My Cermark nightmare

[Kev Williams]

Ok, so not really a nightmare, but this is one of the reasons I bought the Triumph, in hopes that I could do panels like these in the thing. I do these in several sizes, this is the granddaddy, 24" x 30", 1/8" thick #4 Stainless...

DSC02847.jpg


I've done these for years in my LS900, but I have to rotate the plate 180 to do the bottom. This means splitting up the outer border line and re-aligning it for the second pass. The Triumph is big enough I can do two of these at once in the thing. Still teaching this laser to do what I want it to do, but I'm getting there! It's not perfect but pretty close. It's no worse than any panel I've produced in the LS900, and in many ways it's better...

One issue I'm having is heat warp. The 900 would cause some warping, but man, the Triumph really heat soaks the SS. When the logo was only half done, I found the side edges of the plate were nearly an inch off the table! Not very good for keeping the laser in focus. I used weights to hold the edges down and finished the logo, but had to re-run it this morning because of the "seam" in the logo where I re-started it.

I'm trying to figure out an easy way to temporarily clamp the edges of work down. My table is stationary, and I have it zeroed out nicely for flat and X-Y position, so I prefer not to move it if I don't have to... but I'm going to have to move it at least once I think: One idea I have is to get some very small toggle bolts, small enough to fit thru the honeycomb holes (not sure they exist), that I can use with some brass plates topside to act as hold-down clamps. To keep the toggle bolts from messing up the honeycomb from below, I thought about taping a few 1-1/2" or so diameter washers to the table from underneath where I need them that the toggle can go thru, then the washer could act a shim to protect the honeycomb. If/when I need to remove the toggle bolts, they'll just fall into the drawer below, I won't need to move the table to retrieve them.

Another idea is simply to use gravity --- as in, some 30" long x 1-1/2" wide by 3/4" tall lead bars should hold down a plate! I could have them made and paint them easy enough. I have a piece of solid SS that's 12" x 1" x 2" that weighs quite a bit, I use it regularly, but at 1x2" it's a bit in the way sometimes, smaller would be nice- so would more than just one!

I'm pretty pleased with this laser, it's already paid for itself. The learning curve between the Triumph and my other units is similar to learning to drive a 1962 Volkswagen vs. being chauffeured around in a Lincoln Limo-- but I'm getting there!

[Gary Hair]

One issue I'm having is heat warp. The 900 would cause some warping, but man, the Triumph really heat soaks the SS. When the logo was only half done, I found the side edges of the plate were nearly an inch off the table! Not very good for keeping the laser in focus. I used weights to hold the edges down and finished the logo, but had to re-run it this morning because of the "seam" in the logo where I re-started it.

If you are warping 1/8" stainless then you need to reduce power and/or run faster so you don't generate so much heat. I ran a job last summer on 12ga and the pieces were about the same as yours, 38" x 20", they were filled with text from top to bottom and I had absolutely no problem with warping, not even a fraction of an inch. If you have to use that much power/slow speed to get the Cermark to stick then you have something wrong.

Gary

[Mark Sipes]

What setting are you using for the new laser parameters. Sounds like you need to speed the laser travel up.

[Scott Shepherd]

If you are warping 1/8" stainless then you need to reduce power and/or run faster so you don't generate so much heat. I ran a job last summer on 12ga and the pieces were about the same as yours, 38" x 20", they were filled with text from top to bottom and I had absolutely no problem with warping, not even a fraction of an inch. If you have to use that much power/slow speed to get the Cermark to stick then you have something wrong.

Gary

I agree with Gary. There's no way you should be warping 1/8" stainless with that little of graphics around the piece.

[David Somers]

Kev,

Is your table steel so magnets will work on it? I realize your stainless substrate is likely too thick for a magnet to hold through it. But, you could be creative and glue a strong N52 Neodymium magnet to a material. Whatever you choose to glue it to, space it so that the material sits 1/8" high of the base of the magnet and is shaped so you have an overhang to it that extends beyond the diameter of the disc of the magnet. Then put the magnet on the table and let that overhang hold your substrate down.

