Commercially available laser controllers

Electronics related to CNC

Re: Commercially available laser controllers

Postby lasersafe1 » Tue Sep 14, 2010 4:10 pm

The Jpeg file on page 8 of this thread is "the" file that was used for my test and should be used by everyone if we want to compare.
First click on it once to bring up the full size image, then right click on it and do the "save Image as" to your computer and it will download the 2.5" square image at 600dpi. You will know you have the right file if it is 1.7 MB. If you get it at 79KB, then you have only saved the thumbnail version.

The thing about the FSE that seemed most amazing was his comment of doing it in 10 minutes. I had to slow down my raster significantly to 200mm/sec and do a very fine stepover in Y of .005mm per scan. This took me 2 hours to run. If I try to do it with larger stepover, it gets muddy.
lasersafe1
 
Posts: 599
Joined: Fri Nov 27, 2009 8:23 pm

Re: Commercially available laser controllers

Postby lasersafe1 » Tue Sep 14, 2010 4:38 pm

I just found a picture of the FSE Mayan calendar on his website with a U.S. Quarter in the picture. A quarter is 31/32", so my measurements of the engraving put it at 2.2947" across. The test file I have is 2.424" on the actual engraving. Thus it seems pretty clear that he is getting some damn tight engraving at a really fast speed.
Attachments
withquarter.jpg
lasersafe1
 
Posts: 599
Joined: Fri Nov 27, 2009 8:23 pm

Re: Commercially available laser controllers

Postby dirktheeng » Tue Sep 14, 2010 9:30 pm

Tweakie wrote:Sounds like a challenge for dirktheeng. Hows your C++ Dirk ?

The Mach plugin is open source and can be downloaded from the Artsoft website complete with the source code http://www.machsupport.com/plugins.php
Some work has already been done, to correct for the aspect ratio problem, but there is lots still to do.

LS1 - I will make the next one 63mm for a better comparison with your work but it is really the FSE I would like to compare with. What is the diameter of theirs I wonder ?.

Tweakie.


I do ok with most any programming... had to build my own experimental systems for my PhD at Penn State... wrote the control algorythms for reactors, high accuracy torque cells ect... The biggest problem for me right now is that I don't have anything to mess with. I am still engineering my laser system and really tring to keep costs down. I can never spend enough time doing fun stuff like this. I have a wife who loves me dearly that I am choosing to be with in preference to this project many times. This is fun, but my marraige is forever. It'll get done eventually. We also decided to save for a house, so I have to budget my fun a little too. Until then, I can contribute to the forum and share my drawings. I may start a build log soon.
dirktheeng
 
Posts: 616
Joined: Thu Sep 09, 2010 4:49 pm

Re: Commercially available laser controllers

Postby lasersafe1 » Wed Sep 15, 2010 1:13 am

What ??!!!??? A wife that wants to spend time with you!??!? Now that it is college football season, my wife is glued to the set and I get to play with the laser. She's a football fanatic. I still don't know why she married an engineering geek.
lasersafe1
 
Posts: 599
Joined: Fri Nov 27, 2009 8:23 pm

Re: Commercially available laser controllers

Postby Tweakie » Sat Sep 18, 2010 1:31 pm

LS1,

FSE have now confirmed that their Mayan is, as you calculated, 2.3" diameter. I have asked them and it will be interesting to see if they are willing to show a higher resolution image but I doubt that they will. I get the feeling that they can always do better than everybody else but are not willing to prove it. Still we will see.

Tweakie.
Tweakie
 
Posts: 120
Joined: Tue Apr 27, 2010 5:17 pm
Location: UK.

Re: Commercially available laser controllers

Postby fullspeceng » Sat Sep 18, 2010 7:21 pm

Tweakie:

We are a transparent company with nothing to hide. Our image is large enough for you to see the lines are clearly cleaner to the naked eye.

Under 40x magnification they are still sharp.

We are happy to do apples to apples testing against your system.

Here are the factors that must be the same:
1) You must use the same material. We ran this in oak and you ran it on some other material. Suggest something that is easily accessible and we will run it the same.
2) You must run the same image at the exact same size. Our original image had a black border and our logo on the bottom. It is exactly 2.315" x 2.668". Since you changed our image we cannot compare it.
3) You must state 1) how long it takes to setup the image from the time you save the image to disk to the time it starts engraving 2) how long it takes you to engrave the image

Complete these tasks and we will repeat with the exact same parameters.

We are 100% sure our controller is bit perfect. Not only can you see the difference at 1000dpi but we are happy to show the signals captured on a highspeed 1ghz oscilloscope which shows less than 10ns variation between the pulse width on all axes.

Why do people insist on downsampling the image then enlarging it to a different size and using a different material?

