tube current on full power using the lasersaur board - 12.3mA. This is roughly half what it should be. The quoted operating current for a 60W tube is 22mA. Using the test button with a 5k pot set to max gave 27.2mA!

The lasersaur board setup uses only three wires - gnd(4), TTL-H(1) and protect(3). I checked the PWM signal on max (255) - straight lined at 5V as hoped. At about half - 128 - the PW was about 50% - PRF was 30uS. I added a pot to the setup and set to max - 27mA - backed it off until it was 22mA during a test cut (pot across 6,5 and 4 - wiper to 5).

I can now cut hard 12mm (1/2") balsa at 1100mm/min (18.3mm/s) on max pwr - I couldn't cut it at all before without creating a flaming mass :o
The 6mm quality aircraft ply (5ply) - I can just cut it at 200mm/min - but to be honest, the material is really cremated. I tried two passes - lowering the nozzle half way - but that was just as bad - as Tim said, "it's all in the glue" :D

I'm not sure why it is is happening now, but I can cleanly cut 3mm liteply at 4000mm/min (157mm/s) - and the edges are barely marked.

Anyway, I'm happy today - I need to re-read the chinese manual that comes with the PSU (coletech) - looks like you set the max pwr attainable with a pot (current limit) and the PWM gives a percentage of that max. No pot, and it would sem the current is limited internally to about half. I also need to look at Barts schematic - as they say, if all else fails RTFM :D :D

I'd like to hear anyone's attempts at MDF (not chipboard) - I get VERY blackened edges on the stuff, it's not pretty - I can cut up to 9mm (albeit at 300mm/min).

Cheers

Neil

add commentadd comment in the forum

reader comment Comment from: TLHarrell on Tuesday, November 15th 2011 - 10:47 PM
iGull wrote:First off - nice rocket lamp ! Reminds me of Buck Rogers ! BTW, how DO you get the bulb in :?


Thank you. I'm really happy with it too.
There's a double ring just above the bulb socket. The two rings have three rectangular holes in them that have matching clips that lock them together.

Next up, I'm working on designing some permanent fastening methods for flat MDF parts so I can ship flat packed and don't require any glue.

Tuesday, November 15th 2011 - 9:49 PM

Hi All

First off - nice rocket lamp ! Reminds me of Buck Rogers ! BTW, how DO you get the bulb in :?

mikegrundvig wrote:Ugh, I'm really disappointed to hear you can't cut 6mm plywood with a 60 watt laser. Obviously the glue used is critical but still, I'd hoped that 20 extra watts would make a difference. A friend of mine says the FabLab-type place he goes to routinely cuts .25" plywood on a 50 watt laser so I'm just not sure why they can do that and a 60 watt is having trouble.


Mmm - what we need is a level playing field here - something common we can all refer to worldwide - I'm guessing 3mm liteply (laserply as it's occasionally called) - most ARTF model aircraft/boats/castles are made with it - might be an answer ? Other option might be a notepad of graphpaper ? Although I can get access to a laser power meter, from memory, it only went up to 10W at 10.6um - anyway, even between power meters there are huge differences ! (I had problems through my whole working career with calibration pi$$ing contests :D )

mikegrundvig wrote: Do you know you are getting the power you expect? Is the tube old and partially discharged? I really need to cut 0.25" plywood a lot so this is making me think I should go with at least an 80 watt tube. Thanks and good luck!


The tube is brand new from Coletech - however, I have no idea of it's condition and age - label says something in chinese the 11/07/11 which I'm assuming is the 11 July 2011. I'd really like to cut 1/4" plywood too :D which is why I went to 60W ! My experience of laser cutters is low, perhaps the 50W laser your friend has access to is an expensive RF type and not an 'el cheapo' chinese tube like mine ?

twehr wrote:It's all in the glue.
I am cutting 1/4" red oak and birch ply from the big box stores with 40 watts - slow but cuts in a single pass.


Define 'slow' ? Therein lies the rub Tim - expensive plys tend to use phenolic resins - phenol-formaldehydes from memory - also used to manufacture printed circuit board (before epoxy glass fibre) - that might account for not cutting through my aircraft ply (it was 6mm 5 ply BTW, I just checked).
I don't know how many laminations would be in your ply Tim - I'm guessing 3 ? I know the US home trade uses different timbers to Europe, so trying to get a worldwide handle from there could be difficult.
As a quick test, on full power at 600mm/min (10mm/s), I can cleanly cut through a stack of 4 sheets of 3mm liteply at once.

I've done some quick testing - the fastest I can make a clean cut of a 20mm square of 3mm liteply is 2500mm/min - 41.6mm/s - 98.4"/min - 1.64"/s.
I also did some tests on a paper graphpad lying on my desk - it's 5mm graphpaper - sheets are 0.07mm (0.003") thick - here's the results ...

0.76mm @2500mm/min (98.4"/min)
0.99 @ 2000mm/min (78.74"/min)
1.35mm @ 1500mm/min (59.06"/min)
1.75mm @ 1000mm/min (39.37"/min)
I didn't count the sheets, being an engineer, I used a digital vernier gauge - I left it to the student to calculate how many sheets it cut :D

I'm hoping someone can perhaps try a similar test with their 40W chinese tube lasers. I revisited my laser alignment at the weekend and I'm extremely happy with the alignment - everything is down the centre. I double checked the beam placement from the final mirror by aligning the tube to a scribed circle with a cross (mine is the big coletech black assembly BTW) - the tube was moved back up to focus position, the spot was smack bang on the crosshair. FWIW, from the bottom of my nozzle to the cuttable surface is 11mm (I have a 63.5mm lens). I fully cleaned the optics with methanol before I tested tonight.

Something I also did as I had previously seen 'stitching' in my cuts (probably due to using the lasersaur board - I'm guessing the PRF of the PWM signal is a bit slow) was to use the 'test' button to ensure it was at max power (no other connections connected to the PSU other than 240V) - judiciously pressing the button at the start of the cuts :D

Apologies for the long post, I look forward to any replies and suggestions.

Best regards from a wet and windy Scotland

Neil

add commentadd comment in the forum

reader comment Comment from: twehr on Tuesday, November 15th 2011 - 4:32 PM
mikegrundvig wrote:Ugh, I'm really disappointed to hear you can't cut 6mm plywood with a 60 watt laser. Obviously the glue used is critical but still, I'd hoped that 20 extra watts would make a difference. A friend of mine says the FabLab-type place he goes to routinely cuts .25" plywood on a 50 watt laser so I'm just not sure why they can do that and a 60 watt is having trouble. Do you know you are getting the power you expect? Is the tube old and partially discharged? I really need to cut 0.25" plywood a lot so this is making me think I should go with at least an 80 watt tube. Thanks and good luck!

