Retrofit an older Epilog with 3rd party controller? 3D engraving capability

[Andrew Stow]

I just got an old Legend 32 (30 watt) at government auction. Fairly cheaply. I then had to get a replacement (refurbished, not even new!) LCD/keyboard assembly. So as it stands, I've got about 3 grand in this machine.

But one thing I really would like to use it for is 3D engraving. Goal is to be able to do gun stock checkering. Talked to epilog support and they said the 6000 series machines aren't capable of it. 8000 series and up, only.

So what I'm wondering is, how feasible would it be to use a 3rd party controller for hardware pulse generation and something like Mach 3 or linuxcnc for user interface?

I'm talking about disconnecting all the motors, the laser, everything, and wiring them all up to a 3rd party controller and then user-defining every pin in mach 3 or whatever I end up using.

Has anyone done this? Is there even any third party controller software that will do something like Epilog's 3D raster mode? Or for that matter, a CAM processing software prior to sending it to the engraver? What hardware would be a good choice for what I'm trying to do?


Ya'll's thoughts would be most appreciated. Trying not to get stuck in a sunken cost situation here. So if the most reasonable suggestion is "sell what you've got and buy something else" don't hesitate to suggest that, either.

About the only thing I KNOW needs to be done to the machine at this point is at least one mirror replacement and a lens replacement. Worst case scenario would be that my tube is on it's last legs, but I hear photovac does good work at a reasonable price on synrad modules. So I think I might be close to break even on resale in a worst-case scenario.

Thanks in advance

[Dan Hintz]

I suggest running your machine at max power and try to engrave a line of expected depth into your choice of wood (choose the hardest wood you plan on using for stocks)... you may find the speed at which it runs is too slow for your taste, especially after you've thrown a bunch of money into the machine.

[Mike Null]

Take a look at the Epilog web site. They have some good examples of what you can do with gunstock engraving and checkering.

[Andrew Stow]

Thanks, will do.

Do you believe I'm expecting too much from a 30 watt rated machine?

I fully acknowledge the possibility. However, one thing I'm wondering is... what is it that makes "3D rastering" in the newer machines different from "map color to power/speed" in the older ones? I tried asking the Epilog guy that, but I didn't really understand his answer. He said the 3D engraving mode changed the output frequency according to the greyscale of the image (paraphrasing from memory). I asked him if that meant it changed the on/off frequency of the laser and he replied that no, it was still a continuous wave.

Did he mean that the actual wavelength of the beam was modified?

Sorry for the possibly dumb questions.

[Andrew Stow]

Take a look at the Epilog web site. They have some good examples of what you can do with gunstock engraving and checkering.

Right, they are impressive and make me hopeful that what I want to do is possible, but those are with the 3D rastering option. Not included on the 6000 series legend 32. My machine does have an option for "map power/speed to color" but according to the Epilog tech support guy I talked to, they're different. I don't understand how, tho.

[John Pletcher]

In a nutshell.....in standard raster mode the tube pulses, where as in 3D mode the tube doesn't pulse, but stays on and the power level varies according to the gray scale. The darker the color the higher the power resulting in a deeper burn, the lighter the color the lower the power the less burn. The result is a 3D engraving

[Andrew Stow]

Ok, so basically:

The tube power is varied by pulse width modulation of DC to the tube, but the actual emission of light, whether in pulses at a defined rate or in a continuous beam, is independent of that?

Is that a fair assessment of what's going on?

[Dan Hintz]

Do you believe I'm expecting too much from a 30 watt rated machine?
Not necessarily, but I think you need to be aware of a machine's limitations, particularly if you plan on throwing more money at it. You may decide it's like putting a Ferrari's engine in a Yugo... sure she's got lots of power, but the first corner you turn at speed and the wheels fall off. Seeing what the machine can and can't do at its limits will give you an idea of if it's worth fixing up other areas.


what is it that makes "3D rastering" in the newer machines different from "map color to power/speed" in the older ones?
Generally, mapping color to speed was limited to a small number of colors (say, 8), and the colors themselves are generally set (red, green, blue, and straight mixtures thereof). Each color could be a specific power/speed. "3D" modifies the laser's power between two user-defined power levels (say, 0-60P) in relation to the grayscale level of an image... black gets 60P, white gets 0P, middle shades of gray get 30P, etc. You could accomplish the same thing by selecting each colors, but it would be an extremely painful process... the 3D setting simplifies everything significantly.