A material I might suggest for this would be something called Sugru. It is a very workable silicon putty with an open time of 30 minutes or so and a cure time of about 24 hours. It sticks to almost anything tenaciously. You could shape it as you wanted right on the magnet so you have something to grip the magnet with easily, and also to create the overhang that will keep your substrate down securely. You can find Sugru at Sugru.com. In the US a good source for N52 Neodymium magnets is KJmagnetics.com. There should be another thread here where Dave Sheldrake gave a good source in the UK for the same types of magnets as well if you happen to be in the UK.

For what it is worth, I always have a dozen or so packs of Sugru on hand in the fridge for things. Although the 24 hour cure time is annoying, it is so very useful that I don't mind that. I also use N52 magnets of various sizes all the time and keep a stock of various sizes down to 1/16 x 1/16in on hand.

One word of caution. If you do keep magnets like this around, even the lesser ones, please be really careful with them if you have pets or young kids around. Two or more of those suckers swallowed can be life threatening, either in the short term or the long term if they pinch tissue and cause gangrene or create a blockage in your digestive tract. Really cool stuff, but not to be trifled with around kids and animals.

Dave
PS.....I obviously can't speak to the settings others have suggested you reduce. But even once your settings are golden you may still be looking for some stout hold downs for other reasons and the combination of Sugru and magnets might be really useful to you.

[Kev Williams]

For 2-1/2 months I've been fighting this machine to do Cermark nicely. Trying to find a sweet spot hasn't been easy. My biggest problem has been getting the Cermark to go BLACK. I can roast it down to a goldish-gray with no effort at all. And it looks great, other than it's not black. So how do I get it black? I have to defocus the lens, a lot. I've found my lens's pinpoint focal point by running several 3" long vector lines on black anodized aluminum, and change the focus distance each time. I've gotten the same results after several tests so I'm satisfied it's correct. Measuring from the bottom of the lens barrel, 1.48" is perfect. (the lens is roughly 1/2" above the bottom of the barrel). However, I've found that, even with my other lasers, I get better results on Rowmark and Cermark with the lens closer to the work than the pinpoint setting. Typical on my LS900 is about .025" closer. I found out the hard way that doing Cermark SS at the pinpoint setting results in spots where the Cermark comes off, even with lots of beam overlap.

I've found the same to be true with the Triumph, but the focus change is enormous- to get Cermark to go black I have the focus point at 1.35", more than 1/8" difference. I run at 300mm/s speed, and the power setting I used on this is 51. Based on the mV gauge I figure that's about 50, maybe 55 watts. The lettering and detail is thickened up slightly, but nowhere near what I'd figure 1/8" out of focus would do. If I try focusing tighter, I get goldish gray. If I turn down the power it darkens, but not much. The closer to pinpoint focus, the more the engraving looks like record grooves, even at a .04mm gap, which is 635 lines per inch. Each beam overlap just creates a new groove. The grooves refract light, and no matter how black the Cermark actually is, if theres' grooves it's never black. If I bring the power down, just about the time it starts looking like it's getting good, the Cermark quits sticking and starts washing off. No matter how much I experiment, I always end back up at my 1.35" focus and around 50 power. And if I dwindle the power from there, the Cermark starts washing off.

Another issue is "straight line" rastering vs text and/or graphics. I've found I have to totally separate lines from text, because lines burn much hotter, I assume because when doing lettering the laser's always pulsing, but when doing lines it's just ON. Text I did at 300/51, but the lines I do at 400 speed and 30 power. Any more than that and I get goldish-gray lines. If I do text at 400/30, almost ALL of the Cermark will wash off. That said, this logo is a lot of black, and I could probably get by with somewhat less power. Also, I can probably do a lot less lines per inch would would also help with the heat.