-Mike


It's all in the glue.

I am cutting 1/4" red oak and birch ply from the big box stores with 40 watts - slow but cuts in a single pass.
reader comment Comment from: mikegrundvig on Tuesday, November 15th 2011 - 4:13 PM
Ugh, I'm really disappointed to hear you can't cut 6mm plywood with a 60 watt laser. Obviously the glue used is critical but still, I'd hoped that 20 extra watts would make a difference. A friend of mine says the FabLab-type place he goes to routinely cuts .25" plywood on a 50 watt laser so I'm just not sure why they can do that and a 60 watt is having trouble. Do you know you are getting the power you expect? Is the tube old and partially discharged? I really need to cut 0.25" plywood a lot so this is making me think I should go with at least an 80 watt tube. Thanks and good luck!

-Mike
reader comment Comment from: TLHarrell on Tuesday, November 15th 2011 - 12:40 AM
Sweet little rocket. Looks a bit like the lamp I designed.

Image

(On Facebook: White Light Laser)

Monday, November 14th 2011 - 9:20 PM

We have liftoff :D

tintin.JPG
Tintin rocket from Thigiverse


Don't know what's going wrong with the pic, it's portrait on my computer - rotated it and it's just the same !

Spent some time at the weekend testing a pile of materials - lots of success, but many failures ! I had hoped with a 60W laser that it would be able to cleanly cut 6.0mm birch ply - not. This is good quality 'aircraft' 7 ply - not the cheap and nasty 3 ply you get from DIY stores (although I had the same issue with that too :cry: ). I barely made a dent in it - perhaps 3mm if I'm lucky - full power and only a couple of hundred mm/min - lots of black soot though :D

It would have been nice to check a database of materials and their cutting speeds/powers required. I fully realise that each machine is different, but there has to be correlation somewhere. I'll create my own as a matter of course as it'll be needed anyway. I don't see a 'database' of materials in LaserCAD or PHCad - I should have thought that it would be a necessity ??? My step-four cnc router software had all that built-in - feeds/speeds/materials.
If there is any interest in a database, I could lash something up in FileMaker (I can create standalone databases for a couple of platforms) - maybe an online wiki would be more useful 'though - Bart?

However, scored a brace of transparent yellow 6mm perspex sheets originally intended for an optical cleanroom area - makes for a nice transparent cover for the cutter!

I did a full cut of a tintin rocket (thingiverse) from 3mm liteply - cuts very easily on low power - so I'm quite happy with the results. Beam diameter seems to be a tad under 0.23mm - this is with a 63.5mm lens - I'm really not sure what it should be - 50mm lenses seem to be around 0.15mm ??? Perhaps someone can clarify. I've done a focus ramp test and the focus range is quite wide - about +/- 2mm.

I also picked up some old free plastic lighting grid - actually works ok - after loads of cutting, it's barely cut away - it'll do for now.

Cut some 20x40mm extrusion end caps from 6mm smoky acrylic - really fast to cut - but the smell - yukk :D I don't have my extraction fitted yet as there are no skins on. You can't get 20x40 caps for bosch/rexroth - you are supposed to use 2 20x20 caps. I just screwed them on using core screws.
I've cut my hand a few times on the bare extrusion, so they are quite useful.

Cheers

Neil

add commentadd comment in the forum

Thursday, November 10th 2011 - 10:48 PM

Hi All

It's been a while since I last updated the log - mainly due to vacation and general laziness :D

Things have moved on a little since my last post, I've installed some electronics, air and waterworks to get the beast moving (most of the work is temporary, but the stage and X axis cabling is permanent).

I'm now running with a pair of Keling 4020 drivers at present along with a Lasersaur setup consisting of an Arduino Uno (with lasaurgrbl) interfaced to my Macbook Pro running the python serial server and a copy of Firefox (the serial server I believe is to be moved from the host computer to an Mbed ). What can I say, everything I have touched with arduino and lasersaur has worked first time every time - no glitches, no hang ups no reboots - zilch - it just works. Anyone that puts down this arduino setup needs their heads examined. OK, it may (in the future) have some issues with rastering, but the current setup works faultlessly and is a VERY clean interface (Mach take note :lol: ). It won't take much to turn this setup into something very professional (and also CHEAP !). I should really join the lasersaur set, but I only need the software (and I can spend sixty bucks on the skins :D ), so there's not much point - I'll wait until it is fully open source.

Really didn't have any issues installing and aligning the 60W tube - apart from the fact that the original tube I received was broken (the water nozzles at the rear were snapped off) and the new replacement tube I find is NOT 56mm but 60mm in diameter !! Consequently, the tube doesn't fit my mounts :x I haven't informed coletech yet - the tube is 1250 mm long, so it's not an 80W tube (which is what I thought - it does say in chinese if anyone can read it). However, I lashed up a couple of alloy tubes and epoxied the rear nozzles back to the original tube - bingo, no leaks and it all works - AND I have a spare tube - albeit 60mm :D I may be prepared to sell the new 60W tube if anyone is really interested - I believe there is a shelf life.

Had a few semi-major design issues - my original intention was to have a 1240x620mm working area, but due to a faux pas, the energy chain interferes with one of the rear Z mount brackets and loses me 18mm of Y movement - I can still cut 1210x610 (4'x2') sheets 'though - just :roll:

The new Z belt system works fine and is quiet - however, I haven't had the chance to do anything with it yet and just rotate the whole thing by hand for now :D While on vacation in Florida, I was directed to this surplus 'shop' - Skycraft Surplus (more of a warehouse - even has spares for cruise missiles :D ) and scored a proximity sensor for 5 dollars - best part of 70 GBP across here so it was worth the ten minute drive. I have it mounted and it works (within about 5% each time) - I have a small 15mm disc of alloy that will sit on the material - the intention is to build a small utility processor with a stepper driver that can handle all the odd jobs needed. Saying that, a winder handle on the z axis leadscrew and an 11mm gauge works equally well :D

I've attached a couple of vids of the beast working - I still get a kick out of laser cutting balsa ribs - when I consider the amount of wasted time in my life cutting ribs with a broken razor blade :lol:



Next are the skins - my original choice was magnet fittings, but after getting some sample skin material, the magnets came undone - epoxy/cyano/gorilla/silicon - the magnets eventually pop off the rear painted surface - I'm really annoyed - I've moved to screws under duress, but at least they are into small tabs that will fit in the extrusion slots. The removable panels will have thumbscrews. I guess this machine is really a prototype and you have to put up with some compromises, but that's not in my nature :lol: :lol:

Once the plenum floor is fitted (there's a second floor to be fitted above the waterworks and electronics to form a 'smoke' plenum) and the eggcrate table is done, I need to make some decisions about which controller system I will finally use. Tim Wehr has just posted some info on his blog about tests he has done using VMware Fusion on his Mac using the LO Controller - this looks exceedingly promising - anyone need a cheap(ish) 60W tube in the UK to pay for it =;-???

