CNC Laser Shaking when Changing Directions Rapidly.

[Brandon Bonenfant]

Hey everyone,

So I have recently upgraded one of my lasers to a Laguna LCB Full Table Laser (60"x120") and I'm having some issues with the Y Axis.

What seems to be happening is when the laser is moving in the negative direction, and then quickly changing to the positive direction, the gantry starts to shake and give a very bad / rough cut until the machine stops shaking. I've worked with the company to get things as good as we can and it's still not even close to the quality of my other lasers.

I was wondering if anyone has any ideas on how to make this better or if there is a way to add a small delay before abruptly changing directions to avoid the shaking. If I run this machine at lower speeds it is better, but this plate is only being ran at 30mm/s which is about 1/20th of what the advertised speeds were.

I use RDWorks for running this laser and this does the same thing with many files created in various programs. This issues also only shows on curves and does not seem to be an issue when ran on straight lines.
385327
Thank you,

- Brandon

[Dave Sheldrake]

check the stepper motors, you may have a bad wire or duff motor...best guess with what you have said? it's electrical rather than mechanical, also check the rail bearings, one of them binding can cause problems as well although it's not quite so likely

[Bill George]

How solid is the machine, is it on casters or directly on the floor? My guess after checking for loose wire connections is something is not tight enough on the machine and its got a bit of slop and rapid movements cause the machine gantry to shake.

[Brandon Bonenfant]

So the machine is directly on the floor, there are casters, but they are elevated off the floor by the leveling feet. I'll definitely check the wiring in the morning, but I fear it may be an issue with something being loose.

To add more:

When the machine arrived, essentially every screw was loose or had fallen out and had to be reconstructed. This involved disassembling most of the machine and reassembling it with help of one of their technicians. Once this was fixed the gantry still shook like crazy when moving. To fix this I had to tap a few new bolts to secure the gantry better and turn the acceleration down from 500mm/s to about 5-10mm/s. This helped alleviate the shaking a lot, but it was unfortunate since the acceleration killed my cut times and made plates take much much longer.

Once all of this was "fixed" we found the edge quality issue. To add, the gantry weighs 200-300 pounds and is roughly 7' long 1.5' wide and sits on 2 glides that are only about 6" wide. So this thing is not supported that well at all. I'll try to get some better pictures tomorrow to show this a bit better, maybe just supporting it better would work. From the picture it looks to be supported fully, but it does not actually sit on those rails, it hovers above them. 385353

[Bill George]

To clarify for everyone, the Y axis in this case is the movement on the 120 inch or 10 foot long part of the table? Because moving 200-300 lbs is going to cause something to move or shake If its not done right. Laguna has not had very good reviews on some of the lasers, I do not know about the CNC machines but you can check in that Forum.

I think If you could post a video or two it would help. There are some really sharp people on here.

[Brandon Bonenfant]

To clarify for everyone, the Y axis in this case is the movement on the 120 inch or 10 foot long part of the table? Because moving 200-300 lbs is going to cause something to move or shake If its not done right. Laguna has not had very good reviews on some of the lasers, I do not know about the CNC machines but you can check in that Forum.

I think If you could post a video or two it would help. There are some really sharp people on here.


The issue with the movement is on the Y Axis that is the 120 inch portion of the machine.

We have one of their CNC Routers in our shop and it works great and was built well, we went with one of their lasers due to this. Sadly I looked past some of the initial bad reviews due to our past experience with them. So far it has been quite disappointing and an uphill battle with this machine. To top it off, they no longer offer any of their flatbed laser models and removed them from their site about a week prior to my machine being delivered. I'll try to get a decent video tomorrow of it running, but you can visibly see the shaking when it switches directions too abruptly, which is probably due to swaying that much weight around.

Thanks for the responses, I'll check back with the video tomorrow!

[Kev Williams]

As Dave mentioned, a bad wire or connection can cause some 'violent' shake, I've been on the receiving end of this 3 times now, twice with my 80w Triumph...
385368385369

and a few months ago with my 14 year old Gravograph LS900-
385367385366

all three times was the Y axis, and all 3 times it was these bad wires in the 'drag chain'. (Second time in the Triumph was my repair breaking, which I was expecting)...

https://youtu.be/mODbcMcuIfU << this is a link to a video of my machine messing up, it's about 3 minutes, and when I made it I hadn't found the broken wire yet, I was still troubleshooting-- if you fast forward to just before the 1 minute mark I start the machine, and at about 1:12 the 'violence' starts up, and goes for 3 holes, then it runs fine for the most part..

You likely don't have a broken wire yet, and as you'll see if you watch the video, my results were a ton worse than yours, and happened in mid cut, not changing directions...but still, you could have an iffy ground somewhere or a loose wire in a connector, etc, so worth a check...

