BuildLog.Net - Document your builds!


Buildlog Title: Educa's BigFoot

newest first oldest first
Builder: educa
Member Since: 2011-12-22

Wednesday, December 28th 2011 - 10:45 PM

Hi all,

My name is Bart, I'm from Belgium and I intend to make a laser cutter.

This site allready helped me a lot in finding information and now I'll start making my own machine.

The purpose is to be able to accept materials upto 121x61cm (about 4x2 foot, therefore the name of the project BIGFOOT)

I am currently reading a lot of buildlogs and will soon post my ideas and progress.

The machine will not be a 2.x machine, but a lot of idea will be taken from it of course.

add commentadd comment in the forum

Thursday, December 29th 2011 - 9:48 PM

Ok here we go with the first real info.

I see on the buildlogs that you guys seem to use misumi 2020 aluminium extrusions.
I did a lookup and found that 2020 extrusions seem to be 20mm by 20mm big. I use the word BIG, but in my humble opinion this is rather SMALL.

Don't you guys have problems with the stiffness of your frame ?

Here in Belgium I didn't even find a source for 20x20 extrusions. The standard is here 35x35 or 45x45

ImageImage

What do you think? Would 35x35 be enough for my frame ?

add commentadd comment in the forum

Thursday, December 29th 2011 - 10:32 PM

Now my next material of choice is the slides where the movement is going to be. I see that most people here glue V-rail onto aluminium extrusions and then use grooved wheels. While this might be a good (and cheaper) solution, I do have good experience with ball bearing slides, so I'll go with these since they have only about 5 microns of play between slides and carriages.

The gantry axis will be built upon 35x35 alu extrusion and will use SBC linear rail measuring 1250mm. The carriage is 135mm so I won't be able to cut all the 121cm I have on the material plate. That is not such a big deal however since I will certainly be able to travel 1000+mm and I only want to be able to put quarter E1 plates (121cmx61cm) into the machine so I don't have to use a sawblade and can use the sheets directly from the store.

This linear rail is very strong with width of 34mm and height of 29mm precision hardened steel, so I don't think in combination with 35x35 aluminium extrusion that this will bend in any way in the middle.


Image

Image

Image


For my smaller rails which carry the gantry, I'll use THK ball bearing precision guides SHS 20 750mm long with each 2 carriages (only 1 will be used). A Carriage is about 10cm, so 75 - 10 = +- 65 cm movement possible.

Image

Image

This looks like a solution which should be very high precision. It is however not LOW cost .

The rails themselves cost (second hand) about $350 , but then again, I want to be sure that this part is rock solid.

add commentadd comment in the forum

Thursday, December 29th 2011 - 11:04 PM

To be able to move the axes I'll also need stepper motors of course.

So these are the weapons of choice.

Since my machine will be quite big and the rails are also not light stuff, I will use NEMA23 motors with a torque of 3.1Nm . Thats very high torque so I suppose I won't get lost steps when I run the axes.

The power comes from a 400Watt 36V 11A PSU unit

Image

Then I control the steppers with a CW5045 stepper controller capable of driving upto 4.5 A and doing microstepping in both imperial and metric scale
1 , 1/2 , 1/4 , 1/8 , 1/16 , 1/32 , 1/64 , 1/128 and 1/256
+
1/5 , 1/10 , 1/25 , 1/50 , 1/125 , 1/250
This would allow for very precise control, but tests still have to make clear what microstep settings are acceptable, since higher microstepping makes steppers smoother, but reduce torque

Image

The stepper motors themselves are NEMA23 3.1Nm steppers with dual 8mm shaft
They will be driven at bipolar parallel connection and draw 4.2 amps

Image

The pulleys will be 20 tooth T5 pulleys (5mm tooth spacing) and the belt is T5 Polyurethane Steel reinforced open loop.

Image

Image

If the pulley is 20 tooth T5, then 1 rotation is 100mm

The motor is 1.8° 200 steps motor, so I should theoretically get following resolutions when microstepping

@ 1 step: 0.5mm / step
@ 1/5 step: 0.1mm / step
@ 1/10 step: 0.05mm / step
@ 1/25 step: 0.02mm / step
@ 1/50 step: 0.01mm / step
@ 1/125 step: 0.004mm / step
@ 1/250 step: 0.002mm / step

I think 1/25 step will be enough since I will mostly design vector cut stuff in coreldraw and PLT is 0.025mm accurate (1/40 mm)
Also, I don't see a point to go higher resolution for engraving since a single dot will be 0.15 tot 0.20mm anyway.

add commentadd comment in the forum

Thursday, December 29th 2011 - 11:13 PM

By the way, if anyone has any tips/hints for me while reading these blogposts, the please feel free to post them here. I am willing to learn from anyone!

add commentadd comment in the forum

reader comment Comment from: dirktheeng on Friday, December 30th 2011 - 3:04 AM
Bart,

Wow, you got some good stuff so far! I really want to get better pulleys for my system. I'm not a big fan of the plastic ones I have now... not real accurate and I think they are the source of some backlash.

One thing you will want to do is keep the mass of your gantry cart to a minimum because that is what you want to move fast to do engraving... make sure the mirror assembly is good and stiff too.

Also, I think there is good reason why the printer manufacturers use an encoder strip with the print heads with the belt systems. I think that they can't rely on steppers to get to the sub 5thousanths of an inch they need to do really high resolution printing. I am looking into putting in an encoder strip for engraving. so I don't have to worry about the backlash issues. It seem that most people can't hold much more than 20thousanths with this system and some have had some issues with engraving without compensating for backlash.
reader comment Comment from: BenJackson on Friday, December 30th 2011 - 4:18 AM
dirktheeng wrote:Wow, you got some good stuff so far! I really want to get better pulleys for my system. I'm not a big fan of the plastic ones I have now... not real accurate and I think they are the source of some backlash.


I blamed the wobbly idlers until I had the smooth idler. Finally I cranked up the tension on my X much higher than I had previously and it helped quite a bit.

Friday, December 30th 2011 - 11:39 AM

Does anyone know what benefit I would get by using a beam expander like tis one on ebay?

http://www.ebay.com/itm/CO2-10600nm-10-6um-2-5x-Fixed-Magnification-Beam-Expander-f-Laser-Marker-Marking-/120812240030?pt=LH_DefaultDomain_0&hash=item1c20f87c9e

add commentadd comment in the forum

reader comment Comment from: mikegrundvig on Friday, December 30th 2011 - 1:07 PM
Wow, you have an almost identical plan to me with the rails, higher voltage/larger steppers, and even the drivers/power supply - we are both using the same for the most part. I LOVE your pulleys - I was not that impressed with the nylon pulleys from SDP-SI and I feel they are the "cheapest" grade mechanical part in my design thus far.You have chosen MONSTER rails, large enough to run a good sized CNC mill. Is there a reason you went so large? I'm using 15mm width rails at the recommendation of a friend of mine and even they are rated in the hundreds of lbs. Rails also have a pre-loading that's not inconsequential if you don't have much weight on them. The larger the rail, the heavier it gets. I'm a little concerned that the pre-load will be problematic on my gantry head (x axis) as it's going to take more power to overcome than friction.

As for extrusion rigidity, I'm VERY concerned about this and have spoken with the engineers at 8020 (I've got a great local distributor for em so I'm going this way rather than Mitsumi) and they directed me to a load calculator they have produced on their website. It was/is incredibly useful at determining how much flex and torsion you will get out of the extrusion for a given weight/length/load profile. Long story short, the rigidity of the extrusion isn't the problem if you go reasonable. The problem is how you mate various pieces of extrusion to each other. The "core screw" approach combined with mounting plates is what they suggested as being much stronger all around. Also, the "inside corner" joiners get progressively stronger the more of the corner they consume. Specifically, you start with the 90 degree simple ones they are quite flexible. Then they have the ones that form a small triangle and are pretty good. The strongest they offer actually slide into the t-slot itself with a leg between the two pieces of extrusion. You can just see them in this video. I'm not positive these are stronger than the basic triangle design I linked before but the guy in that video swears his frame is bulletproof and as he needs far more consistent accuracy at a higher speed than us, I suspect he is right. For reference, steppers are not fast enough for his application so he is using large DC servos. Supposedly these are the strongest for the outside corners and look nice too. You can still support these with inside corner brackets as well to ensure an even stronger joint.

Whew, lots of stuff. I'll be watching your thread like a hawk! Thanks!