It is frustrating for us to suggest a simple apples to apples comparison and people insist on deviating from the test plan so the two results are meaningless.

Scientific testing requires scientific methods.
fullspeceng
 
Posts: 38
Joined: Thu Dec 17, 2009 10:13 pm

Re: Commercially available laser controllers

Postby Tweakie » Sun Sep 19, 2010 12:14 pm

We are a transparent company with nothing to hide. Our image is large enough for you to see the lines are clearly cleaner to the naked eye.
Under 40x magnification they are still sharp.


Fullspec,

I am not doubting that your system is excellent, in fact I am awaiting the results Bart's testing before placing an order for one of you boards and software.
What I am doing, is looking at other systems and software producing the same results (as good) as you have achieved. Perhaps if you would be prepared to post a higher resolution image (40x magnification would be nice) so that we could all see just how sharp your image is then a better comparison can be made.

Essentially what you are saying in the above quote would be just like me saying that the the lines in this image are clearly cleaner than yours (which, of course, they are not but you cant see that from the picture).

Please post a higher resolution image.

Tweakie.
Attachments
DSC00503.jpg
Tweakie
 
Posts: 120
Joined: Tue Apr 27, 2010 5:17 pm
Location: UK.

Re: Commercially available laser controllers

Postby lasersafe1 » Mon Sep 20, 2010 1:01 am

Okay, so I just cut some gears with my laser and was inspecting them with my electron microscope, but there was a dust mite in the frame.
10_matchar_2.jpg
10_matchar_2.jpg (5.75 KiB) Viewed 19817 times


Seriously... I did the best I could with the calendar file. The reason I cut off the edges and FSE company logo was the fact that I didn't want to become an advertiser by default. I'm happy I could run the file with a nice high resolution, but it took far too long and I don't think I want to put my engraver through another 2 hour test. I don't think FSE is making any false claims. He clearly has the bit conversion control handled well. I'm looking forward to testing one.

Clearly there are going to be advantages for some controllers over others, but there will be limitations as well until the software can be upgraded for the perfect controller. One advantage I like with the LightObject controller is the ability to run files without even booting up the PC. I also like running files that have mixed raster and vector cutting within the same run. I am finding disadvantages though. I have had the software crash if I try to process a file that is too big. Apparently it also takes much longer to do a run than with the FSE controller.

Does everyone have a flatbed scanner? If we are interested in looking for missing pieces or jagged edges, then it only makes sense to scan them in for an image. The last scan I did was at 1200 dpi. The whole calendar would have exceeded the buildlog limit, so I cropped it to only show the region of interest. The wood I used was Maple because it has a super tight grain and very few dark grain lines across.
lasersafe1
 
Posts: 599
Joined: Fri Nov 27, 2009 8:23 pm

Re: Commercially available laser controllers

Postby Tweakie » Mon Sep 20, 2010 6:22 am

LS1,

Them dust mites - they ruin everything :lol:

I keep looking at that calendar picture of yours and try to make up my mind about the wavy line. It is obviously not a resolution issue, but it looks for all the world like it could be caused by non-linear motion from part of the drive train and I just keep thinking mechanical.
Have you tried any other (perhaps Simpler) single line designs and had the same result ?.

Tweakie.
Tweakie
 
Posts: 120
Joined: Tue Apr 27, 2010 5:17 pm
Location: UK.

Re: Commercially available laser controllers

Postby Tweakie » Mon Sep 20, 2010 11:43 am

This image has been cut to FSE's dimensions in Oak and is probably equal to their quality but as they refuse to put up a higher resolution image, despite my repeated requests, we will perhaps never really know.
I had not cut anything into Oak before and it was quite a learning experience. It requires about 30% less power than engraving laminate and is considerably less noisy (hard engraving laminate makes quite a racket when you start chewing into it with a laser) also, as others had suspected, the sharp corners char more readily than the flat surfaces and this has the effect of smoothing the contours of the image.
This example was cut with a vertical (Y Axis) stepover of 0.08mm and horizontal (Y Axis) feedrate of 4500mm per minute. The total time from loading the .BMP to completing the work was 19 minutes (not as fast as FSE but my X Axis also carries the Z Axis which usually holds a 3 Phase spindle router motor. This combined mass limits my acceleration / deceleration and maximum federate over such short distances – thus the extra time).
Although I can continue to reduce the vertical stepover I think that any possible increase in resolution is probably outweighed by the increase in job completion time.
What do you Guys think – is this as good as the FSE example or should I cut another with smaller stepover ?.

Tweakie.
Attachments
DSC00515a.jpg
Tweakie
 
Posts: 120
Joined: Tue Apr 27, 2010 5:17 pm
Location: UK.

PreviousNext

Return to Electronics

Who is online

Users browsing this forum: No registered users and 8 guests

cron