That all said, I'm leaning more towards your software just needs adjusting, specifically the accel and decel rates...
These 2 photos show before and after results with my Triumph, the before engraving is the upper-left run, the after the lower-right run...
385370 385371
-the first run was at factory default settings, the second run was after playing with the accel and decel rates. I couldn't tell you which ones I played with, and probably wouldn't matter anyway since it seems all different far-east laser mfr's use different terms for the same thing ;)...

At any rate, my first runs show similar wobbles as in your photo. Slowing down the rates made a drastic improvement in the second run, still not perfect but I lost a lot of the wobble...

It's possible your software is set up to move a gantry about that weighs about 500 pounds less than yours! The machine DOES need time to slow down and accelerate all that mass... :)

.

[Bill George]

I guess my Free advice for the moment is if the machine is still in Warranty I would either have them fix it or come and pick it up. All the hardware and screws were loose when it was delivered and the factory tech came out to repair, and now the unfixable issues I would consider looking elsewhere. It would be interesting how fast the large Kern laser runs without issues?

[Dave Sheldrake]

If the gantry has been stripped or removed my money at this point is Crab gantry if it's not the motors, cut a square of 200mm x 200mm and check the corners for angle, at the factory the rails and gantry's are aligned with lasers much like tracking a car's rod ends, the bigger the machine gets the more critical it becomes due to the small bearing area

[Brandon Bonenfant]

Thanks for the replies everyone,

I'll post an update this afternoon once I test the things mentioned in the previous posts! Hopefully something helps alleviate the unknown on this thing. The return policy may be the way to go soon though, trying to chug along since it is hard to keep our shop down for 8+ weeks. (5-8 weeks for refund / pickup, then who knows how long to get another machine on order). Thankfully it still does shapes without curves fine, helps knock out many of the orders we have.

I'll let everyone know once I get it testing a bit this morning.

[Brandon Bonenfant]

Alright,

So first off, there do not seem to be any issues with the wiring / motors. All wires are tight and the issues with the cuts happen at the exact same location on the objects with curves every time. So I don't think it is a skipping motor.

I realigned the gantry by removing the belts and pushing it towards the stopping blocks while talking to a technician, the same way they align it. (Sadly they do not use laser precision when aligning these).

The gantry does have some play in it and you can hear it rattle during hard turns. I may be able to add a few more bolts to help secure it better. If this does not work I will probably just go for the refund on the machine and eat the lost time. This 100% seems like a major problem and there does not seem to be a good fix for it. There is a reason they discontinued these and the build quality is probably the reason.

I did play with the acceleration settings and this did help a lot of many of the pieces, however it moves at a snails pace and I cannot get it to run over 30mm/s without seeing similar issues. (Better than it looked before but still not good enough).

I'm going to play with it this morning and try to secure it better, but I think it was said perfectly above, they did not make a machine capable of moving a 300lb gantry smoothly.

I'll see what happens and try to post once I figure something out or decide on the refund.

Thanks everyone,

-Brandon

[Bill George]

Well if they can not fix and you can't use the way it is, I personally would return as you have decided. Of course the next question, outside of paying $100k+ for a Kern which might be for you the right thing to do, where else do you get a good large format machine? I would contact Kern and ask, if you have a production line running you want quality.

[Kev Williams]

...I did play with the acceleration settings and this did help a lot of many of the pieces, however it moves at a snails pace and I cannot get it to run over 30mm/s without seeing similar issues. (Better than it looked before but still not good enough).

I still think most of the problems are speed-settings related, and the factory should be able to help you with that.

My Triumph has several parameters menus, these 3 are the user, crafts and factory parameters...
385537385538385539
The first contains more 'crafts' parameters in the top section, and lots of speed settings; space speed, space acc, space acc acc (?), initial speed, min acc, cut acc, cut acc acc, sweep acc, speed ratio, carving wrap accel...I'm not sure what they all mean, but the adjustments I've made have affected the speed when running curved shapes, and slowdown rate when the machine comes to a corner where an abrupt direction change is about to happen. These adjustments have helped as the pics above show, but they didn't affect the machines "main" speed...

Second menu is also 'crafts parameters', this particular menu focuses on the speeds of arcs and circles, 'curve discrete' which is, best I can tell, how to tell the machine to "add" more phantom nodes in an arc to enhance curve smoothness, and 'carving backlash', which fine tunes the left-right/right-left raster alignment. The "small round speed limits" are fantastic once learned, you can run 300mm/s straights, and when a rounded corner (1/2" radius for example) is encountered, it will run at your entered speed limit, say 40mm/s, then return to full speed once out of the arc.