-Mike

Friday, December 30th 2011 - 1:53 PM

Mike, the reason I took such large rail for X is because thats the one I could buy on ebay. These rails are extremely expensive so you take what you can get :-)

The steppers should have no problem at all at 3.1Nm to pull this 1 carriage.


For alu extrusions, this is what I can get here locally. They have also very strong connectors.

http://alumes.be/pdf/catalogus_nl.pdf

PDF is in Dutch, but drawings are drawings :D


I'm still not sure if I'll take 35x35mm or 45x45mm rails.

45x45 is double the price of 35x35

add commentadd comment in the forum

reader comment Comment from: dirktheeng on Saturday, December 31st 2011 - 2:54 PM
educa wrote:Does anyone know what benefit I would get by using a beam expander like tis one on ebay?

http://www.ebay.com/itm/CO2-10600nm-10-6um-2-5x-Fixed-Magnification-Beam-Expander-f-Laser-Marker-Marking-/120812240030?pt=LH_DefaultDomain_0&hash=item1c20f87c9e


it will lower the energy density of the beam on your optical components and make them run cooler... not a problem with these lower power lasers though and even if it is with 80+ watt lasers, it is almost cheaper to get better mirrors.

Saturday, December 31st 2011 - 4:23 PM

OK then its not of good use for me.


I was thinking of also mounting a red laser pointer to see where the machine will start engraving.
Instead of using a beam combiner (expensive + 1 more step to align + some power goes lost with such a combiner) I would mount the laser pointer onto the moving laser head and let it point down.

Then I can use the red pointer to point to the working table and if I know the offset between the laser and the real laser beam then my software can calculate the difference and start cutting at the exact place where the pointer was previously pointing to.

Price for this all: $5 or so

add commentadd comment in the forum

reader comment Comment from: BenJackson on Saturday, December 31st 2011 - 8:04 PM
This guy had a clever solution for the 2.x laser: http://danielbauen.com/make/index.php/l ... n-extra-m/

Basically a pivot arm that drops a laser down between the CO2 tube and the first mirror when the lid is open:

Image

Personally I am avoiding any additional weight on the laser carriage (it needs to accelerate very fast for engraving) so I haven't done anything like you describe. I've actually been considering the "knife edge mirror" technique that's described in another forum post.

Sunday, January 1st 2012 - 7:27 PM

That looks like a nice solution.

Maybe I should consider something like that.

add commentadd comment in the forum

Sunday, January 1st 2012 - 7:37 PM

Here I am again with a dillema.

The always returning dillema of HOW MUCH laser power to put in your machine.

I do know that 80Watt will be a lot more powerfull then 40 Watt, but I also have the following problem.


I intend to make the software for the machine myself. That is I want to make my own sort of cnc controller which accepts gcode + a subset of my own to define engraving.

The software will run on windows machines, but communicate through (probably) and arduino to some electronics.

I want my software to be able to both CUT and ENGRAVE in the same session.

For example a few posts ago in this thread you see my dragon image. This dragon is made out of vector engraved lines + vector CUT lines. Or course in real life I would first engrave and then cut.
But.... cutting is preferably done at full power and engraving possibly not.

So, at what power setting should I do engraving?

I intend to use the PPI technique for cutting, but isn't it so that for engraving you have to provide both very short pulses and secondly turn the power of the laser way down?



It's still not crystal clear to me how this exactly works. The Power supply has both a TTL High and Low input, but also a way to setup laser power by an analog voltage.

Does this mean that if I control with an arduino, it is best for me to connect 1 output pin of my arduino to the TTL Low (or is it high) and then connect the arduino through some DAC to that 5V analog input on the PSU to control laser power?

If for example I use a cheap 8bits DAC (or even an R2R ladder) then I could setup the voltage to the laser PSU in 256 steps (0->5 V in +- 0.02V increments) and in case of a 40 Watt laset I would control the power from 0->40 in 0.15 Watt increments.

Is that how it works? Or does such a PSU not react to 0->5 but to a smaller range of voltages? Maybe there is a minimum voltage needed to shoot a laser at all ?
I would really appreciate if somebody could xplain this a little.

Thank in advance,

Bart

add commentadd comment in the forum

reader comment Comment from: awesomenesser on Sunday, January 1st 2012 - 7:56 PM
educa wrote:...
If for example I use a cheap 8bits DAC (or even an R2R ladder) then I could setup the voltage to the laser PSU in 256 steps (0->5 V in +- 0.02V increments) and in case of a 40 Watt laset I would control the power from 0->40 in 0.15 Watt increments.

Is that how it works? Or does such a PSU not react to 0->5 but to a smaller range of voltages? Maybe there is a minimum voltage needed to shoot a laser at all ?
I would really appreciate if somebody could xplain this a little.

Bart


Yes that would be how it works there is a minimum turn on but you could calibrate the software to add that offset.

One recommendation I would not use and arduino to power for your custom controller. I would recommend an FPGA, AVR32, XMOS or just about anything else faster than an 8bit AVR. Also depending how you do it you might have issues with the speed of communication from the pc to the micro controller. MACH3 which transfers the direct pulses has to use custom realtime drivers to be able to achieve those speeds. I would recommend that you just transfer (USB to RS232) your gcode and engraving information to the controller and it handle the pulses to the motors. I have many projects which transfer from a custom c# windows application through usb to a FTDI FT232RL chip which then outputs TTL 5V serial data for transfering data to a AVR32 (I have a little information on my site http://robertgivens.com/mediawiki/index.php/SyncPOV_Software).

This is a large project to undertake and I look forward to seeing the results.
reader comment Comment from: BenJackson on Sunday, January 1st 2012 - 8:05 PM
educa wrote:For example a few posts ago in this thread you see my dragon image. This dragon is made out of vector engraved lines + vector CUT lines. Or course in real life I would first engrave and then cut.

I think when you say "engrave" most people are going to think of raster engraving (rapidly sweeping back and forth to produce a bitmap).

When you make vector lines like your dragon it's just like the cutouts except for the power and speed.

educa wrote:But.... cutting is preferably done at full power and engraving possibly not.

You may want less than 100% power even for cutting. Something thin like paper, for example. With an 80W tube more things will be "thin" for you than for me.

educa wrote:Does this mean that if I control with an arduino, it is best for me to connect 1 output pin of my arduino to the TTL Low (or is it high) and then connect the arduino through some DAC to that 5V analog input on the PSU to control laser power?

TTL low and high are identical and just for convenience. Yes, you will want analog control of the IN terminal. A DAC isn't really necessary. An AVR can easily PWM that pin fast enough.

Also only some of the voltage range is usable. On my laser it doesn't fire below 0.5V and full power is about 3V. It's also nonlinear. About half the power is between 0.5 and 1V and the other half between 2 and 3V. See the graph in my buildlog.

Sunday, January 1st 2012 - 8:21 PM

ok if the avr can pwm this fast enough, then its 1 DAC saved in the wallet.

When I say ENGRAVE I really meant raster engraving. thats why (in the case of my dragon) I specially mentioned vector engraving.

In the photo's above the dragon is cut on the edges, but the vector engraving lines are also cut quite deep, but I suppose with the right settings (and probably with some PPI tricks too) I could make the laser MARK the wood without cutting in it ?

In the end I would like to make a machine which can engrace (be it raster of vector engraving) some pattern on wood and then cutout the contours in 1 machine operation, so the machine will switch between engraving and cutting in my software.

add commentadd comment in the forum

reader comment Comment from: TLHarrell on Monday, January 2nd 2012 - 2:48 AM
I've been using BOTH raster and vector engraving. On my machine, when vectoring MDF, I set to about 2% and 100% speed. You can see the results of both on the fin parts shown in this other post (sorry for cross posting, but I didn't want to upload the photo again) http://www.buildlog.net/forum/viewtopic.php?f=16&t=751&p=6828#p6828

Monday, January 2nd 2012 - 11:44 AM

It looks very good. I allready saw that rocket lamp by the way, thats a very nice creation.

I have another problem.
For my machine I will use ball bearing linear rails. These are quite thick and very strong things, but I wonder about the motor to drive these carriages.


Please let me make some calculations and if I'm wrong please correct me.


The motor I will use is a nema23 3.1Nm stepper motor.
So technically speaking this motor would be able to hold 0.316112026kg (316 grams) at a distance of 1 meter.