This pic shows some of my testing--

DSC02849.jpg


Closeup of a couple of identical R's -- the bottom shows grooving. It's probably as black as the one above, but the grooves make it look gray. And no grooves in the one above, and it's only barely thicker. Also, note pieces of Cermark missing from the bottom, this is what happens when I get close to 'good' when more in focus and lowering the power.
DSC02850.jpg

[Scott Shepherd]

Kev, have you done a proper power grid yet, or just tests like what you are showing? If you haven't done the grid, take 45 minutes and do it. You'll find the sweet spot a lot easier with the grid, but under no circumstances should it be looking like it's looking. If so, that's a power/tube quality issue, not a cermark issue.

[Gary Hair]

How are you applying the Cermark and how thick? My guess is you are applying it too thick and if you are using a spray can then I'd recommend airbrushing it because I have never seen anyone able to apply it thin enough with a spray can. If you are airbrushing it on then you should have a light enough coating that you can still see the shine of the metal through the coating. If the coating is completely opaque then you have way too much and will get the results you are describing - a black mark that washes off.

To get Cermark to mark consistently you have to do a few things, very few things actually, but you have to do them correctly or you will get bad results.
1 - Clean the metal with denatured alcohol
2 - Spray the Cermark to a very thin layer - so thin you'll swear it couldn't possibly work.
3 - Run a power grid to determine optimal settings

That's it, just 3 things!

Details:
1 - use a microfiber cloth and wipe the stainless until there is no longer any black on the cloth. You must have a clean surface or the Cermark will turn black but not stick.
2 - "thin" is a relative term but in this case it really means "very, very thin". Spray with an airbrush so that you have enough of a coating that you can still barely see the metal through the Cermark.
3 - mark a series of 1/4" squares on the same material you will be marking for the job. Use 100% power, 600 dpi, IN FOCUS, and vary the speed, only the speed! Ferro recommends starting with speed equal to the power of your machine. I have found this to be way off, but it is a starting point. So, for your 80 watt machine you should have the first square at 80% speed, the next at 75%, then 70% and so on. After lasering from 80% to about 30% you should get a scouring pad and scrub the part to see what sticks. If it looks like a range of speed settings seem ok then I would do it again varying the speed within that range in 2% increments. Scrub and see what sticks. This should give you the exact settings.

I have never heard of the problems you are having with focus, that seems to me like it's a bandaid to cover something else.

Try out my suggestions and let us know what happens.

[Dan Hintz]

I'll also add...

For thick (heat-sinking) metals, I use 1,000 lpi/dpi (in your case, 0.025mm scan gap)... 600 lpi/dpi is okay in many cases, but when you want to be sure, 1,000 is it. When you start getting grays, that tells me you're overcooking it. When you take it out of focus so much, you're spending a lot of energy heating up the metal, not the Cermark, hence the warping. You want a fine focused beam to pour all of your power into the Cermark and underlying metal before it has a chance to dissipate into the rest of the metal. Once you find the sweet spot for a focused beam, I think you'll find you're pouring a lot less heat into it than you were previously.

[Mike Null]

I do several thousand dollars a year in Cermark jobs and I have encountered a similar situation to Kev with respect to .125" ss warping. In most instances the Cermark process works beautifully but this one job on 1/8" ss will not run consistently but the bowing of the ss occurs every time.

I have worked the power grid, varied the application of Cermark (I air brush) but I have only 10 out of 50 good pieces. I have varied dpi up to 1000 but my standard is 600.

The piece is 2.5" x 11" with a 1/8" border. The border is the cause of the bowing.

The point is that Cermark is not foolproof regardless of all the various tricks.

[Bill Cunningham]

You may have to split up the area into opposing quadrants and stagger the marking to dissipate the heat and prevent warping.

[Ernest Martin]

I used to have problems doing the Cermark product. I don't use it a lot but I think one of the biggest things is to make sure it's not to thick. It's not one of those cases where more is better.

[Kev Williams]

First, thanks for all the input!