Cheers

Neil

add commentadd comment in the forum

Friday, September 30th 2011 - 9:23 PM

New joined T5 belt and pulleys arrived this afternoon from beltingonline.com - pretty sharp service.
Can't really see much of a join in the belt - it's an off-white colour belt and you can only see a slight degradation in texture on the outside face - I gave it a good pull test by hand (very technical :D ) but couldn't separate it (probably impossible by hand anyway looking at the original steel belt specs).
Centre drilled all the pulleys to size on the lathe and drilled and tapped for grubscrews. Removed all the original chain sprockets (what a pain) and refitted and re-adjusted the new pulleys. I added an idler wheel to give the drive pulley a bit more 'bite'.

New T5 Belt.JPG
New T5 belt and idler


What a difference ! All nice and smooth (relatively speaking :D ) - can even turn it all by hand - I'm really wondering if a z stepper is all that necessary - a small handwheel to catch one of the top M8 domed nuts would work equally well without all the attendant driver/PSU/software etc. Need to add another strut between the two 20x20 leadscrew supports as there is a little movement there.

I had given some thought to the z driver - rather than drive this axis from the main software/firmware (yet to be decided), I had intended creating an isolated z axis system with autofocus detector and integrated stepper driver - probably driven from a picaxe. Using 'yet another' arduino seems a bit of a waste of technology when a picaxe is only a couple of pounds compared to twenty-odd for an arduino (and it uses basic rather than c :D ) A single encoder 'pot' would work fine as the manual control and either a microswitch or capacitive sensor for the autofocus - all the ones I've seen on cutters use something like a small removable metal block for the sensor to operate on - I haven't seen a 'real' autofocus system as such.

Cut and fitted some mounting brackets for the x axis microswitches - I much prefer an old-fashioned microswitch as an end sensor to the optical variety.

add commentadd comment in the forum

Wednesday, September 28th 2011 - 5:17 PM

I just ordered the 10mm T5 joined belting and pulleys from beltingonline.com. I'll give that a try and see if it is any better. I'll need to add a wee idler to give the drive pulley a few more teeth to play with.

I built a dummy laser to get a feel for the whole alignment ...

photo.JPG
Dummy laser


Apologies for the iPhone pic. The dummy is a length of 40mm alloy tube and a couple of turned end caps from some acetal. The laser (a rather expensive red alignment laser that was a spare for some scrap equipment) is an 11mm small black tube with focus lens - laser is a bar - I just roughly focused it through a 1mm hole in the front mount - the laser is mounted at the back. A 60W laser tube is quite long - 1200mm and 56mm dia.
Trying to align the stupid laser through the front 1mm hole was a royal pita - even though everything was coaxial and accurately turned, the laser in it's own mount was not - to be expected I suppose. I don't for one minute think that the CO2 laser will be coaxial within it's tube either :lol:

Aligning the rest of the system took about five mins - I was pleasantly surprised, but the coletech adjustments were very accurate. One problem was in the way the mirror mounts are made - when mounted on a flat surface, the lower screw adjuster interferes with the surface - I added a 0.5mm brass shim underneath the mount - there are three mounting holes - 1xM6 axis and 2xM5 locks - this is enough to let everything move without raising the mirror too much from it's main axis.
All stayed aligned and focused from min xy to max xy and no-one more surprised than me !!

Onwards and upwards ...

add commentadd comment in the forum

Sunday, September 25th 2011 - 7:18 PM

BenJackson wrote:You can always "join" them by connecting loops by a shaft with two pulleys. You could connect the two sides front-to-back and the two back rails to each other and the motor, for example.


Yes, when I started on this project, that was one of the options - in fact, I have about three A3 sheets with various flavours of this idea :D
It's probably one of the better ideas as it cuts down on long belt lengths, but trying to get four belts tight without putting idlers on them becomes painful - not to mention all the mounts/brackets/pulleys :lol: I also have some ideas with counterbalances to hold the table up (using shafts) and then just pull the table down.

I spent the best part of this morning again looking at the chain problem. I tested a piece of chain on its own in 'vertical mode' - it was just garbage - either the sprockets are badly sized or the 6mm 04B chain is - the chain would jump as the sprocket teeth entered the space between. I've decided to give up on the chain - I'll punt it to some fighting robot guy :( It's not as if the chain and sprockets were cheap either !

Moving on from this, I double checked the 'friction' on each of the four Z leadscrews - they are all very easy to rotate, so a good belt would definitely work.
I made a concerted attempt to look at different belt styles and options but couldn't get anything at 2980mm. However, I actually found a UK company that can supply joined PU belting ...
http://www.beltingonline.com/polyuretha ... elting-38/
Surprisingly, the belt is CHEAPER than the flipping chain !!!! Why didn't I see this before :o It does say that the strength of the belt is approx half of a contiguous belt 'though. Swapping out a set of sprockets for a set of pulleys and swapping the chain for a belt is an easy out for now - it might even work :D
I don't suppose anyone has any experience of 'joined' belts ?

BenJackson wrote:There's no burn problem on the bottom because once you go through the focus lens the beam converges at the focal distance and then diverges after. Before the final lens you can burn things with the laser from quite a ways away. I measure about 3 focal lengths below the focal point on the 2.x laser, so the beam has 3x the diameter of the original beam or 1/9th the power density of the original beam.

I do think you want a cleanable surface on the bottom: I cut some plywood with terrible smelling glue and the smell lingered until I wiped down the bottom skin.


Had it been my turn for the family brain cell that day, it might have occurred to me that the beam power density would be appreciably smaller at the distance of my bottom sheet :D - thanks for that - seems that brain cell volume is inversely proportional to age :D
I have some 7.5mm laminated plywood - called 'waterboard' I think (has a real formica skin on each side) - they use it for shower cubicles etc - I think that will be ideal - easy wipe etc.