My pulley is a 20 tooth 5mm pitch pulley, so 100mm around = a pitch diameter of 100/PI = 31.83mm or 0.03183 meter. So the pulley would be pulling the belt at a distance half of that = 0.01591 meter

1/0.01591 = 62.85, so I assume I then get a pulling torque of possibly 62.85 x 316 grams = 19860.6 grams or 19.8606 Kg


Can I then safely assume that my motor will have no problem at all to pull my bearing block ? I think that I need less then 1Kg of force to make the bearing block slide in horizontal direction.

Hopefully my calculations are correct.

add commentadd comment in the forum

reader comment Comment from: TLHarrell on Tuesday, January 3rd 2012 - 1:43 AM
Sounds to me like you've got plenty of pull. Make sure your belts can handle that amount of tension. Those linear rails, unless they have a major preload on them, are fairly close to frictionless. The bigger number to deal with will be mass calculations for acceleration.

Tuesday, January 3rd 2012 - 8:58 PM

Ok here we go again.

Today my X axis rail arrived from Germany.

Man this thing is HUGE. It weighs about 8 kilograms and is very very heavy.
Also, the linear ball bearing slide has (unfortunately) mechanical artifacts in the rail which make it not run very smooth.


So..... I allready settled with the ebay seller and the rail goes back. Meanwhile I allready bought another rail (this time 1400mm long) of the type THK RSR12WZM
This rail is a lot lichter (only 24mm wide and 8.5mm heigh) and the carriages only weigh <70grams

Still they are top quality (I allready have some of those here which measure about 30 cm, but its the same rail and carriages and I can really make them fly)

So its waiting again for that rail to arrive.


In the meantime, I have discovered http://www.beltingonline.com/ to be a site to bookmark.

I found there that they have T2.5 AND T5 belt. I really wonder if T2.5 wouldn't be a good solution for me instead of T5, but that largely depends on some stuff I'll have to ask to beltingonline (mainly the fact that I don't have the tools to accurately make my own bore + grubscrew holes.

If I use 2.5T 20Tooth pulleys, then 1 rotation of my motor in full step would be 50mm.

The motor is 200 steps motor, so 1 full step would be 50/200 = 0.25mm

The driver (a CW5045 microstepping driver) can do 1/10 , 1/25 , 1/50 step to the 3.1Nm stepper so theoretically that would give me a resolution of 0.025 , 0.01 or 0.005 millimeter (where 0.025 is the default resolution of HPGL files which only work upto 1/40th of a mm)



Life looks nice again.

Unfortunately I also got the idea to begin to draw my machine in sketchup and that is not so easy at all :)
Therefore, the first thing I'm going to do now is read a good sketchup book :)

add commentadd comment in the forum

reader comment Comment from: BenJackson on Tuesday, January 3rd 2012 - 9:23 PM
Just buy Alibre Design Standard.

Tuesday, January 3rd 2012 - 9:57 PM

Well not to act stupid, but its $199 and I don't want to spend that much on this part of the job :)

The design is more like a proof of concept try.

add commentadd comment in the forum

reader comment Comment from: BenJackson on Tuesday, January 3rd 2012 - 10:27 PM
But it's a vastly superior tool if you want to make mechanical designs. Perhaps HeeksCAD has caught up enough to get you by if you really need free.

I have also used SketchUp to do what you're doing and it's night and day. Alibre (and other parametric solid modelers like SolidWorks) really capture the intent of what you want to do ("put a hole 3mm from the edge" or "align this notch with this other part") and everything ripples through when you change the original constraints. Just try to resize a hole in SketchUp... Plus when you're done with your design Alibre is ready to output design documents, 2D sections for machining (or direct connection to CAM), etc. I also output STL to directly 3D print designs.

Alibre could easily save you $199 by helping you find interferences that might otherwise lead to scrapped parts. Being able to actually build the assembly such that you can drag the laser carriage and gantry around and see where everything goes will save you a ton of time as well (and probably lead to simple tweaks that improve your cutting envelope).

Tuesday, January 3rd 2012 - 10:35 PM

I see.

I just found that there is a 30 day trial version available so its actually quite risk free.
If after 30 days I'm still not convinced you can always ask for another 30 days :)

Now even if I use alibre or sketchup, I need to be able to learn it.

Are there any ebooks on alibre you could advise ?
I'm going to look at the tutorials now

add commentadd comment in the forum

reader comment Comment from: BenJackson on Tuesday, January 3rd 2012 - 10:59 PM
There are tutorials and once you get the "idea" of parametric solid modelling it's pretty intuitive. Even if you don't know a function exists you will suspect it should so you'll know to look for it.

If you see any tutorials on: http://alibre.dk/training_support/freea ... aining.asp which are not free (grayed out) it may be that someone like me has had access to them via one of Alibre's renewal incentive programs and has... viewed them... permanently... and might be able to recall them in vivid detail in the form of a .mp4...

A few of us Alibre users have uploaded designs to Thinigverse you could look at. Another advantage of parametric CAD is that you can look at those designs and see just how we did everything.

Alibre also has decent forums and Bart might be willing to put up a CAD/CAM forum here.

BTW if you have not picked a 2D CAM you might consider CamBam -- there's a plugin that one of the users made that lets you transfer sketches (the 2D foundations of most parts in Alibre) directly into CamBam. It has a very, very generous free trial. In fact, it is so generous that I have 75% of it left and I really should just buy it.

Wednesday, January 4th 2012 - 12:43 AM

Ben. I'm currently downloading the alibre design user guide to study it.

a few questions to you:


1) did you design your machine in alibre design? care to share?

2) do you know if the personal edition is enough to simulate gantry movement when you turn the stepper motor axle in the assembly?

3) Can this software also easily design the timing belt somehow?

4) Any idea if there is a parts library somewhere which could contain usefull parts for our purposes, like alu extrusions, nema 23 steppers,....

5) Any idea if you can export the completed machine design to some kind of 3d format usable for rendering in for example blender?

add commentadd comment in the forum

reader comment Comment from: BenJackson on Wednesday, January 4th 2012 - 1:30 AM
educa wrote:1) did you design your machine in alibre design? care to share?

2) do you know if the personal edition is enough to simulate gantry movement when you turn the stepper motor axle in the assembly?

3) Can this software also easily design the timing belt somehow?

4) Any idea if there is a parts library somewhere which could contain usefull parts for our purposes, like alu extrusions, nema 23 steppers,....

5) Any idea if you can export the completed machine design to some kind of 3d format usable for rendering in for example blender?


1) Not the laser cutter. I built bdring's design.

2) I don't know if there's a rotary-to-linear constraint. I don't think the personal edition is missing any constraints at the assembly level, though.

3) I've never designed a belt. Bart (Dring, that is) would be a good person to answer this. He makes lots of renderings with belts. I'm assuming they're cosmetic. The software could definitely model a belt as an extruded section that reaches from A to B and update that dynamically, but you'd probably have to reason from that about overall belt length. I'm not sure.

4) McMaster has lots of their parts online. I imagine others do as well. Misumi and 80/20 make their profiles available via 3D warehouse 3rd party sites. Things like steppers aren't that hard to model yourself if you just want the right shape. If you want them to be "pretty" you can find STEP files with a lot of detail (but then motor body length varies so beware of that)

5) You can export to lots of 3D formats including STL and I think whatever that 3DStudio format is. The pro version has more import/export formats as well.
reader comment Comment from: mikegrundvig on Wednesday, January 4th 2012 - 3:22 AM
I can't speak to Alibre but in SolidWorks, belts and the like can be actually added as needed and it will properly handle gear reduction, etc. I use this to tell me needed belt lengths for instance. You can also use it to determine if there are collisions with moving components and the like.

-Mike

Wednesday, January 4th 2012 - 11:48 AM

Well I read some documents yesterday and tried alibre (well, the basics)

It looks like a nice program. There are still things I didn't find a solution for however, but it does look nice.

This is my first drawing. Some 35x35mm alu extrusions which I then aligned and mated to form an (strange) assembly :)

Image



I wonder now. This assembly is made with 3 times the same part and the part is 10cm long.

But what if I want a part to be 10cm , another 20cm and the third 50cm? Would I need to draw 3 different parts for that and save as individual part files ?

add commentadd comment in the forum

reader comment Comment from: mikegrundvig on Wednesday, January 4th 2012 - 2:54 PM
Again, no Alibre expert but look for the concept of "configurations" for the part. Something that lets you control the configuration of the extrude step to create multiple identical parts with other that difference. Then in the assembly (the drawing with multiple parts) you choose the configuration you need for a given instance of a part.