[Michael Hunter]

On the Legend 32 "map colour to power/speed" is used for vectors only.
I use this to draw lines on switch panels before cutting the holes and outline.
(My setup is green to draw and black to cut, but you can choose any combination you like).

Before completely rebuilding the machine - have you actually tried raster engraving a greyscale image?
Although not quite the same as the "3D" engraving of newer machines, it is possible to get some pretty good results.

[Joe Hillmann]

I am not quite sure what you are trying to do but is sounds like playing with shading may get what you want. Take a look at this video and then do some playing with your laser. You may not have to change anything on it.

http://www.youtube.com/watch?v=zpl7o7ulWAA

[Andrew Stow]

Generally, mapping color to speed was limited to a small number of colors (say, 8), and the colors themselves are generally set (red, green, blue, and straight mixtures thereof). Each color could be a specific power/speed. "3D" modifies the laser's power between two user-defined power levels (say, 0-60P) in relation to the grayscale level of an image... black gets 60P, white gets 0P, middle shades of gray get 30P, etc. You could accomplish the same thing by selecting each colors, but it would be an extremely painful process... the 3D setting simplifies everything significantly.

Great info, thanks.

It sounds from your description that the 3D setting/automatic power range via greyscale theoretically might be added to the pre-8000 series machines with a driver and/or firmware update. If so, it would be interesting to see some smart fellow come up with a software hack to do this. This person not being me, since I know practically nothing about programming.

For my purposes tho, I will give the manually setting colors thing a try. It shows only pure R,G, and B options by default, but it appears that I can add more color definitions as well.

For simple checkering, presuming a unique power value per .001" of depth cut, I would only have to define 50 or 60 discreet colors in a range to have a smooth, adequate depth changes for my purpose, I think.

It would be impractical for 3D engraving things like photos, but might be doable for checkering patterns.

Michael Hunter: I'm a complete newcomer to laser engraving. I've built CNC routers and done a little work with them, but all this making wood disappear with light stuff is pretty new to me!

Still waiting on a couple of things to get started, so right now I'm just trying to get familiar with some of the issues surrounding my goals.

Thanks for all the great suggestions, guys.

[Dave Sheldrake]

Did he mean that the actual wavelength of the beam was modified?

The term frequency is being used to cover different meanings, Frequency & Wavelength are directly related in terms of physics of a laser but not the in same way when discussing PRF / PWM in a different context.

The wavelength is fixed by the gain medium and internal mirrors (in our case CO2 gas) and while other wavelengths (outside of 10,600) are emitted they aren't *used* as such.

cheers

Dave

[Dan Hintz]

It sounds from your description that the 3D setting/automatic power range via greyscale theoretically might be added to the pre-8000 series machines with a driver and/or firmware update. If so, it would be interesting to see some smart fellow come up with a software hack to do this. This person not being me, since I know practically nothing about programming.
It's unlikely someone will write a driver for an old machine, so don't get your hopes up. I will likely end up writing one for whatever fiber laser I end up with as I know I will require functionality not offered by known machines... but it will not be a pretty process.


For simple checkering, presuming a unique power value per .001" of depth cut, I would only have to define 50 or 60 discreet colors in a range to have a smooth, adequate depth changes for my purpose, I think.
Most substrates are not linear in nature when it comes to power versus engraving depth. On top of that, substrates like wood are variable in nature from one pinpoint to the next (latewood/earlywood, i.e., rings, are one perfect example), so even with the same setting you will get variable depth for the same power. So, trying to determine 50-60 different settings in an attempt to create a checker pattern will be an exercise in futility as it will never pass from one piece of wood to the next (or even from one square inch to the next). It's not an easy problem, so I would think long and hard about sinking money into this problem.

[Chuck Stone]

If your image is grayscale (not converted to 1 bit bitmap) try the color mapping
from black to white. Set the black at full power, white at no power, and fill in as
many other colors as it will allow.. at various power levels. (if it lets you choose the
colors)
While not true 3D, if you can change those settings it should certainly be enough
for a checkering pattern.

EDIT: When you fill in those other colors, make them shades of gray.
D'OH! I forgot that part..

[Andrew Stow]

Most substrates are not linear in nature when it comes to power versus engraving depth. On top of that, substrates like wood are variable in nature from one pinpoint to the next (latewood/earlywood, i.e., rings, are one perfect example), so even with the same setting you will get variable depth for the same power. So, trying to determine 50-60 different settings in an attempt to create a checker pattern will be an exercise in futility as it will never pass from one piece of wood to the next (or even from one square inch to the next). It's not an easy problem, so I would think long and hard about sinking money into this problem.

