PDA

View Full Version : Double Image



Tim Bateson
02-21-2017, 10:43 PM
The Fusion M2 Dual uses most of the same mirrors and same lens, but I'm getting 2 totally different results. The CO2 is producing a double image, but the Fiber is getting it right. Any ideas? This is just simple text, not a graphic.
354570354569

Kev Williams
02-22-2017, 2:26 AM
Found some Fusion pics online- looks like the mirror exit & entry 'holes' are nice shiny aluminum-? I'm thinking part of your C02 beam is just grazing the edge of one of those holes, most likely the one over the lens, and it's altering part of the beam path-

That's my guess anyway! If it is a reflection causing it, try taping over the mirror hole over the lens with some painters tape, let the laser burn thru it for a short spell and see if the lettering stops doubling while the tape is buffering the beam--

Scott Shepherd
02-22-2017, 7:58 AM
Looks like one of two things to me, a software bug, or a bad mirror or lens. The reason it would work on one and not the other would be because it hits the mirror/lens in a slightly different spot. I'd inspect the mirrors and lens under magnification, all mirrors, going back to the beginning, and if they are all okay, then I'd say it's a glitch in the driver, in which case I'd reinstall the driver and software.

Just my guess.

Chris DeGerolamo
02-22-2017, 11:43 AM
If you vector with the CO2, does the beam still appear to be split/ give double cuts? The answer to this question may help determine if it's hardware or software.

Kev Williams
02-22-2017, 12:52 PM
If you engraved it axis swap like this-- 354609 (but I'm assuming you didn't), this would be a LR-RL backlash alignment error, caused mechanically (loose belt, set screws, loose lens) or the software needs an adjustment.

A cracked mirror might cause it I'm sure- however, probably not the lens-- I have 2 cracked 2" Triumph lenses, one is full-on Y crack that intersects right near the center, and it engraves as well as the others...

I'm hardly a software person, but I believe simple hpgl coordinate data is fed to the machine analogously? So to get wrong data from the computer seems unlikely- As for the machine's brain/controller sending doubled-up data to the laser, I assume that's possible, but every time my controller has ever sent wrong data to the machine (any and all of my machines!), the result is more drastic-- laser gantry slamming the wall (LS900), a lasered straight line during a pass followed by no action (ULS), regular engraving turns into a straight line to infinity (CNC machines), and stepper motors going nuts (all of em)... so I'm just guessing it's simpler than that, glass or mechanical problem-

I also guessing the fiber hits a different first mirror than the C02 does? If so, the C02's first mirror could be loose or cracked-

Scott Shepherd
02-22-2017, 2:13 PM
Kev, there are issues from time to time with it being driver related. I've seen it on ULS machines over the years and I think Epilog too. It's very sudden, it normally wasn't happening one job and then the next job and every one after does it. It ends up being something driver related. A reinstall of the driver has fixed that condition many times from people I've seen have it happen. Not saying that's the issue here, but just that I've seen it happen many times. I think it's happened to me once or twice over the last 7-8 years owning a ULS. Reloading the driver solved it each time.

Kev Williams
02-22-2017, 4:59 PM
Makes sense-- and anything's possible with these machines !

About 3 years ago I corrupted my LS900's controller by trying to upload new firmware from Gravo7, and it plain didn't like it. Reloading the older firmware 'takes', but it won't undo whatever damage I did. The result is, engraving from Corel is totally unaffected, but I can no longer engrave anything vector from Gravostyle. I CAN run raster engrave jobs from Gravo, but if I do, then I can no longer run vector jobs from Corel either, even if there was no vector items in the Gravo job. So whenever I run ANY job from Gravo, afterwards I have to delete and reload the LS900 driver so it will work with Corel again...

Tim Bateson
02-23-2017, 5:39 PM
Mystery solved - Loose lens. CO2 moves faster than Fiber so it wasn't "rattling" enough during Fiber jobs to see the issue. Added another paper thin washer to the top of the lens & problem went away.