-Mike
reader comment Comment from: BenJackson on Wednesday, January 4th 2012 - 5:34 PM
mikegrundvig wrote:Again, no Alibre expert but look for the concept of "configurations" for the part. Something that lets you control the configuration of the extrude step to create multiple identical parts with other that difference. Then in the assembly (the drawing with multiple parts) you choose the configuration you need for a given instance of a part.

You can't create (but you can manipulate) configurations with Standard. I didn't find configurations to work well for extrusion lengths. I think a different part for each length with a different p/n (and mfr p/n) is the way to go. Plus one of your rails might be "extrude to geometry" instead of "to depth" so that it automatically spans the front of your machine.

One of my initial mental hurdles with Alibre was learning to relax and love the giant part lists.

Wednesday, January 4th 2012 - 6:12 PM

partlists ? where ?

add commentadd comment in the forum

Sunday, January 8th 2012 - 8:39 PM

Hi,

For my machine I want to begin with a 40watt laser tube but the machine will have place for 40watt , 60 watt or 80 watt at a later point.

Now I'm looking to buy my laser head since first thing I want to fully make is my X axis.
Therefore I found the following offer by cnccoletech

http://www.ebay.de/itm/CO2-Laser-Head-M ... 230d0c8a2c

My machine will have a cuttable area of aprox 48" by 24" running on precision ball bearing rails. I can machine my delrin parts on a high precision cnc machine, so alignment should be doable.

Now I read here and there that 48x24" is quite big so its better to have good quality optics and mirror mounts with fine thread pitch.



So here are my questions now:

1) The mirror mount supports mirrors of 20,25 and 30 mm (I think I have to choose on order). What would be best for me ? I personally think I would be best off with 25mm? Any advise ?

2) The mirrors are SI Coated gold. For 40Watt thats certainly not a problem, but will this withstand 80watt also at a later point ?

3) For lens I think best option for general usage is 50.8mm focal distance ? I do plan on cutting too however, but I will hardly cut anything thicker then 5mm (1/5") and materials are acrylic and plywood/mdf. I still have to ask cnccoletech what diameter the lens has because probably thats also important ?

4) The lens is probably ZnSe based. Is this ok for 80 watts ? (it certainly is for 40 watts)


Kind regards,

Bart

add commentadd comment in the forum

reader comment Comment from: artwood_decor on Monday, January 9th 2012 - 2:17 PM
[quote="educa"]1) The mirror mount supports mirrors of 20,25 and 30 mm (I think I have to choose on order).[/quote]

For 25mm mirrors and lens I got a quotation USD220+USD40/shipping.

Monday, January 9th 2012 - 3:24 PM

You got that quote from cnccoletech ?

add commentadd comment in the forum

reader comment Comment from: artwood_decor on Monday, January 9th 2012 - 3:34 PM
Yes.
It was for this listing:

http://www.ebay.ca/itm/CO2-Laser-Head-M ... 230d0c8a2c

and here is my request:
1x Laser Head
1x High quality lens, focus length 50.8mm
2x Reflect mirror mount 25mm
3x High quality Si coated golden mirror, diameter 25mm

So the price goes up with diameter...

Monday, January 9th 2012 - 3:45 PM

and thats ok upto 80 watts of co2 laser?

add commentadd comment in the forum

reader comment Comment from: artwood_decor on Monday, January 9th 2012 - 4:08 PM
Didn't ask about power, but my assumption was the beam diameter goes up with power. (I might be wrong)

Monday, January 9th 2012 - 5:47 PM

Is that lens then also 25mm ? or smaller?

Did you buy it from coletech ?

Regards,

Bart

add commentadd comment in the forum

reader comment Comment from: artwood_decor on Monday, January 9th 2012 - 6:01 PM
Good question but I have no answer :oops:
Didn't buy it yet.

Monday, January 9th 2012 - 6:35 PM

Ah OK, I am going to ask quote now and the immediately order, since I do believe the measurements of these pieces are interesting for making the frame, so it best in my opinion just to buy them :-)

Today mu 1.4m THK linear slide arrived and it is running smooth as butter.
So I now have my 2 750mm THK slides for the Y axis and 1 1400mm THK slide for the X axis.

Next step is the 35x35 and 45x90 alu extrusions and then the cnc will have work to do.

I'll also order now the 3.1Nm nema23 stepper and aluminium pulleys + T5 belt.


Life is good!

add commentadd comment in the forum

Thursday, January 12th 2012 - 12:18 AM

Today I designed from scratch the 35x35mm aluminium extrusion profile I'll use for the machine + the 750mm thk ball bearing guide in detail.

Sketchup works nicely for this.

Results below.

add commentadd comment in the forum

Friday, January 13th 2012 - 7:24 AM

My x-axis stepper motor arrived today.

Man this is a beast. It has 3.1Nm of torque and I managed allready to make it rotate 10 times per second on 1/25 microstep mode.

1/25 microstep = 5000 steps per rotation. On a 20 tooth T5 pulley that'll give me 100mm of movement in 5000 steps or 0.02mm accuracy (more then enough for me). Thats 0.0007874" accuracy

At this high microstepping and relatively high stepping rate (The motor driver gets 50000Hz steps) I still have soo much torque that I burned my hand trying to hold the shaft to stop it (note to self: don't try this anymore, it hurts)

So now it's waiting for the T5 open ended belt + the pulleys to arrive and then we can continue on the mechanical part.

add commentadd comment in the forum

reader comment Comment from: vincenttor on Tuesday, January 17th 2012 - 10:40 PM
i like this idea allot,
I wanted to make my excisting laser from FSE to a bigger format like 1200x900mm.
How are you planning to move the axis ?

something like this ?

Planning to use these rails but still thinking how and what i should do with the y axis because it looks pretty heavy moving the stepper fast with engraving.
Image

Tuesday, January 17th 2012 - 10:58 PM

I will just use T5 belts with 20 tooth T5 aluminium pulleys on 3.1Nm Nema23 stepper motors.

My working area is theoretically about 1300 * 640 mm , but I'll limit that to around 1220 x 610 mm , thats 1/4 of an official E1 plate of material and that is also the size in which I can buy my material without having to put it in a saw to make smaller pieces.

I will also use 45x45mm aluminium extrusions instead of 20x20mm for the buildlog machine for 2 reasons.

1) I cannot buy 20x20 here, only 35x35 and 45x45

2) My machine is a lot longer then the buildlog machine, so I surely must use stronger alu extrusions.

add commentadd comment in the forum

Wednesday, January 18th 2012 - 11:03 PM

Ok today I finished the initial design of the aluminium extrusions for my machine.



Currently I use 45x45mm extrusions and honestly I think these are a little overkill. I think I'll change to 35x35mm extrusions which are a lot cheaper and lighter and still a lot stronger then the default used 20x20 here on the buildog machine (who is smaller then mine)

I Also included a higher res picture

Image

A little detail, but the machine itself is 1600x1200mm big and the working area is around 1240 x 630 mm
The Z table will have an aproximate lift of 350mm

add commentadd comment in the forum

Monday, January 23rd 2012 - 12:48 AM

Here is my most updated design, which will probably be the final design for the machine.

I use very strong materials and parts, but thats mainly because they are quite eqsy to mount and surely won't hurt for the strongness of the machine itself.

Video is available in full HD on youtube. Comments are more then welcome

add commentadd comment in the forum

reader comment Comment from: Greolt on Monday, January 23rd 2012 - 1:08 AM
Just some comments. Only my ideas. Take them or leave them. ;)

How about getting the table lift belt drive below the floor of the chamber. Makes for a cleaner design and should make cleaning easier.

The space allowed for the laser tube is very large.

Smaller pulleys on the Y axis drive. You will not get useful resolution as drawn.

Greolt

Monday, January 23rd 2012 - 6:53 AM

Greolt, thanks for idea, here my replies to them.

1) Mounting the table lift belt will indeed be quite hard. I don't see however how I would get it below the chamber because that would mean I would have to go below the table top (the brown is a 38mm thick MDF tabletop).
I do expect however that once I mounted this, it won't need a lot of servicing, since this is 15mm wide polyurethane T5 belt with steel reinforcements.

2) The space for the laser is indeed quite big, but that also has its purposes. First of all I want space to go upto 80Watt laser tube. Secondly, in the lower chamber below the actual laser there is place for the electronics, suction system + water cooling system which will all be internal, so that space will surely be used :)

3) About the pulleys. These are T5 pulleys. 20 tooth on the X axis with a direct 8mm mount to stepper motor shaft.