The last menu is the 'factory' parameters, which are also adjustable. In my machine I've increased the max X speed to 800mm/second. Not sure if any of the other adjustments I made were in this menu or the other. (these menus are all non-adjusted factory figures)... Anyway, I'd love to experiment further and figure out changing some of the Acc-Acc numbers actually do :) --all I need is to find a few hours with nothing to do... lol

[Jerome Stanek]

Maybe change the ramp settings to a little lower

[Dave Sheldrake]

talking to a technician, the same way they align it. (Sadly they do not use laser precision when aligning these).

there's the problem, that's entirely the WRONG way to sort out the alignment on a large format laser, it MUST be done with accurate equipment or you will never get the thing to cut correctly...if a companies tech was telling me to do it like that I wouldn't be going back to them at any point


The gantry does have some play in it

There shouldn't be any lateral play , neither axial or torsional on a gantry that big...the least it will cause is resonance problems, in worst case it will trash the bearings

[Brandon Bonenfant]

Hey everyone,

So I managed to adjust the settings and got the cuts a ton better! There was still a bit of play, but not nearly as much as before.

So I ran a test plate of a bunch of circles and the first plate went flawlessly........

but then.....

The second plate started great and then did almost exactly what Kev Williams' machine did!!!! So it looks as if there is an issue with the Y-Axis Wiring!!!!!!!

I'm really hoping this will help alleviate the remaining issues I am seeing.

I ripped apart the drag chain and did not see any visible damage on the Y-Stepper Motor Wire, nor did I see any issues when tracing it back to the control panel and power connections. This problem seems to happen more and more as I run the machine, but I cannot find the tear!

Does anyone have any tips / tricks on finding the issue with the wiring without straight replacing the entire wire?

Thanks so much for everyone's help so far!

- Brandon

[Kev Williams]

On my machine, they used off-the-shelf "Instar" steppers- nothing at all wrong with them- but they come with wiring that's too short, so Triumph just added to the wires. How I figured out where to look was because I finally noticed the machine always messed up at around 16 to 18 inches down in the Y axis, so I slewed the gantry down 17" and started pulling the chain apart, and where the wires were spliced was just going up into the loop. I slewed back up towards the top, and fiddled with the wires, and sure enough, it would go nuts every time I fiddled... The wiring LOOKED FINE, I had to remove nearly all the chain clips(?) so I could pull the wires out and apart, and that's when I found the black wire drooping in the middle. The solder joint broke, but worked fine as long as the wires laid flat...

SO, check for butt splices in your stepper wires. Also, check where the wires run thru the cabinet, could be a wire is against a hole with a sharp edge, it wouldn't take many trips back n forth with a little tension on the wires to cut into the insulation and short the wire to the cabinet...

FWIW, after I fixed mine the first time it did it again about 4 months later- solder don't bend! Second repair, IIRC I just hand twisted the wires for about an inch and finished with heat shrink tubing.

And-- when the LS900 started doing the same thing, I knew exactly what it was! And the machine showed me where the problem was as the wire was arcing off the cabinet, bright sparks are easy to see! Those wires weren't spliced, the ground wire I assume deals with the most heat, and I suspect the insulation wore itself down rubbing against the chain...

Good luck with it, I really hope you don't have to send it back!

[John Lifer]

I'm stepping in and saying if it looks to be the Y axis only, then get replacement wires out, cut and installed.

I've a cheap 3d printer I bought and put together a couple of months back. Super fine wires on the darn thing. 26 gauge max on most of the thing, some 30 or smaller. Added drag link run on the x axis and within a week, was having issues with fan stopping. Replaced wires, all the way as I could get close, but wire was broken inside insulation and no apparent damage. Same thing for heating cable a few days later. Replaced that one and changed up the links a bit, so far no further issues. So if you think it's wiring, You might spend a little time trying to find, but I'd spend the effort on just replacing.

(Yeah, it will be a pain) But at least you think you might have solved it.

[Bill George]

As a former air craft electrician I can tell you solder for splicing wires that flex went away eons ago. All we used was good grade Mil Spec crimp on splices and terminal connectors. It was crimped twice, once for the wire and the other for the wire covered insulation. Some of the commercial grade crimping tools did it all in one operation. T&B brand is one I try to buy now as they are the closest to Mil Spec.

[Brandon Bonenfant]

Thanks everyone,

The wiring issue seems to be fixed now. I ended up spending a bit of time looking for the issue and ended up replacing the entire line like mentioned above. They had 4 spots where the wire was solder from the stepper motor to the control board. Seemed like they built it is steps and used small pieces of wires. I ended up getting replacement wire from the local hardware store and replaced the run with a single pieced and crimped and heat shrank it together. Seems to be working well for the last 4-5 hours!

There is still some issues with the shaking, but I'm currently messing with belt tensions and such to try and alleviate this as much as I can. I'm getting very very close to having it running perfectly!

Thank you all so much for everything! This really is an amazing community of people!!!

I'll try to post some before / after pictures of the cuts once I have it squared away completely!

-Brandon