Cheers

Neil

add commentadd comment in the forum

reader comment Comment from: BenJackson on Friday, September 23rd 2011 - 8:30 PM
You can always "join" them by connecting loops by a shaft with two pulleys. You could connect the two sides front-to-back and the two back rails to each other and the motor, for example.

There's no burn problem on the bottom because once you go through the focus lens the beam converges at the focal distance and then diverges after. Before the final lens you can burn things with the laser from quite a ways away. I measure about 3 focal lengths below the focal point on the 2.x laser, so the beam has 3x the diameter of the original beam or 1/9th the power density of the original beam.

I do think you want a cleanable surface on the bottom: I cut some plywood with terrible smelling glue and the smell lingered until I wiped down the bottom skin.

Friday, September 23rd 2011 - 8:03 PM

Just in from the workshop trying yet another approach to this Z table issue.

I was correct in thinking that the chain 'droop' was causing all the problems. I clamped four scrap pieces of 18mm MDF board just underneath each of the sprockets so it guided the chain in and out of the sprocket (the gap under a sprocket to the nearest extrusion). The jam-ups ceased - still pretty noisy as expected. I think a sprung idler to adjust the tension rather than just using the stepper position to adjust it may also alleviate other issues - too much tension on the stepper for one - occasional 'dinks' for two (where it will occasionally catch on a sprocket tooth - being sprung means it takes the slack up unlike a tensioned stepper mount where it will just jam).
Short of going for the four steppers and associated headaches, I'll lash up some decent acetal guides and take it from there for now (cheapskate ;) ).

I'm also thinking that a full board with cutouts for the screw shafts and maybe covered with thin polypropylene or acetal sheet would work too - could even form the base of the upper 'plenum'. I wonder if anyone that has been using their .2x has noticed any 'burn' on the shiny blue bottom skin of their cutter ? In which case, I could get away with using MDF or ply laminate sheet - maybe with tinfoil covering as a barrier to any laser burn problems.

Goodness, if only they made 4.5m belts :D

I don't suppose anyone knows how to join toothed belt - is there a magic spelle involved ?

add commentadd comment in the forum

Friday, September 23rd 2011 - 4:56 PM

BenJackson wrote:The Prusa Mendel (3D printer) uses 2 Z motors because they realized that a stepper motor was cheaper than any combination of belts and so on that would let you drive two Z from a single motor. You just connect them to the stepper driver in parallel (could work for your 4 as well).

I don't quite follow your chain problem.


Chain problem - one of my problems is size ( :D ) - my table is 1240x630mm (about 4'x2') - the chain just droops across that expanse - it then grabs on the sprocket and jams. Increasing chain tension just serves to stall the stepper (even a hefty size 23 or my battery drill). The acetal guides may help align the chain to the sprocket.

OK on the mendel - that's certainly an option - replace the leadnuts with sleeves and the screws with rod then add a screw at each end with a stepper each. (or fours steppers which would mean no real mechanical changes - just a stepper mount) - the cost is becoming painfull :D

Cheers

Neil

add commentadd comment in the forum

reader comment Comment from: BenJackson on Friday, September 23rd 2011 - 4:21 PM
The Prusa Mendel (3D printer) uses 2 Z motors because they realized that a stepper motor was cheaper than any combination of belts and so on that would let you drive two Z from a single motor. You just connect them to the stepper driver in parallel (could work for your 4 as well).

I don't quite follow your chain problem.

Friday, September 23rd 2011 - 3:29 PM

Hi Ben

I have put a post in the mechatronics section with an idea about adding an extra bearing to the original vee wheels - have a look and see if it would work on the .2x ...
viewtopic.php?f=13&t=791
I'm looking at retrofitting it to my cutter if it causes me any continuing issues (so far, it hasn't since adding the belts).

On my Z stage front today - garbage :x I thought things were all going too well :D
As expected, the horizontal chain is causing a problem - mainly because Mr Chain prefers to be in the vertical plane.
It was a pita to fit too. Unsure which way to move - can't get a closed toothed belt about 4.5 metres which is why I chose the chain in the first place.
Really want to have the centre of the stage open, so no single vertical leadscrew in the centre with a rail for each corner - those things always end up being lopsided. I don't fancy a sarrus linkage - I have a little experience of something very similar and it needs to ensure that the bearing tolerances are tight - too much construction for me too - I prefer to kiss :lol: One option is to put acetal supports on the chain to ensure that the feed to the sprockets is kept aligned. One other option is four steppers of course :|

Anyone with an elegant solution to this issue is welcome to comment !

Cheers

Neil

add commentadd comment in the forum

reader comment Comment from: BenJackson on Thursday, September 22nd 2011 - 9:30 PM
I think all of my V-wheels have come unglued at this point. When I caught the gantry with the lid stop that popped a bunch of them on Y. I accidentally ran the X carriage into an object I was engraving and popped a couple of those.

I think the "trapped" V-wheel design from Makerslide is a good idea.

Thursday, September 22nd 2011 - 9:13 PM

More work over the past few days - and not a few c0ck -ups too :oops:

Seems that no matter how much 3D drawing you do, there's always something that bites you in the @ss :lol:
I'm kind of lazy when it comes to the fine detail in a model - I tend just to ensure that everything will fit accurately and leave the screws and nuts to themselves - I had to re-cut the belt mounts for both axes because of this - the pulleys then had to be mounted on the opposite side. On top of that, the travelling mirror mount was interfering with the top screws on the LH carriage - I just flipped it over and re-drilled the holes at 45 deg - sorted :lol:
And when I flipped the X axis stepper - I had to cut the lower vee rail - that's what God made Dremel discs for :D

P1000525.JPG
Aligning X axis to reference bar with a pair of 321 blocks


There's a matching 321 block at the opposite end. This was used to align the belts and pulleys for the Y axis. I'll recheck them tomorrow as one of them looks a little slack - although the alignment still seems orthogonal.

P1000526.JPG
Y axis belt mount


P1000528.JPG
Coletech laser head with air assist


Used 10mm aluminium tube to reduce mass and (hopefully) reduce any tendency to whip - turned alloy ferrules on ends held in with cyano.

P1000527.JPG
Y axis stepper


Coletech optical parts - reasonable quality - silicon mirrors with gold plating - adjustment screws about 0.25 to 0.5mm pitch.