20 tooth T5 = 100mm for 1 rotation. The motor is 200 steps 3.1Nm and in 1/25th microstepping (plenty of power and running very smoothly) this gives me a resolution of 100 / 200 / 25 = 0.02mm (1/1270 ") I think that that is plenty for a lasercutter. HPGL PLT files even have only a resolution of 0.025mm (1/40th mm)

On the Y axis I couldn't go with 20 teeth pulleys, because I use taperlock mounting method. These ar here 40 teeth pulleys, so
at 1/25th step they give me a 200/200/25 = 0.04mm (1/635") resolution and normally I should be able to force this motor in 1/50th step because the 3.1Nm motor is very powerfull and speed on this Y axis is relatively lower then on the X axis. So resolution would then be 0.02mm in both X and Y

For Z axis the leadscrew is 20x4, so 4mm per rotation. Even with the stepper in half step mode there I get 4mm / 200 / 2 = 0.01mm resolution, so I don't see any problems there.




So to rephrase. I understand your point about the difficult to mount table lift belt. I don't see a direct solution to it however since I cannot make 20mm holes in this alu extrusion or I'll loose all of its power.

add commentadd comment in the forum

Monday, January 23rd 2012 - 7:00 AM

By the way, I just had another look, but I can change the belt if I loose up 2 alu extrusions on both sides (the Z rail is still more then enough supported then) and then unscrew the bearings. Then i can just change the belt with no problems at al.

add commentadd comment in the forum

Friday, January 27th 2012 - 9:30 PM

Based on Greolts comments, I decided to redesign my complete machine structure to have hanging pulleys for the up/down table.

Design is not ready yet but pictures/video will follow.


In the meantime I have a question about exhaust fan. Its very important to extract all possible smoke from the machine and put it outside my house :)

I have a 12mm hose measuring about 2 meter and then I can blow out of the window.



Now the question. I read somewhere that you need about 100cfm , but would a fan like this one http://www.frozencpu.com/products/8147/ ... 6c435s1109 be something interesting ?

Would it make sense to put 1 fan IN the machine and 1 at the end of the hose in a push/pull config?


And what if I put 2 of these fans on top of eachother, am I right that this will NOT increase the cfm rate, but it will almost double the presure? And what would this be good for to increase this pressure?


Would I rather put 2 of these on top of eachother and then have 1 exhause hose, or maybe better have 2 of them besides eachother and then give each one a hose?



On the other hand I could also use a EFD inline ducted fan which is also used in model airplane business.

1 nice piece is http://www.hobbyking.com/hobbyking/stor ... oduct=4242 which still needs a motor but I think this beast could pump out HUGE amounts of power.
They also exist smaller.

The only big problem I have is that these motors seem to be rated mostly in grams of thrust instead of cfm, so its difficult to compare.
On the other hand I', pretty sure these brushless motors will outperform the pc ventilators by a very huge amount and they also offer the nice feature of being very easy to speed control (something some pc ventilators can also do of course)






I wonder, any hints/tips are more then welcome.

Bart

add commentadd comment in the forum

Sunday, February 12th 2012 - 11:37 PM

Today I took some time to have a look at implementing PPI in my software which will run on an atmel microcontroller.

I looked at Dirktheeng's arduino code and found out how he did it.
It is actually quite easy to do.


My machine has 20teeth T5 pulleys on the axis and I have 200step steppers configured in 1/25th step (even at that high microstepping they have incredible torque since they are 3,1Nm motors)
Now if I calculate right this meant that 1 step on the stepper motor will give me 0,02mm of travel on my axis (100mm/(200*25)) the 100mm being the travel of my belt with 1 rotation of pulley.

If 1 step is 0,02mm then theoretically my maximum ppi will be 25,4/0,02= 1270PPI if I let the laser pulse on each step.

But. because there is a difference in timing between a step which only moves 1 axis and a step which moves both X and Y, this will not be reachable. If moving a single axis counts for 1 step, then moving both axes at same time counts for aproximately 1.42 steps (Pythagoras)


I could actually put my stepper driver in 1/50th microstep mode and probably still get more then enough torque, so then I would be able to achieve upto 2540PPI , but is this really needed?

When I look at Dirks code now, then it looks like he calculates travel and then if that travel is over a certain toggle value, he shoots the laser for 1 pulse and resets that travel to begin counting again. I wonder however why he zeroes that cumulative travel variable, because if its OVER a certain value, then it means that it is at that value + a remainder, but by zeroing that remainder is lost ans so possibly also the next ppi calculation will be off by some amount.

Or is this all not exact rocket science and are aproximations more then enough?


In Dirks code I see a float ppiX which makes me think he has a 157steps per mm of travel on his machine. I have 50 steps per mm and if I set microstepping to 50 I have 100steps.

Would that mean that PPI is possible on my setup? Just by counting steps and adding 1 or 1.42 to a cumulative counter and looking for overflows passing a certain toggle value?



What are actually common PPI values used to cut through acrylic and wood ? Is it values like 1200 ? Or is it more like 400PPI ? This makes a big difference in accuracy here for my calculations.



For example, if I draw a straight line with only 1 axis and I shoot the laser every 5 steps then theoretically I would get 508PPI

Correct me if wrong, but doesn't that mean that when I setup a cumulative counter on my arduino or avr (still have to decide) which adds 1.0 for very step moved in only 1 direction and 1.4142135 for each move in 2 directions in the same step, then when I fire the laser every time this cumulative counter gets bigger or equal to 5, I will aproximately fire 508PPI with my laser?
That would make implementation very easy. Of course in that case I am hardcoding some values in my microcontroller, but it would make everything so simple and the hardcoded values can always be recalculated for another machine setup.





I hope somebody can shine a light on this a little.

Thank you,

Bart

add commentadd comment in the forum

Thursday, February 16th 2012 - 10:00 PM

Because there are some actions to be done with timing, I decided to move a little of the PPI logic off to a second atmega chip (end result will probably run on a $1.5 chip)

For PPI I have done tests in shooting times.

De motion control system will be an arduino, but the pulses to shoot the laser will go through the little extra chip.

This chip will understand 3 shooting modes initially (it is possible that I'll make this software configurable)


For PPI I will need a 3ms pulse
For Engraving I will need a lot smaller pulses like 0.4ms
For continuous work (dunno what I'll need it for if I have PPI) I'll have continuasly on.


So it works like this.

Whenever I want the laser to shoot, I let my arduino give 1 pulse to the extra chip. That chip receives the signal and aproximately 5.5 microseconds later it starts to shoot the laser and acts indepenent of the arduino.

After 3ms flat (I can do this accurately with about +- 104nanoseconds play) the laser automatically shuts off.


This means I will take off a lot of processing hunger. It will be taken away from the arduino and put onto the separate atmel chip.

I even have space on that controler to implement full power control for the laser in an 8 bits setting, so theoretically I will be able to set the power of the laser from 0 to 100% in 256 steps or aproximately steps of 0.4%)


Now at this moment these timings of 3ms and 0.4ms are fixed programmed into the chip, but I'm looking to let the chip communicate with the main arduino so I can download other settings at runtime.


This will make the motion control so much easier and I really don't need a DSP for it as I can create accurate timings which only defer about 104ns from the wanted timings.


Thats looking promising for a $1.5 chip :)

add commentadd comment in the forum

Wednesday, April 18th 2012 - 10:06 PM

Roughly 2 months have passed by and it was very silend on my buildlog, but it was certainly not silent here :)

A lot of part have been ordered and I have now begun assembling my aluminium extrusions.
The machine is quite big (therefore its name bigfoot) and I must say I'm thrilled with the quality of the extrusions I bought. They are mostly 35x35mm (yes, the normal buildlog 2.x extrusions are only 20x20) and for some parts I use 45x45mm

There is around $1300 of aluminium extrusions used in this beast and I think you can lay a cow on it and it will not bend :)

The slides are all ball bearing linear THK precision slides and the z-table is working in precision spindles. Every part needed is cnc-milled very accurately to ensure no oddities.

Image
This is a left view of the machine

Image
Front left view. Notice the big lid to open the case (the lid will disable the whole machine when opened for safety, because there also happens to live a 7yo boy here)

Image
Lid opened

Image
View on a few linear ball bearing thk rails. These still need a lot of mounting using custom milled aluminium pieces, but thats not for this week :)

Image
Another view on ball bearing rails (the one on top will be X and the other one is 1 of the 2 Y axes). You can also see the up-down-spindle on this picture

Image
Closeup of the up-down-table mounting system. Precision spindle with steel net keeping the table in place and being able to lift it.

