PDA

View Full Version : focus issue with the fiber, likely of my own doing-



Kev Williams
10-20-2017, 11:21 PM
Me and the BIL re-arranged the fiber setup yesterday, it's been sitting on not-the-flattest table in the world so we stuck a flatter piece of plywood under the thing. It's also been propped up on pieces of 2x4's and shims and stuff to level and steady it. We weren't being exactly dainty with it, but it IS a 'portable' unit so I figure it should handle some light abuse ;)...

anyway, got it set up, and for once I got the top of the scanhead and the top of the table dead to rights center-bubble. The distance from the worktable top to the lens housing is the same all around the lens and table...

So I had a 6x12" aluminum license plate to engrave an oval logo that worked out to 6.3" long x 4.08" tall. After the first pass, the bottom 1/3 of the oval-logo was barely engraved while the top 1/3 was bright white. I ended up raising the table quite a bit just to get the bottom to start engraving decent. This is across the 4" height, and I have a 6" lens--

SO I just just drew up some skinny ovals and stretched them along the X and Y axis's to determine the absolute max work area allowed by the mirrors: 159.1mm X, 167.2mm Y. Beyond this, the mirrors would clip the ends of the ovals. Doing this I also found that the X and Y dead centers are very close, X to the right less than a mm, Y to the top by about half a mm

So I ran a continuous 6" tall x about 1.5" wide oval at zero power over some black anodized, and kept shimming up the bottom of the table and re-adjusted the central focus until I got an evenly marked oval. I then measured the distance from the top of the oval to the lens, and the bottom, and it's a 5mm difference!
Drawn to scale in Corel it looks like this:
370101
Otherwise, everything is fine!

So, I'm thinking this is must be just an optics thing, but I'm trying to figure out what's going on. We didn't just DROP the machine or anything, but it was jostled some. So if for example, when we moved the machine, the-- I have no idea what it's called, but the 'tube/housing' at the end of the fiber cable that's clamped down on top, if that housing moved and the incoming beam hitting the mirrors is now skewed a bit, could that account for the wacky focus, even though the beam seems to be centered up nicely in relation to the work area AND engraves okay?

--I've noticed with my Triumph 'dumpster' when using a hand-held laser pointer aimed at he mirror above the lens, that the final dot position moves very little in relation to where and how the beam hits the mirror, so I assume the lens is pretty forgiving as to where it centers the beam spot. And so I'll also assume (probably wrongly ;) ) that these fiber lenses are a bit forgiving too, at least concerning position. But the nearly 1/4" difference in focus distance across 6" of Y axis space has me stumped. X focus is fine...

I haven't pulled the covers and looked inside yet, and I also haven't done a 'size-wanted v size engraved' test yet, which I'll do shortly, got other fish to fry at the moment :D

Any ideas?

Scott Shepherd
10-21-2017, 9:08 AM
Aligning a fiber is a trick for sure. It's hard to align something that's invisible :) Just for reference, the beam at the top of the unit, is about 6-7mm in diameter when it goes into the galvo head, so no point in trying to figure that out. What I have done in the past is close to what you are describing. I draw a square, as large as it will go (200mm, 250mm, or whatever), them put a sheet of anodizing aluminum on the table. Put the power down to where it barely leaves a mark. Engrave the box. You'll see what you already described, it engraves some places, not others. If it's left to right, break the bolts loose on the galvo head and rotate it. Do it all again. If it's front to back, you'll need to shim the housing. What you should end up with is a square that engraves exactly the same as the lowest power.

That's the best I have been able to come up with.

Kev Williams
10-21-2017, 1:04 PM
This laser has always had slightly less power towards the bottom, but not like this! The other 'option' in my mind is if the all the beam isn't hitting all of one or both mirrors due to initial beam angle. Or maybe the beam isn't well-centered entering the scan head itself and part of the beam is hitting the wall around the hole..? -if memory serves, the hole the beam enters is like 1/2" dia. or less, which doesn't leave acres of margin space for a 7mm beam to pass thru, so maybe the head just needs to be slid up or down a smidge? Beam obstruction could also come into play when shimming the scanhead to tilt it slightly. It WOULD be nice to be able to see the beam-- maybe a thermal imaging camera is in order? :D

I've been needing to 'shim' the scanhead big time anyway (like moving it 6-12" out from the box) so this might be a good time to get a spacer made and make that happen too...