P1000529.JPG
Three main optical parts from Coletech


I'm sure someone has used this in place of thermal receipt paper - Dymo labels - they're thermal, so you can print targets on them.
I'm selling the patent - you saw it here first :lol: :lol:

P1000532.JPG
Laser targets


All the mechanics seem to line up fine - no jam-ups apart from some cyano on a rail which I scraped off - really pleased with the rail alignment.
Had an issue with the black acetal rings popping off from the bearings - nothing will REALLY stick to acetal - I scraped the acetal and roughened the bearings with a dremel. The main problem was me playing with the X rail and it not being parallel to the reference rail - bit of side pressure. It's been perfect since putting on the belts and Y axis stepper shaft. I will be looking to buy the new version with dual bearings at some point, but I'll keep the current rails as they are perfectly OK

More work on the Z axis table and the dreaded chain drive tomorrow hopefully.

Cheers

Neil

add commentadd comment in the forum

reader comment Comment from: dirktheeng on Monday, September 19th 2011 - 10:00 PM
iGull wrote:This is a long shot, but you're not the same Dirk the Danish engineer that used to work for QC Optics in Connecticut are you ??



no I'm not... I'm dutch

Monday, September 19th 2011 - 11:56 AM

Was busy at the weekend cutting the rest of the acetal parts...

P1000523.JPG
Acetal Component Parts



Made a few minor changes to the final parts after cutting out some test pieces in MDF. Some of the holes were incorrectly sized, but I just opened them up in the drill press. I've added the silicone strips to the laser tube mounts - the strips are in a fetching red colour :D Nothing really adheres to silicone strip (after all, that's what it's designed to do I suppose :D ) - I used double sided tape. Tape probably won't last long - most man-made 'self-adhesives' seem to fall apart after a few years :x

There was some chatter on the lists a while back about belt pulleys and M5 screws I think - I've always used 'shoulder screws' for those type of jobs (there's one in the pic above) - they come in various sizes and have an accurately ground shaft (unlike the normal rolled screws) - they're cheap too - I think I got those on evilBay as I was buying some other bits at the time.

Spent what seems a fortune on screws/washers/nuts this morning - all stainless as they were nearly as cheap as nickel plated or black (in the hope that they won't corrode with any of the nasty gases that might be emitted - probably an overkill). I also spent another mini fortune on Bosch M4 tee nuts for all the parts and accessories. Having been in engineering all my life, all those things were available to hand (so to speak :D ) - being retired means having to get your own - I even had to purchase a reel of solder recently - something I've NEVER done :D :D

Pic also shows the Coletech optical parts - bit better quality than the standard chinese offerings (with finer adjustment on the mirrors etc), but nowhere near as good as European or US made 'industrial' optical components - you get what you pay for I guess and they'll be fine for what I'm using them for.

I've used Bart's vee wheels with the single bearing for this job - I think I'll upgrade them to the dual bearing versions when they become available. I think the new ones will be much more rigid when the machine size gets bigger. The single bearing ones will be fine for now 'though - they worked first time with no jamming on the large rails I have - I've never experienced that before ! There always seems to be a long tweaking period involved with regular linear bearings. They are extremely quiet too.

Cheers

Neil

add commentadd comment in the forum

Saturday, September 17th 2011 - 9:46 PM

Hi Dirk

This is a long shot, but you're not the same Dirk the Danish engineer that used to work for QC Optics in Connecticut are you ??

Anyway ...

dirktheeng wrote:I can attest to this... when I used my leveling feet to adjust the frame without the skins on, I could go from end to end with a level foot and still have all 4 on the table.


Ouch, then you do need to stiffen up your frame :D

I take on board your concerns for the stiffness of my 'wee' laser cutter :D

I have owned an accurate machinist's level for a number of years and use it at each stage of the build process. If I screw one foot too high, then another one comes off the floor too - my frame is pretty stiff. The main frame, the 'optics table' is built from two sections of 20x40 - spaced at 80mm. The frame is held by eight core screws at each corner. The frame corner 'legs' are 20x40 as are the lower cross members. Everything else is 20x20. I'm satisfied that it's rigid enough (I think :D ) - sans skins. I'm wondering if there is an issue with using diecast brackets ? For sure, they certainly make life easier during the build, but I'm a tad concerned when they are used singularly. The one flexible part is the lid - but to be honest, that doesn't matter ( I haven't put the 6.35mm perspex sheet in yet so it might get better :D )

dirktheeng wrote:I think you need to look into using much better mirrors than we have with very fine pitched adjusters. Even with the small travel we have now, just barely touching the adjuster on the mirror moves the beam significantly at the end. I can't immagine trying to adjust the beam with the mirrors we have if my travel were double what I have now.


I'm using 25mm Coletech mirrors and mounts along with their air assist final lens assy - not the slightly cheaper variety on the .2x (I'm not castigating the .2x BTW - far from it !!!) - the screw pitch looks to be around 0.25 to 0.5mm on the mounts FYI .

dirktheeng wrote:I think that is why most of the mounts you see have 3 screws around the perimeter. Moving the laser laterally isn't as critical becaue you can move the laser head forward/backward, but getting it up and down is not really easy... better to move the laser up/down.


:lol: :lol: - a bit of my history - I'm sixty years old and have worked in the semiconductor industry for the last 20 years and the previous 20 in the commercial military electronics business as a systems engineer and test equipment design engineer. My first experience of lasers was designing and building the laser PSU (using selenium stick rectifiers :D ) back in the late sixties then designing and building the mechanical mirror mounts to hold the primary and secondary mirrors (they weren't integrated way back then :D ). You had to align them manually with your last good eye and a bit of toilet paper :D I have extensive experience of aligning laser systems in lithography systems - mainly LMS and exposure lasers - I have spent many painful hours aligning other people's systems. One of my pet hates is trying to align a laser using the three screw 'system'. I think it must have been a woman that invented it (did I just say that - the PC nazis will be after me :D ) - honestly, why would you want to offset the laser at 45 deg during an alignment when what you want is an orthogonal adjustment ??? My brain could certainly never connect with this. The best system (IMHO) is a dual XY stage - but like all things, compromise (cost usually) is a necessity - so we use something else (have a look at Newport's XY optical stages for laser/whatever alignment - then look at their horrendous prices). While my system may look a bit agricultural to the 'uninitiated', there is some madness in the design - it's based on friction. The tube is VERY well supported in the two circular mounts, unlike the .2x system where the laser is ONLY supported on two 'points' (I am bashing the laser mounts, sorry Bart - I think you got them upside down ?). When the tube is loosely 'snicked up' in my mount, you can insert a flat bladed driver or allen key into the space between the frame and the mount and tweak the laser mount position - adjustment is available in X & Y - INDIVIDUALLY and NOT at 45 deg - laser alignment then becomes very much easier.