Image
Another view of the T5 pulleys. All mounting had been custom made

Image
Here you see a top-down picture showing the up-down table, but currenly there is not yet a perforated steel plate on it + the honeycomb material which will be 0.787" thick.



These are very exciting times because I am now spending all my time milling all these individual parts to mount the machine. Everything is in high quality aluminium so this machine will be extremely rigid.

add commentadd comment in the forum

reader comment Comment from: jkarpy on Wednesday, April 25th 2012 - 1:44 PM
I noticed you are using the Light Object Pro laser head mount. Are you willing to share the CAD file?

Wednesday, April 25th 2012 - 5:16 PM

Of course you can have the CAD model.

As a matter of fact, this site is a community to help eachother, so I will share my complete CAD model. It is made in Sketchup8 and shows my complete machine.

I hope somebody can do something with it.

Maybe my machine is not 100% recommended to build unless you have a nice budget, because I use very high quality 35x35 extrusions cut to a precision of 0.01mm + ball bearing slide rails (all THK) + all other parts are machined out of machining-aluminium with high rigidity. It is a bigfoot and its price tag will reach to the $4000 when finished, so I certainly didn't try to make the cheapest one.

On the other hand, I wanted to build the machine from scratch with a large cutting area and pro building materials.

Price was not important for this one :)



I attached my Google Sketchup 8 CAD file so please tell me if you could download it. The laser head + mirror holders are in the CAD file. I didn't buy the holder at lightobjects by the way but on ebay at CNCOLETECH
Attachements...
educa_bigfoot.zip
Google sketchup CAD file of my bigfoot lasercutter.
(1.91 MiB) Downloaded 1831 times

add commentadd comment in the forum

Thursday, April 26th 2012 - 12:03 PM

10 downloads allready of the cad file and no comments.

Is the cad file so good or is it just awfully bad? :)

add commentadd comment in the forum

reader comment Comment from: sports.racer on Thursday, April 26th 2012 - 2:20 PM
educa wrote:10 downloads allready of the cad file and no comments.

Is the cad file so good or is it just awfully bad? :)



More like 12 downloads. :)
I've added your parts to my Sketchup model. They seem so big in my design but the dimensions are right.
Thanks for the parts!
Attachements...
Capture.PNG

Thursday, April 26th 2012 - 2:27 PM

Looks nice,

I'm currently busy with mounting all parts so that will still take a considerable amount of time since all the special alu parts have to be milled individually.

Also looking now at a way to do limit/home switches using slotted optoswitches to eliminate mechanical contacts (+ they sound more sexy) :)

add commentadd comment in the forum

Friday, May 4th 2012 - 12:23 PM

There is a reasonable big chance that I will try to power the complete machine with a beaglebone

For the people who don't know a beaglebone, it is a 720Mhz arm cortex board which runs linux, has 256Mb ram, SD, Ethernet and there is are 2 extra onboard controllers with each 200Mhz speed which can do very precize timings.

Currently learning a lot but this device looks like a very capable candidate to do very high speed signal processing for the laser, especially the speeds needed for raster engraving and PPI.

Image

add commentadd comment in the forum

Friday, June 15th 2012 - 7:11 PM

Hi guys,

The beaglebone adventure is going quite well and I have almost working code with full PPI support at steprates upto 60kHz

Today I did another test, since the Y axis is completely ready on my machine and so I decided to do a speed test.

The speed on video is 500mm/s . I was able to go upto 700mm/s before you clearly hear that the stepper would like to divorce from me :)

Anyway, I cannot imagine this axis ever to run at speeds above 200 or 300 mm/s , so the results are superb.

add commentadd comment in the forum

reader comment Comment from: evokanivo on Friday, June 15th 2012 - 8:41 PM
Would you mind explaining what portion of the required processing the board is doing? I can't tell what software it's running or what it interfaces with. Are you building off the standard starting images?

I'm currently doing my own pulse generation and would like to get over 100kHz on bare metal, which is above the limits of my current dev board. So thanks for letting me know about the Beagleboard - it's cheaper and more powerful.

Friday, June 15th 2012 - 9:49 PM

Actually I don't use beagleboard but beagleBONE. Ist of the same family.

The big problem with these kinds of boards is that the linux os isn't realtime, so you cannot get very accurate timings.

Fortunately the beaglebone also has some kind of mini CPU called PRUSS which runs on itself at 200mHz and can accept commands from the c code running on the beaglebone.


I cannot give a lot more information currently as I have to give credit to somebody else for the low level stuff.
A person from Holland is building a cape for beaglebone + the needed PRUSS driver.

Once that is ready this person is going to make the cape available commercially, but we are nog in betatest.


I just bought a beta board and am going to assist in the higher level code to run lasercutter. The lower level high frequency code is still his stuff.


So.. in future I will be able to tell more.


By the way, what you see on the movie is a testrun with EMC2 under linux, not yet the bone since EMC2 has such a nice testing routine that I didn't care to create it on the beaglebone just to be able to discover possible speeds and accelerations :)

add commentadd comment in the forum

Wednesday, June 20th 2012 - 9:10 PM

While the Y axis was running smooth, it was time to start working on the X axis.

See how smooth this one will run

add commentadd comment in the forum

Monday, July 9th 2012 - 9:35 PM

Time for a few updates.

The machine now works in all 3 axes.

Some videos I made since the working of the Y axis are the working of up-down table , the X axis + the alignment of mirrors while testing with a red laser pointer







I don't like the sound in the last movie. Its because my energy guiding chain makes this noise and I need it to guide the air tubing for the air assist. Maybe I should try to find much smaller energy chain for this...

add commentadd comment in the forum

reader comment Comment from: Techgraphix on Monday, July 9th 2012 - 9:49 PM
educa wrote:I don't like the sound in the last movie. ...

Maybe it's gone when you put the airhose in it.. Maybe also a wire for the Z-autofocus-sensor and one for a Pilot-laser-light? In that case you shouldn't go to small..
But: it's not a musical instrument... some mechanical noises are considered normal..

Kees

Monday, July 9th 2012 - 9:56 PM

Airhose is allready in it. Pilot laser is going to come through the mirrors as a drop in replacement when lasertube is off

The only thing which might interest me is autofocus, but I still don't know how on earth I would do that. Ideas ??? I though I was going to do that with a focus gauge, but autofocus is a lot nicer of course

add commentadd comment in the forum

reader comment Comment from: Enraged on Monday, July 9th 2012 - 10:07 PM
can the autofocus be a subroutine you can run? if so, you could just use a microswitch on the laser head, and the have table rise up until it touches, and then back off a set distance to the focal length.

Tuesday, July 10th 2012 - 11:19 AM

I can run whatever I want. the controller is running at 720Mhz and the dedicated timing mechanism is running at 200mHz, so plenty of space

add commentadd comment in the forum

Sunday, November 4th 2012 - 7:11 PM

Today is a very important day for my machine.

I managed to let the machine cut something.

Currently cutting is done with linuxcnc. The wood is 4mm thick poplar triplex and I cut at 15mm/s with 100% power on a 80W reci tube and around 400PPI

The cut is ok, but I might be able to reduce charring some more. That will need some experimenting and for that I'll wait until my air assist compressor arrives and my fume extraction (630CFM) arrives.

The demo object is 70x60 mm



I think I'll need to reduce cutting speed and up my PPI settings, but for that any hints/tips are very welcome.

Earlier today I also cut 10x10mm rectangles out of this 4mm thick triplex and the final cut-out rectangle was 9.8x9.8mm, so I really get a very nice 0.2mm (0.0078") kerf.

The picture below is by the way after coloring in by my son

Image

add commentadd comment in the forum

Sunday, November 4th 2012 - 9:21 PM

I start to love Dirktheeng more and more.

Have a look what a difference I got (left compared to right) while cutting 4mm poplar with different PPI settings.

Image

add commentadd comment in the forum

Sunday, November 4th 2012 - 9:39 PM

And a last one for today. Dinosaur is about 2x2 inches big

Image

add commentadd comment in the forum

reader comment Comment from: jv4779 on Monday, November 5th 2012 - 3:10 AM
It looks like you are cutting and marking at the same feed rate but different laser powers? If this is the case you can use the same power for both and just go faster on the engraving. For example I cut 1/8" acrylic at 7 ipm and 40% duty cycle, but I do marking at 60ipm and the same 40% duty cycle. This an example where duty cycle is much easier to control than PPI, just go faster to cut less deep.