Kev Williams
10-21-2017, 4:39 PM
OK-- I pulled the scanhead off and removed the sides so I could explore :)

First thing I did was look down the pipe to see if there was anything interesting-- there was...
The red light and its mirror- which is pretty cool by the way, a transparent mirror that lets the fiber beam and all other light pass thru, and still reflects the red laser. Anyway, I didn't care for the position of the light an mirror, they seem to hang down a bit low in the pipe to suit me, encroaching into the fiber beam path zone...
370138
So, how to raise the light & mirror up? I thought about shims, then I noticed the top cap of the red laser fixture has 4 mounting screws , and 4 adjusting screws in the corners...
370139

Well, that's easy enough- I turned in each screw 1 full turn, then tightened down the cap-
It didn't move it much, but the led and top of the mirror are definitely more out of the line of fire now!
370140

Also in pic 2, note the snout of the laser tube, it WAS nearly a full inch from the face of the led case, I moved it to within about 1/8", I figure if nothing else it'll help with dust and keeping bigger bugs out! ;)

Also note the knurled ring screwed to the threaded snout on the front side of the led case. Why the threaded snout I'm not sure, other applications I guess, and whether the ring is on or off won't affect a thing. It was off for the mirror adjust pics. I screwed it back on just barely, then had a look--
--And this may be the cause of my problem:
370141

Note how low and to the right off-center the laser assy is to the scanhead mount hole- That's pretty bad! And- probably of my own doing, because the first time I ever broke into the machine, the 4 screws on the front panel under the hole (last pic), I loosened those in case I needed to... turns out I didn't need to, BUT they hold up the plate that holds everything seen in pic 2! And I didn't exactly get those screws tightened very tight I just found out... SO-- could be our jostling from yesterday let the support plate drop... ?

So I used a screwdriver gently as a pry bar to move everything up/left while getting the panel screws nice & snug--
-the laser assy is now centered MUCH better!
370142

I'll definitely have to re align the laser size & shape and re align the red laser, but hopefully this does the trick!

Scott Shepherd
10-23-2017, 8:52 AM
Good luck Kev, sounds like you might have figured it out. It's a shame there isn't an easier way to see the beam path. Sadly, it it's going to mark anything at a 7mm diameter, and once it's into the mirrors, it's too late. I've seen Chinese newspaper clippings used to shim these things up before :) I think you'll be pleasantly surprised at how much better it runs when it's aligned.

Kev Williams
10-23-2017, 2:15 PM
I'd LOVE to figure out how to 'see' the beam....

After I moved the red light, I had a look at where it was hitting the mirrors; it wasn't even close to their centers. Dropping the light back down didn't change much, so I decided to use the corner screws to actually adjust it. After a few minutes I had the beam as close to dead center of both mirrors as I could get it, with the mirrors parked at dead center...
370261

This a 1/20sec shot of the mirrors doing a max sweep at 3000mm/sec, I was surprised how much of the 2nd mirror is utilized, which tells me laser alignment IS fairly critical at least!
370260

After doing this I knew I'd have to re-align the red to match actual, which I did. But man, is the actual a long ways off from the red now, the red tracks around 2-3mm to the right of the actual beam now, before it was like .020" inch or so.

So then, if my red light is hitting the mirrors dead center, where's the actual beam hitting them? I tried to figure that out by placing various materials at different power settings in the unfocused beam path, and nothing would mark. What I didn't try was tin foil, but I did try hitting some magic marker on a laser brass, nothing. The laser's beam path would be easy enough to change with paper shims here or there, but you can't adjust what you can't see!

As it is, it's working much better than before but it's still leaning towards wanting a closer focus distance at the Y bottom than top. For 90% of what I do it'll be fine. For now ;)

Dave Sheldrake
10-28-2017, 12:45 AM
It WOULD be nice to be able to see the beam-- maybe a thermal imaging camera is in order?

Laser beams have no mass Kev, hence no temperature :)

Kev Williams
10-28-2017, 1:45 PM
Laser beams have no mass Kev, hence no temperature :)
--always a fly in the ointment! or in my case, getting roasted onto my lens!

so then, how DOES one line up a fiber to the mirrors? Just keep fiddling & shimming until the beam hits where the red dot is?

Would the beam show us as heat if it's hitting say, a thin piece of polycarb?