Rigidity in the rest of the system is also important - especially in the 'optics table' and anything attached to it. I'm using 6mm acetal plate (because I had two sheets really :D ) - it also machines well and is very rigid.

/rantModeOff :lol: :lol:

I cut and finished some acetal parts on my CNC this evening (before the first glass of wine :D ) - the belt mounts and clips mainly along with the secondary mirror mount - all looking clean so far.
The rest of the stuff should hopefully get cut tomorrow.

It's getting to an exciting stage in the project now when all the electronics/mechanics/liquids/air come together - then you see all the faux pas that you made :lol:

Cheers

Neil

add commentadd comment in the forum

reader comment Comment from: dirktheeng on Saturday, September 17th 2011 - 2:27 PM
TLHarrell wrote:As others here have stated, the frame is a bit more flexible without the skins than they had believed. You may want to bolt on at least some of the skins to provide a shear support to each direction of the frame and only use the magnets where you really need the accessibility.


I can attest to this... when I used my leveling feet to adjust the frame without the skins on, I could go from end to end with a level foot and still have all 4 on the table. After putting on the skins, the frame is solid and I have to be precise with my feet otherwise they lift off the table and it rocks. I would be sure to level and square your frame b/4 you tighten down the bolts on your skins. It will lock it in that config once tight.

Also, your frame is significantly bigger than the 2.x laser with travel distances that look to be much longer. I think you need to look into using much better mirrors than we have with very fine pitched adjusters. Even with the small travel we have now, just barely touching the adjuster on the mirror moves the beam significantly at the end. I can't immagine trying to adjust the beam with the mirrors we have if my travel were double what I have now. I would also suggest adding some micro adjustability to your laser mounts. It would suck to have to try to move the laser up/down to hit the last mirror in the middle without that. I think that is why most of the mounts you see have 3 screws around the perimeter. Moving the laser laterally isn't as critical becaue you can move the laser head forward/backward, but getting it up and down is not really easy... better to move the laser up/down.

Tuesday, September 13th 2011 - 12:07 PM

iGull wrote:One issue with using core screws is that you have to be ahead of yourself all the time - one out of sequence screw and you have to rip it back down again

Yes, I was right, I forgot to insert the upper Z table mounts which necessitated removing the optics bench then drilling and slipping in the mounts. When I put it back and trued it all up, I discovered I had put the bench back in '@rse for elbow' :oops: and noticed that I had drilled the core screw holes incorrectly on one side :oops: Not a productive afternoon ! However, a new day and taking some care has taken it to the next level.

Here's a picture of what core screws look like - you just screw them into the end of the extrusions, drill an access hole in the opposite piece then tighten them up using a Torx T25 driver.

P1000516.JPG
Core Screws


I found that using my little Makita impact driver made the job a whole lot easier !!!

I turned down four acetal inserts with M8 threads for the Z lift nuts - they are just turned from 20mm acetal barstock and pushed into a 20mm hole in the extrusion. The inserts are held in place with a couple of M6 grubscrews down the extrusion. I think there will be enough slack in the arrangement with M8 threaded rod and 8mm skate bearings above and below to allow the system to work. The threaded rod is rolled and is a tad under 8mm diameter.

P1000515.JPG
Z Lift Inserts


P1000519.JPG
Z Lift mechanism 1


P1000521.JPG
Upper Z Lift Bearing


The framework is pretty much complete now - I added some short stiffeners at the centre of the rear hinge support bar and one at the front of the frame. I'd like to add four more short stubs of 20x20 around the periphery of the optical bench for good measure (but I'll be using brackets, I've had enough of tearing it down :lol: )

P1000518.JPG
Main Frame Updated


This is the arrangement for mounting the laser tube - this is in 6mm MDF for test purposes - the inner of the opening is lined with 1.5mm silicone from a baking sheet I purloined from the kitchen drawer :roll:

P1000522.JPG
Laser Mount


I have +/- 5mm adjustment in X&Y - hopefully the laser will be fairly axial.

Thought I'd attach a pic of a cleaner I've used for a lot of years - especially for granite optical tables etc - but it works really well for cleaning anodised aluminium extrusion - called Chemsearch 'Sparkle' - it is an abrasive in a foam - the foam has a whiff of WD40 about it. No idea where you can buy it now - try Google.

P1000517.JPG
Sparkle


A few minor mods to the acetal parts before I commit to cutting.

Cheers

Neil

add commentadd comment in the forum

reader comment Comment from: TLHarrell on Thursday, September 8th 2011 - 10:12 PM
I just looked up these "core screws" you've been referring to. Hadn't seen them before. I'll have to consider them when I start my project.

Thursday, September 8th 2011 - 9:28 PM

Managed to do some work in between other jobs today (I'm supposedly retired !).

Frame reasonably complete now - regular coke can perched on frame gives scale to the job - it's about a european single bed size :D
I assume that coke cans are the same size the world over - like Macdonalds burgers :D

Workshop is getting a wee bit cramped now. I'm really pleased with the rigidity of the frame, but there are a couple of areas where it can be improved (the internal skins haven't been fitted yet of course, but it's probably prudent to have it rigid without the skins).

One issue with using core screws is that you have to be ahead of yourself all the time - one out of sequence screw and you have to rip it back down again (bit like skin screws and knitting :D ) Only used four brackets so far too.

Cheers

Neil
Attachements...
P1000512.JPG
P1000513.JPG

add commentadd comment in the forum

Wednesday, September 7th 2011 - 4:39 PM

Started building the rest of the frame today - lots of access holes due to using core screws rather than brackets. Built a simple chipboard jig to ensure relatively accurate positioning. Side frames completed.

Started gluing up the vee rails using my thin cyano method (well not mine, I'm sure hundreds of aeromodellers use the same process). I've attached some pix showing the process - gives you as much time as you need before committing.

Essentially, the process involves cutting all the materials to size and then give them a good wash down with cellulose thinners. Mark up the positions of the vee rails on the extrusion, then tape the rails to the extrusion starting at the centre of the extrusion then over the top to the other side. I used blue 3M masking tape. During the taping up, you can double check the straightness and lie of the rails. Make sure that the tape bridges over the gaps between extrusion and rail as we are going to run thin cyano down the gaps.
I double check everything before nuking it with the cyano - doing one rail at a time is sensible - you can also check using a vernier gauge that all is straight - mine were within 0.03mm (about 0.001 inches) across the length.
When happy with alignment, just tilt up the assembly and run thin cyano down the internal corner first until no more wicks in - then do the upper edge of the vee - start at one end and dribble some on the junction letting it run down and wick in - when it stops, move further down. Do the same for the other edge.
Don't overdo it and get cyano all over the place - be frugal. When it's all done, remove the tape - NONE of it should be stuck to anything or you have been overly heavy on the cyano. Wipe everything down with a cloth dampened in cellulose thinners. I used Thin Zap Cyano (pink) BTW - I'm sure that other thin cyanos will work just as well.