Second, I'm gonna toot my horn a bit... I've been engraving SS with Cermark for 13 years. The knife in the pic below, I did 11 years ago on my LS900. I spend almost $1000 a year on Cermark. Lasering SS is at least 1/4 my workload and I work 16+ hours a day. I'm pretty familiar with the good, bad and ugly of LMM-6000 Cermark!
blade.jpg


My problem at the moment isn't so much with Cermark, it's trying to figure out how to get good results with Cermark from my Chinese laser! I want to point out something about the LS900, which is , when lasering Cermark, I NEVER use less than 100% power. Never. And 20% speed is about as fast as I ever go. If I use less power and/or more speed, I get iffy results. I do watch backs at 100 power, 12 speed, 1000 LPI and 1200 DPI and get beautiful results, and the backs never warp, and rarely get very hot, but mostly because of the small detail. My typical SS settings are 100 power, 18 speed, 500 LPI and 800 DPI. This gives me great results, but these settings are right close to the upper limit. Any less resolution, more speed, or less power and I'm flirting with a non-stick issue. I engrave several 1x3x20ga SS ID plates daily on the 900, they're nearly covered with logo and text, and they do warp a bit. The 900 also warps the large panels, but very little compared to what the Triumph did. Finally, even if I try, I can't get the 900 to roast Cermark like the Triumph does without breaking a sweat.

So I ran a power grid yesterday. As I've noted, I've found my pinpoint focus point (on anodized) to be 1.48" to the bottom of my lens barrel, but my best results with Cermark (so far) is focused at 1.35". So I split the difference and ran the power grid focused at 1.42". Grid incremented from 80 to 10 power, and 150mm/s to 500mm/s. I ran the grids vertically by speed, and changed the power on the fly with the keypad while rastering
downward. I used a .08 gap, which is a bit wide at 317 lines per inch, but that still allows the beam paths to overlap round 50%...
DSC02853.jpg

Found out 2 things--

One, there are no good results to be had at this focus setting. Look at the grid, every square from 20 power up and 150 thru 350 speed is a nearly identical goldish-gray.

Two, changing the power settings on the fly didn't work. It WAS changing the power, it was evident on the mVolt meter. But the results of a CHANGED power setting were different than the results of the same setting sent to the machine in the first place. I can only assume it's how the controller handles computer input vs direct input?

To explain the grid: The first 5 columns were done at the 1.42" focus. The last 3 were done at 1.35" focus. Starting with the 400 column and changed focus, I got blacker results. But that column is skewed because of the power changes, and for some reason, after keying down the power at 30 to 20, it went up to 40, then back to 30, which should've been 10. Note the 400/30 and 400/10 squares were both done at 30 power, but the 10 square is noticeably darker, which is evidence that keying in power changes isn't accurate. Finally, the 450 row was also power-changed on the fly.

The 500 row is the only row that I sent each square to the machine individually. And for the most part, good, almost identical results. Working from 80 power down, the 50 power square finally is a shade darker. and the 30 power setting is visibly the darkest. Yet at 20 power, all the Cermark washed off. That's a pretty big jump, from near-perfect to it-won't-stick-whatsoever within a 10% power setting! The 40 power setting is definitely in the running.

So, this does show that I should be able to get good results, and with a lot less heat. I'm going to do another grid test, but I'm going use text instead of solid blocks this time, and each text block will be sent to the machine individually. I'm also going to run the power settings from 60 to 25 in 5% increments.

[Gary Hair]

I don't understand why you are changing the power... I appreciate your 13 years of experience with Cermark but I think you are making this more complicated than it should be by throwing too many variables into the mix.
1 - In focus
2 - 100% power
3 - 600 dpi
4 - very thin coat of Cermark
5 - vary the speed

[Ross Moshinsky]

500mm/sec = 19.7in/sec. 100% speed on an LS900 is about 70-80in/sec. At 20% speed, you're running at about 14-16 in/sec but only running with 30W. At 500mm/sec you're basically running at speeds that a 30W laser should be running.

I have no experience with this laser but the data says you should run at 500mm/sec (which is about as fast as it's recommended to run a Chinese laser) and any power over 30W. Your experiment more or less confirms this. It also coincides with the general opinion that you are getting greyish results because you're putting too much power into the Cermark.

Obviously this way of calculating things isn't an exact science, but it puts you in the right ball park and from there you can tweak.