Dave Sheldrake
10-28-2017, 10:29 PM
Would the beam show us as heat if it's hitting say, a thin piece of polycarb?

Yup, shows up very nicely, you can also use a digital camera in IR mode (night shoot) and with a bit of settings tinkering (on the camera) see the beam quite well.

FLIR works if you have slightly smokey air as well in the room, the beam heats the particles and can be *seen* that way but dust and smoke around the galvo head when it's open can be a bit dodgy to say the least ;)

Scott Shepherd
10-29-2017, 11:43 AM
so then, how DOES one line up a fiber to the mirrors? Just keep fiddling & shimming until the beam hits where the red dot is?

Would the beam show us as heat if it's hitting say, a thin piece of polycarb?

Kev, my take on it has always been math/alignment related. You can measure with calipers the distance from the base it all sits on to the center of that tube that the beam comes out. Write that down. Then measure anything else you can to see how that lines up. I've made a few pieces that allow me to transfer that measurement to the entry way into the galvo head. If the math adds up, then it's going to be in the center, or pretty close. I always said that when I get a chance, I'm going to make a tube on the lathe that slips over the beam exit tube that is necked down to slip into the hole going into the scan head. I'd pull the red dot mirror system off, align the beam exit and glavo entry with my alignment tool, then I'd put the red dot mechanism back on.

I've seen some crazy alignment issues over the years, coming out of China. I've seen towers cut crooked, so then they tried to make the head and beam square by shimming just about everything on the machine. Hey, how about just skimming the bottom of the tower so it's square? Then you wouldn't have had to shim everything else in the chain. It's a wonder to me that they ever work after the things I've seen. Just some basic knowledge of alignment would go a long way for them.

Kev Williams
10-29-2017, 5:16 PM
one other goofy alignment issue I have, is the engraving isn't square to the scan head, it's off about 1/16" inch across 6",
looks like this-
370566
and since it was the same way with the other scanhead, I'm assuming the beam is entering the head on an angle. It's not a big deal, for plate & stuff I have a corner setup that's aligned to the engraving, and most other stuff needs to be aligned on the fly anyway. But I wouldn't mind it being right! ;)

As for shimming the machine, I have it propped up about 1-3/4" off the table, on wood blocks. I've tried leveling via shimming the machine from its feet but a teensy bit of shim at the bottom is amplifed by the 16" to the head. Too frustrating! What I did that's working well is, I got the machine fairly level and steady, and I just put screws in the 4 corners under the scissor table...by adjusting those screws I got the table so it's near dead-on plumb-level-square with with lens frame. Anywhere I measure from the table to the lens is identical as measured from the opposite side of the table. That's helped!

I need to try a little math as you suggest, I might be able to find that off-angle issue and fix that. Then, I need to zero out the red light adjustments, then see if I can get beam to fire at the same place the red dot is, that SHOULD put the beam pretty close to the centers of the mirrors. I hope!

Scott Shepherd
10-29-2017, 7:30 PM
Kev, if memory serves me correctly, you can change that skew of the box. I'll have a look, but I seem to recall that's a setting. Under "Field" on the Parameters page, you can adjust the shape of it all on both mirrors. I think you need to tweak those for each scan head mirror and you'll get that dialed in, then you can map the red dot over it all pretty easily.

Kev Williams
10-30-2017, 11:56 AM
Yes you can change the skew, but- as far as I can tell- only to align the Y axis sides square to the X axis top and bottom, I haven't been able to get the angle of the X lines to change.

Scott Shepherd
10-30-2017, 12:49 PM
Are you adjusting both Galvo's or just 1? One controls X, one controls Y, if memory serves me correctly.

Kev Williams
10-30-2017, 1:43 PM
I've adjusted both but to be honest I probably need to experiment more--
here's the menu-370612 -these are demo-screen factory defaults btw, not the machines current-
I just now noticed the "angle" adjustment I have the arrow pointed to, I wonder if that'll square up the engraving? Maybe?
The trapezoid and angular settings are what I've messed with so far, but yes, I do need to experiment with both mirrors more...

also-- anyone know how to SAVE a 'correction file'?

John Lifer
10-30-2017, 2:04 PM
Copy the folder? Several config files can't 'see' a couple with notepad, expect they might contain some of this setup. I'd get it correct and then just copy the whole folder to a zipdrive or other storage disk/space.