I've attached a pic of a pretest I did on some scrap extrusion and vee rail - I haven't managed to get it off using a hammer and length of ash timber all held in the vice - so no issues with strength here. Please ensure you do your own tests before trying this method as 'your results may vary' :lol:

Cheers

Neil
Attachements...
P1000510.JPG
P1000509.JPG
P1000507.JPG
P1000505.JPG
P1000504.JPG
Test Piece.JPG

add commentadd comment in the forum

Tuesday, September 6th 2011 - 12:20 PM

Yes, you are absolutely correct, but I just don't like the appearance of external skins - and having worked in the semiconductor business for 40 years, I have had my fill of removing system panels that are attached with screws (especially on a Saturday late night emergency call) LOL :D :D

When designing the frame, I doubled up the 20x40 optical bench members and kept them 80mm apart, so that the X & Y axes and laser tube mountings have a rigid 'optical table' to work from (they also don't interfere with the skins). The corner supports are 20x40 as are the lower cross members. In between stuff is 20x20 - there's some movement for adding more struts for lower doors etc which would increase rigidity somewhat. The z table is also attached (at the top) to the optical table, so there will be a modicum of 'common mode' movement.

If there is any drastic movement, I can add thin aluminium diagonal struts internally across the lower back and sides where access is not too much of an issue. The panel skins are also a tight fit inside the frame aperture which should help matters. The electronics rack is lower right on pull-out slides.

Just need to suck-it-and-see and hope it all works out as planned (as if it ever does :D :D :D ).

Made a start to the framework this morning and built the lid - it's HUGE - I'll need to come up with a neat strut arrangement to keep the lid open - gas struts are definitely NOT on the agenda - another throwback from Saturday night callouts (I have a list of stuff like this to shy away from - unfortunately, fans are in there too :lol: ).

Cheers

Neil

add commentadd comment in the forum

reader comment Comment from: TLHarrell on Monday, September 5th 2011 - 11:24 PM
As others here have stated, the frame is a bit more flexible without the skins than they had believed. You may want to bolt on at least some of the skins to provide a shear support to each direction of the frame and only use the magnets where you really need the accessibility.

Monday, September 5th 2011 - 5:49 PM

Aluminium extrusion eventually arrived this evening - took two of us to lift it all into the workshop :D
Using Bosch Rexroth rather than Misumi - the supplied stuff looks pretty good and the cutting is certainly as good as I can do on my chopsaw with an aluminium cutting blade. KJN were substantially cheaper than the other suppliers and the stuff was cut, not raw lengths. I haven't checked the accuracy of their cutting yet - fingers crossed :roll:

I need to lash up a drilling jig for the extrusion as I am using mainly core screws rather than brackets. There are a few reasons for this, brackets get in the way of skins and stuff if you want a clear access to the frame. The brackets are also about 1.30 GBP each including fixings, whereas a core screw is only 0.29 GBP - quite a saving when you add up the sizeable number of connections needed. I guess that using brackets makes it a bit of an easier 'meccano' type construction 'though :D
It shouldn't take long to drill all the access holes using the jig (only about 500 :lol: ) - it'll have a distance gauge stick built in. I'm using a few 'corner cubes' - on the lid and the table - leaves a neat finish and allows full access to the corners. I'll post some pix of course when the construction gets underway.

I don't have the issue with having sufficient nuts set into the frame to attach the skins - the skins will be held in with cheap 8mm neodymium magnets - the magnets are let into the back of the skins (skins are 3mm) - the fixed magnets are held in a 6mm strip of MDF (I may use some acrylic sheet if I have some scrap lying about - I suppose I could cut it with the laser really :lol: ). The strips are then just dropped into the extrusion slots and held in place with a little flexible adhesive - the corners are clear as there are no brackets.

Something that no-one seems to have mentioned (at least I haven't seen it ?) is tapping of the extrusion ends - bosch rexroth have small 'wings' in the holes which means that you can't use a regular tap - the flutes get jammed in the wings - you need to use a 'spiral tap' where the flute is a spiral not unlike a twist drill - the thread remains the same of course. I'm assuming that the Misumi stuff is similar ?

One thing I couldn't find was a decent set of substantial feet for the system - the ones I got are just the regular 25mm diameter plastic 'buttons' and M6 metal thread that are supplied for 6mm slot extrusion.

Looking forward to getting some real building done now rather than virtual stuff on the computer !

Cheers

Neil

add commentadd comment in the forum

reader comment Comment from: BenJackson on Sunday, September 4th 2011 - 5:50 PM
The 1/2" honeycomb from honeycommcore is actually quite economical if someone wanted to resell it. An unexpanded 4x8' sheet is $80. You can easily saw that in thirds (down to 16" x 8') and with some effort you can split those into 16"x4' which you can expand by hand and still get two full tables (plus waste due to expanding by hand). So a $80 purchase yields 12 table surfaces, or about 6 resellable (unexpanded) units.

If I had it to do again (and I didn't have a lifetime supply of 1/2" cell table material!) I might try to get smaller cells just to have more even support, but I suspect it would be harder to expand and MUCH harder to trim.

Sunday, September 4th 2011 - 3:36 PM

Nice link, I like this part ...

Exclusive new powder coated surface reduces bounce back and makes clean-up even easier


It's probably powder coated anyway when they buy it (most likely it's lighting grid) - let's turn this into a feature :D

I'm not sure what plastic is in powder coating - maybe some chemists among us can help - and whether it has any 'nasties' in it?

Unsure about those "economical prices" 'though :D

I'll be actively looking for the powder coated stuff in the UK - I was keeping away from coated stuff because none of the big laser manufacturers seemed to use it on their machines - thanks for that.

add commentadd comment in the forum

reader comment Comment from: naPS on Sunday, September 4th 2011 - 2:42 PM
That aluminum egg crate looks very similar to the en-gravs product. You can find that here : http://www.cuttinggrid.com

Sunday, September 4th 2011 - 11:10 AM

Finished cutting the test parts from MDF yesterday (took a while, my leadscrews are only 3mm pitch). Something I haven't really noticed before are the number of facets on circles/arcs etc in Sketchup - the default is 24 and it telegraphs into the finished cut of course. FWIW, you can change the default value easily - when you select say a circle, just type in the number of segments you want - I tried 48 and 72 - to be honest, 48 is fine enough for this sort of job. I use the phlatscript (or sketchUcam as it's called now) - produces gcode from the sketchup file - works extremely well - and good for me too as I use a Mac :D One of the nice features of sketchUcam is that it produces 'tabs' to keep the objects in situ while cutting. Only found a couple of minor errors in the cutfile, but worth doing just to check tolerances etc.