Also the rapid movements look very slow. I would expect a laser cutter to be 600-1000 ipm on rapids.

Jeremy

Monday, November 5th 2012 - 7:08 AM

Jeremy,

At this moment I haven't upped my rapids for 1 very simple reason. The machine can rapid on Y without problems at +- 1400 IPM and on X it can even go 2350 IPM.

However. Even if the machine is constructed of 35x35mm (1.4"x1.4") aluminium extrusions. When my Y axis moves fast (my acceleration is quite high on this one, at 236inches/sec²) then the whole machine is wobbling.

This is mostly because of 2 reasons:

1) The machine is still in skeleton state. When al electronics , air assist motor and fume extraction are built-in, then there will be a 2mm thick steel plating around the machine which will make the aluminium extrusions a lot more rigid and allow me to go rapids.

2) Because the machine is quite big (machine itself is 63inches long, 47 inches deep and 23.5 inches high) fast movements create a huge amount or inertia. The machine is mounted onto 2 tables and these are only standing on 4 feet.
The result is that this table is not that stable and as soon as the machine does rapids, the table begins to wobble too. (I even think that the table wobbles a lot more then the aluminium extrusions, but that is something to find out.) Once I'm ready, I'll mount plates under the table edge so that the table surface will be in full contact with the ground and not only by 4 feet. (40mm MDF plate mounted between table top and ground), you should be able to just press it underneath and it will sit there, blocking the table from wobbling.




So thats the main reason why I don't do rapids now.


The reason why you see burned edges is because I was still experimenting there with PPI settings and the cut you saw is at 100% power, 400PPI but at 35IPM. I reduced that speed to 16IPM and now the result is perfect.

The only problem I still have is rastering. The movie yousaw above is using EMC2 (not 2.5.x) with Ben Jackson config. I can get raster engraving working a little bit there, but not completely and I'm certainly not able to engrave for example 4 pictures in 1 laserjob. The laser engraves the first one, but with the next 3 it does all sweeps with X and Y axis, but nothing is burned onto wood.


So, next step is to format a machine with latest live cd and then try out your way (your config) for raster engraving.

Like I understood it now, with your system everything is in 1 single gcode file. (At least its possible).
So for example if I want to engrave a 5"x5" picture (raster engrave) and then cut it out, I could combine the engraving and cutting in 1 single .ngc file and run on linuxcnc with your config ?

That would certainly rock a lot. I don't really care if it would make the gcode file very big, thats my computers problem, not mine :)


So.... that'll be my next step to see how I can solve raster engraving to get consistent results without having to worry if the laser is going to shoot, or only goint to sweep but nothing shot :)


Kind regards,

Bart

add commentadd comment in the forum

Sunday, November 18th 2012 - 10:29 PM

Today I finally finished installing an intel atom board (D525MW) with linuxcnc and jeremy (jv4779) 's linuxcnc configs.

The first engraving was a success, but the quality is still not what it should be (which will require some trial and error)

Image

Now I must admit that until now, I haven't yet searched for the exact focal point of my laser. I always place a 1/2" piece between my sheet and laser head to measure that height, but I really don't know if 1/2" is the correct distance. Dure it looks nice, but if its not 100% ok then I'll never find the best possible quality.



So... I'll have to find out a way to find optimal focussing. Ideas are welcome !

add commentadd comment in the forum

reader comment Comment from: jv4779 on Sunday, November 18th 2012 - 11:02 PM
If you have a motorized bed you can just move it down as you sweep, then find the smallest spot size and measure where it is in your sweep, then calculate what the W value is, move to that W with G0 W? and now the bed is at the correct height. To keep this distance measure from the laser carriage or nozzle to the top of the work and that is your best focus distance. In this gcode it moves the bed down .5 inches over 10 inches. So if the best focus is at 2 inches into the line, your best W is 0.1. G0 W0.1 and measure.

Be sure to touch off your W axis before running this. Touch off at a point above good focus.

%
M4 S100
M68 E0 Q20

G20
G61
G0 X0 Y0 W0

M62 P0
G1 X10 W-0.5 F600
M63 P0

G0 X0
%

Based on your raster picture it looks like you have some slop in your Y axis belt. The first raster line is wavy. Also looks like you can go higher DPI as the black areas are not getting overlapping dots.

Jeremy

Sunday, November 18th 2012 - 11:11 PM

I don't know if its SLOP in the Y. I rather think that it is my machine wiggling because of the fast G0 move to start. I'll try to replace the initial G0 move with a much slower G1 move.

The reason I get this is because the whole laser X axis is quite heavy and if I move too fast the whole machine wiggles a little.
The motorized bed then has time to wiggle a little too, so thats probably the problem. Also the machine doens't have side plates mounted yet so I guess that will improve stability too.


Thanks for the hint about finding focal. I'll test it out since I have motorized Z table. I believe that in your config that is then the W axis ? what axis number would that be for the parameters in the ini file ?


Kind regards,

Bart

add commentadd comment in the forum

reader comment Comment from: StigOE on Monday, November 19th 2012 - 1:01 PM
Another possibility to find focus is to use a slanted piece of wood on the bed and cut a line. Find out where the thinnest line is and measure the distance from the area up to the head or the plate the head is mounted on.
reader comment Comment from: jv4779 on Monday, November 19th 2012 - 3:17 PM
educa wrote:Thanks for the hint about finding focal. I'll test it out since I have motorized Z table. I believe that in your config that is then the W axis ? what axis number would that be for the parameters in the ini file ?


W axis is AXIS_8. In Ben's 2x_laser config he had the table on U axis, but in conventional CNC machines U is parallel to X, V = Y, W = Z. So I changed it to W. A side benefit of W is that when doing a top down render of the tool path, W doesn't effect the X/Y positions.

Jeremy

Thursday, December 13th 2012 - 5:05 PM

I am having some unforeseen problems with my machine.

I tried to cut a 1" circle using the different microstepping possibilities of my stepper motors.

The motors are 1.8° nema23 steppers. The biggest problem is that 1 full step on my stepper is 0.5mm of travel.

So I tried 25x microstepping and thought that I might get 0.02mm step resolution.

The result is something completely different. Instead of 0.02mm step resolution I get the results you see on the picture (taken with different microstep settings on the motors)

Image

As you can see, in 1/4 step mode you clearly see all the steps of this motor and the cut is UGLY. 1/8 isn't better and from 1/16 and 1/32 I start to get better results.

its getting BETTER, but not yet good :(

The reason is probably because in microstepping mode steppers always tend to want to go into fullstep position. So if you are inbetween steps, you get actually not the wanted resolution but something inacurate.



I was looking at planetary gearboxes, but it seems like these would not be a good solution for me (I wonder why).

I was also considering using 0.9° motors and a 10:1 reduction between motor and axis by using 2 steel spur gears with 12 and 120 teeth.

And my last possibility is of course the most drastic one .... ballscrews.


Any ideas / hints / tips ?

Kind regards,

Bart

add commentadd comment in the forum

reader comment Comment from: BenJackson on Thursday, December 13th 2012 - 6:18 PM
Sorry I'm not getting this out of the history, but what's the controller? A microstepping controller should easily be able to hold a stepper motor between "detents" as long as it's enabled. BUT the microstepping positions (other than about half) aren't really linear and you can't rely on them for accurate positioning (just smoother positioning).

I'd also say that given the nice clean cut you're getting in plywood that you're using pulsed cutting and the edges look more like the artifacts of a really low PPI than stepper positioning errors.

Thursday, December 13th 2012 - 6:47 PM

the cuts you see are 400PPI, I also did 800 and 1200 PPI and didn't change a lot.

I must admit that the cut edge is indeed very nice (absolutely no burn) and for a lot of purposes the cut quality is more then enough. Its just that for things like acrylic this side finish is ugly. For plywood I guess its more then ok.


1 relatively cheap thing I can do is that I can replace my 1.8° steppers with 0.9° steppers. That would certainly not make the quality worse and if its really a problem of microsteps pulling towards full steps, then this would even be a good step.


There exists also stepper motors with 0.45° / step resolution. These are manufactured by Lin Engineering (1 company in the whole world I know of) and are quite expensive. Around $300 for 1 motor with around 1.8Nm torque

add commentadd comment in the forum

Thursday, December 13th 2012 - 7:12 PM

This http://en.nanotec.com/steppermotor_st5909.html looks like a nice stepper motor to try out.

add commentadd comment in the forum

reader comment Comment from: kbob on Thursday, December 13th 2012 - 7:42 PM
educa wrote:The motors are 1.8° nema23 steppers. The biggest problem is that 1 full step on my stepper is 0.5mm of travel.