It sounds like those same issues would be a factor with the "3D engraving" setting on a newer machine, as well.

[Dan Hintz]

It sounds like those same issues would be a factor with the "3D engraving" setting on a newer machine, as well.

Some of the issues are, for sure. If you attempt to create a specific depth and tie it to a color, how do you test it? What part of the wood do you test on, because it will be high in one spot and low in another? And the moment you need to adjust the min/max power, you need to adjust all of the colors... a royal pain. Again, it's possible, but extremely painful.

[Mike Null]

If you take a closer look at the art on the Epilog site you will see that it is 3d art which can be engraved with most lasers.

There are ways to draw you own 3d art for laser engraving using CorelDraw. Much cheaper and quite satisfactory when compared to the options you are considering.

[Andrew Stow]

If you take a closer look at the art on the Epilog site you will see that it is 3d art which can be engraved with most lasers.

There are ways to draw you own 3d art for laser engraving using CorelDraw. Much cheaper and quite satisfactory when compared to the options you are considering.


Where can I learn more about this, Mike?

I opened up Epilog's sample file in corel last night and played around with it a bit. There's a TON of discreet color values in that diamond checker pattern.

What I'm wondering now is, can I create an image that ONLY uses a certain, user-defined small set of colors in corel? If so, then it would greatly simplify doing the "faux 3D engraving" Dan is suggesting might be possible. It wouldn't be nearly as smooth, but it would be a starting point. From what I understand, in order to use an image like corel's sample with color mapping, I'd have to create a specific color mapping setting for each and every color value used, right? Painful doesn't begin to describe it.

Dan: My hope is that using somewhat uniform color/grain/species wood will mitigate the variable depth of cut issue. But for sure, I will take a piece of mesquite and use the machine to cut some lines at various powers, wide enough to get the depth gauge on calipers in them, and see if I can measure the depth of cut for a given power. Like you said, it won't transfer from piece to piece very well, but if you read Epilog's description of how they did their gun stock with the diamond pattern, they had to run it three times to get the final depth they wanted. I think the biggest issue to avoid is varying depths of cut across a single piece.

On wood for gun stocks, my aesthetic opinion is that engraving detail into intricately figured wood is a waste of interesting wood, anyways. Better to engrave details into straight, clear walnut than curly maple, in other words.

[Mike Null]

3d drawings are quite time consuming and not something I am skilled in or have an interest in. When I have the motivation I fool around with various of the post script fills and generally feel that good results are doable but not worth my time.

I would search the internet for laser engraved gun stocks.

Ask somebody to run an ordinary black and white photo on wood for you to see what is possible with virtually any laser.

[Rodne Gold]

3d engraving on a laser is such an imprecise variable thing , it's all but useless.

[Chuck Stone]

3d engraving on a laser is such an imprecise variable thing , it's all but useless.

..unless you find a market for it.

[Chuck Stone]

Dan: My hope is that using somewhat uniform color/grain/species wood will mitigate the variable depth of cut issue.

You might look into using a stabilized wood for this, too. Once the wood is
impregnated with a resin, the variations in depth should be minimized.

[Andrew Stow]

Here's a couple of tests I made this afternoon.

http://i.imgur.com/U27kZBL.jpg
http://i.imgur.com/5pcA7Tb.jpg
http://i.imgur.com/IlTYKAh.jpg

The dark wood is mesquite. The lighter is birch veneer plywood. The top part of one of those was ran twice.

That was at 50% power 50% speed.

Next I'll try that 3D checkering image file on a flat piece of wood as a raster and just see what it does, like ya'll are suggesting.

Thanks for all the help. For a 13 year old machine, I'm frankly a bit surprised the laser even fired.

[Andrew Stow]

3d engraving on a laser is such an imprecise variable thing , it's all but useless.

I hope it's not insurmountable.

I started out in this kind of thing with 1911 gun grips. Haven't made any in a few years, but I had more than a few people ask me if I could engrave checkering or designs on them. Always had to turn them away because my homebuilt CNC router was purely 3 axis and doing a rotary attachment was just a bridge too far for my available time. Got only so-so results using a fine V-bit on a curved surface and was quite limited in how far from the perpindicular point I could get with the vector.