I'll overdraw the parts to up the facet count, but I don't think I'll bother recutting a test set. I found a new sheet of 6mm clear polycarbonate I forgot I had - maybe I'll make them clear rather than white :D

add commentadd comment in the forum

reader comment Comment from: TLHarrell on Sunday, September 4th 2011 - 12:23 AM
I've actually considered messing around with some metal slat blinds from IKEA. They're way cheap. I picked up a small one for like $5. Just haven't had the heart to chop it all up yet.

Saturday, September 3rd 2011 - 10:16 AM

While waiting on my test MDF parts being cut this morning, I've been searching through alternatives for honeycomb. One of the lasersaur machines in the US has been using aluminium eggcrate - used for lighting and ventilation in suspended ceilings. I've found a fairly local company that sells it 'up here' ...

http://www.clanma.co.uk/eggcratecore.php

Probably cheaper places too - haven't seen it on evilBayUK 'though. I'll make some enquiries on monday - looks like a good alternative - steel would be nice.
I've also seen it used in vertical cleanroom partitions for laminar flow.

add commentadd comment in the forum

Friday, September 2nd 2011 - 10:35 PM

Should never have mentioned that the metalwork was to arrive today - obviously cursed - despite emails to the contrary, the stuff is not going to arrive now until Monday - a pox on them I say :-))
Maybe that'll give me time to do a dummy run on some cheap 6mm MDF before I run the semi-precious acetal sheet through my CNC router.

Forgot to mention that the table centre can be easily replaced with various tops - I'm looking for some cheap alternatives to honeycomb. My experiences with honeycomb section is that it gets 'dinged' very easily - a simple steel grid would probably be better (10mm?) - could then use magnet hold-downs if necessary.
I was looking at the Lasersaur site tonight and noticed they are using what looks to be drywall corner bead for a base - quite a practical idea if you can get a small enough section (20mm ?). However, checking the local cost of corner bead would put it in the same ballpark as aluminium honeycomb and firstborns.

I'm fortunate to have a lathe to turn the 8mm steel shafting down to 6.35mm to suit the bearings - 1/4", <rant> why do we still use these archaic imperial measurement systems :-)) </rant>

Cheers

Neil

add commentadd comment in the forum

reader comment Comment from: TLHarrell on Friday, September 2nd 2011 - 5:02 PM
Very nice. I'll be keeping an eye on this one as I too would like to build a 24"x48" machine once I get my little 40W unit producing a stream of revenue (total income so far... $65). I've got an interesting idea for the table lift though. I intend to use Sarrus linkages in the table corners instead of rods or rails, and use a dual cam lift in the center of the table.

Friday, September 2nd 2011 - 1:47 PM

Hi

Thought I'd start a quick buildlog as my metalwork should arrive today.
System has a Coletech 60W laser along with their better quality optics etc. Laser and ancillaries arrived a few weeks ago - end jacket broken on the laser unfortunately, but another arrived within 5 days. I have to mention the excellent service I have received from Coletech - at all times the communication has been rapid and accurate ('Shirley' must be a great worker to have in the office :-) ). I ordered my parts from China on the Wednesday afternoon - they were on my doorstep on the Friday afternoon - from China - and using DHL !!! That is outstanding service - I can hardly get that service from suppliers here in the UK.

My machine is similar to the 2x and the Lasersaur (among others) - 1220x610x200mm (48x24x7.8 inches) cutting area. It is floor standing with all the ancillary stuff underneath - air assist pump/cooler/electronics and extract. There's a 'pod' on the front which will contain an LCD and some control buttons eventually.
Rails are vee rails (RS) with Barts original acetal vee wheels - I'd have liked to use the new makerslide rails, but the shipping would be horrendous. It's conceivable that I might convert at a later stage as the bosch stuff is compatible.

Skins are set into the frame using magnets rather than outside with screws (the magnets are let into the dibond skins [similar to aluply]). The side skins are quickly removeable to allow access for long material. There's also a lower 'crumb tray' to catch all those parts that you forgot to add tags for :D

I'm using Bosch/Rexroth extrusions rather than misumi - I can get it all cut from KJN without having to buy stock 3 metre lengths and work out my own wasteage - all online ordering too.

One annoyance is having to use 6mm chain and sprockets for the Z axis lift - I just couldn't source a closed belt that was big enough. I'm sure it'll be fine (if a tad noisy :-) )

I'm currently at a crossroads regarding the controller - I really want to be able to use it with the Mac (and definitely NOT using a PC emulation :-) ) Current options are the LAOS board - I have one of the beta boards which I'm building - it's essentially a CUPS printer which converts to a form of gcode - acts like a normal printer interface - mBed controller onboard. Other option is the Lasersaur unit which is arduino based with a web based interface and importer - I lashed one up - it worked first time and has a VERY clean interface - currently uses svg format files as input - converts to gcode and thence to grbl on the arduino. How this will shape up to 'engraving' will remain to be seen (the arduino isn't that powerful), but the vector work is very clean and fast.
I don't want to go down the DSP OEM board route at present as they are all PC only (and will probably remain that way for some reason). I have a DeskCNC board (PC based :-) ) which works REALLY well on my old CNC router and I have been reliably informed that there are Mac and Linux versions to appear. It has PWM laser output and can also handle 'engraving'.

I'll attach some rough renderings (definitely not up to the 'Bart Dring School of Industrial Design' standard LOL but they'll suffice for my needs :D )
They're also partially incomplete as I don't get too anal about small things like shaft coupers etc.

Cheers

Neil

PS Blackhole was chosen for a couple of obvious reasons, it creates blackholes and that's where all my funds have gone :D
Attachements...
Screen shot 2011-09-02 at 14.32.57.png
Screen shot 2011-09-02 at 14.34.47.png
Screen shot 2011-09-02 at 14.34.04.png
Screen shot 2011-09-02 at 14.33.21.png
Screen shot 2011-09-02 at 14.32.36.png

add commentadd comment in the forum

About Us | Contact | ©2010 BuildLog.Net