Bart



Can you reduce the diameter of the motors' output pulley/gear? (I tried to find a photo of the drive in your buildlog, but if it's there, I missed it. Are you using a belt and pulleys?)

For comparison, the pulleys in the 2.x design are 20 teeth × 0.08 inch/tooth = 1.6 inch/rev (40 mm/rev). A 1.8° motor gets 0.08 inches of travel (0.20 mm) per full step.

Edit: I downloaded the Sketchup CAD file. It looks like you are in fact using a belt and pulleys.

Thursday, December 13th 2012 - 8:22 PM

I use steppers nema23 with 20teeth T5 belt (so 5mm per tooth) that makes 0.5mm travel per step on the stepper.

If I would replace the 1.8° stepper with 0.9° steppers then I will effectively double my resolution.

I was looking for 0.45° steppers too but they seem to be extremely expensive (although I might still consider them because as you can see my machine is quite an achievement for me and I want to make it perfect now).


I was first looking for planetary gear reduction, but thats not a good solution (they told me) because of backlash and the fact that I wouldn't reach acceptable raster engraving speeds.

On the other hand, the quality is also not THAT bad. Its just a search to get it better:)


yesterday I put my 100mm lens instead of the 50.8mm (2") lens and tried again and suddenly a lot of mu problems were gone. I suspect that this is because the 100mm lens makes wider cut and the cuts overlap better making the error vanish more.

I don't think higher PPI will solve this, because I tested it upto 1200PPI. I must confess I did not do ANY test without ppi. Maybe thats a fault of me.

add commentadd comment in the forum

reader comment Comment from: cvoinescu on Thursday, December 13th 2012 - 10:41 PM
FWIW, I agree with educa and disagree with BenJackson: they really look like stepping artifacts rather than PPI artifacts.

Some controllers are better than others at microstepping. Surprisingly, the $12 Pololu A4988 boards are quite good. One important thing to consider, with any controller, is that its operating point must be well into the current-limited region for microstepping to work. So if your power supply is 12 V and your motors are rated 8 V, 1.5 A, that won't work very well. Ideally, the voltage of the power supply will be about an order of magnitude above the rated voltage of your steppers. If your supply is 48 V and your motors are rated 4 V, 4 A, that's great.

Low-impedance motors (those with low voltage ratings but high current ratings) tend to be better at microstepping than higher-impedance motors, even after adjusting for power supply voltage.

Drivers are optimized to run close to their rated current. If you're driving 1.5 A motors from a 8 A driver turned way down, the driver may not have very good control over the current, and the microstepping may suffer.

Another important thing is that the torque of the motor is very low when its position deviates only a fraction of a full step from the commanded position. If your carriage doesn't move extremely easily, the motor will stay still as you microstep away, then it'll jump a number of microsteps at a time.

Finally, the detent toruqe interferes with microstepping, to the point that at very low current the motor may move full steps at a time anyway. As with the drivers, the motors microstep better when driven with currents close to their rated current, and less well when the current is much lower than that.

So, short of replacing drivers and motors, your best shot is to increase the driver current, if possible.

If replacing motors, definitely get 0.9 degree steppers, and do not oversize them too much. Both motors and drivers run better (as far as microstepping is concerned) when driven close to their rated current (let's say at least half the rated current for the motor, as a rule of thumb; for the driver, it depends, some regulate low currents well, some don't), with a power supply many times higher voltage than the motor rating.

Thursday, December 13th 2012 - 10:49 PM

The motors I currently use are http://www.cnc4you.co.uk/index.php?rout ... duct_id=65 and I drive them bipolar parallel at 4.2Amp with a 36V PSU

The stepper drivers are http://cnc4you.co.uk/resources/CW5045.pdf


If you look at my cut quality then the errors I see are really quite small. I wonder if I should replace my 1.8° motor with 0.9° or 0.45° motors.

Both 0.9° and 0.45° motors are nema23, but the 0.9° is around 70 euro and the 0.45° is around 250 . So a little big difference :)

add commentadd comment in the forum

reader comment Comment from: cvoinescu on Thursday, December 13th 2012 - 11:04 PM
That's a big motor there! Specs look good to me... Maybe the driver isn't very good -- I have no experience with it at all, but I remember I read recently about someone with a Chinese driver with a Toshiba chip who got bad microstepping performance and then switched to the much smaller A4988 and it fixed things for him. But he was using NEMA17 motors and a 24 V PSU, so much smaller. Could you borrow a different driver from someone else?

Thursday, December 13th 2012 - 11:13 PM

a4988 is only 2 amps. I don't think my motor will do a lot with that unfortunately.

I do understand that my driver might not be the best on the market, but thats difficult to test isn't it :)

Especially when you see that most drivers aren't cheap.

Thats why I hope that maybe just changing to 0.9 or 0.45° motor would also help me a lot.

add commentadd comment in the forum

reader comment Comment from: fantasy2 on Saturday, January 5th 2013 - 11:25 AM
Hello!

Where did you buy that timing belt for the Z-axis? I cannot find that length anywhere.

Saturday, January 5th 2013 - 11:33 AM

http://www.beltingonline.com/

add commentadd comment in the forum

Monday, January 21st 2013 - 12:16 PM

I'm thinking of exchanging my 3.1Nm nema23 motors for other ones.

I currentlyhave a problem where I think i see the stepper motor step size in my cuts. Its not superbig and accuracy is good, but.... If I see it I'm not happy

My current motors are 1.8°/step and I will put in 0.9°/step motor instead. That should increasemy precision.

I'll use nanotec motors

http://en.nanotec.com/steppermotor_st5909.html


If that is still not enough the I can double my resolution again by usin T2.5 belt and smaller pulleys instead of T5 belt.

Its a lot of work but has to be done...I want absolutely perfect quality in the cuts.

add commentadd comment in the forum

Monday, December 30th 2013 - 2:36 PM

About 1 year since I did an update on my buildlog.

1 very important change has been made to the machine. I replaced my 1.8°/step motors with 0.45°/step motors from Lin Engineering. The result is a lot better then before.

In the last year I have done a lot of comercial work making personalised gifts and selling on my site www.lasermania.be

In 2014 I will do it a lot calmer. From time to time sell something, but not in mass anymore.This is a hobby and if I wanted I could sell $1000 to $1500 a week leaving me no more time at all to do anything else.

add commentadd comment in the forum

Sunday, March 2nd 2014 - 2:08 PM

A little follow-up now.

The lasercutter is completely ok as a project and everything works as intended.

Time for a new challenge.


So.... this time I will make my own 3D printed based on the same principles I built my lasercutter on.

High quality above cost. So, I'm not going to build the cheapest 3dprinter but I'll try to have as good as possible quality by using inductry grade components like THK rails again + an all-metal construction of high precision materials.

The buildlog will be hosted at www.bigfoot3dprinter.com , which is my website for this build.

add commentadd comment in the forum

Tuesday, April 8th 2014 - 6:15 PM

Hmm, I forgot I had a movie showing something very nice.

Engraving with laser. A kind of engraving I did a while ago before I had a co2 laser engraver. It is using a laser diode.

The result was absolutely marvelous, so here is the link to my machine/results

https://www.youtube.com/watch?v=xfhDgUTKdb4

add commentadd comment in the forum

Thursday, April 17th 2014 - 8:38 PM

Today I tested if I could do anything related to laser sintering.

First demo here: http://youtu.be/VN0lS5CSkmk

add commentadd comment in the forum

Sunday, October 16th 2016 - 7:21 PM

So today I opened my new website.

www.laserbeaver.com

On that site I will be documenting the complete build process of a 40W laser cutter from zero to hero, including the complete hardware, but also the programmation of custom firmware and host software. It will be an incredible adventure and I inspire some people.

The site contains a complete development blog + a knowledge base for laser cutter users .

add commentadd comment in the forum

reader comment Comment from: Halfdead on Tuesday, October 18th 2016 - 1:11 AM
educa wrote:So today I opened my new website.

http://www.laserbeaver.com

On that site I will be documenting the complete build process of a 40W laser cutter from zero to hero, including the complete hardware, but also the programmation of custom firmware and host software. It will be an incredible adventure and I inspire some people.

The site contains a complete development blog + a knowledge base for laser cutter users .



I look forward to more updates.
About Us | Contact | 2010 BuildLog.Net