Chinese Laser with Servos

[Tim Drinkwater]

This is my first post but I've lurked off and on for awhile. Thank you everyone for the valuable information that you share.

A bit of background. I'm shopping for a laser machine to mark anodized aluminum. I own a machine shop and we machine a lot of parts that get anodized and our customers would like us to mark them as well.

I have pricing from Trotec, ULS and I'm still waiting for Epilog. I really like the Trotec machines but the ROI on the big three just isn't there for my needs at this point. If I get enough work on the Chinese machine I'll probably buy a Trotec for marking and use the Chinese machine for cutting but that would be down the road.

Thus my endeavor into the Chinese machines. I've been talking with Weike and Shenhui. Weike has been a bit more responsive but Shenhui has a bit better pricing. Presently I'm leaning towards the Weike LG1200. I've asked Weike about upgrading to servos on the x and y axis. The cost to upgrade to belt drive servos is $2650 extra which includes the Leetro 6575 controller. They said it would be faster and I assume a bit more accurate.

So what do those here think. Is it worth upgrading to servo motors and drives? Pros and Cons?

[john banks]

My 100W RECI tube does turn off neatly enough to raster over 400mm/s, so this limits us whereas the steppers run 1000mm/s so are not the limiting factor in our case.

[Tim Drinkwater]

That is very interesting. I hadn't considered the response time of the laser itself being the limiting factor.

I had been planning on a Reci Z2 (90 watts nominal 100 watts max). Are the Reci tubes the same as those used in CO2 galvo machines? It would seem that a laser for a galvo would need a very fast response time.

[john banks]

Would be interesting to know the same re galvo. Some other users with RECI 80W have not seen the same thing, but both my tubes and power supplies do the same. I had accidentally typed does instead of doesn't in my original reply, my tube is too slow but I think you got my message.

[matthew knott]

you can get galvo machines with Reci style (glass DC) tubes, how they perform i have no idea, probably ok as the galvos are so fast that they can skip between non-marking areas then a slight delay can be programed at the start of engraving to compensate for the response time of the laser, it will slow the marking process down but you would not notice it unless you had a high speed machine to compare against side by side.

If i was you Tim and you want to see how things go first dont bother with servos, more expensive, more complex, and a pain if they go wrong, stepper motors are simple, very reliable and do a pretty ok job.
Also maybe look at a fibre galvo, as i suspect it might be better for what you are trying to do.

CHeers

[Dan Hintz]

I see no reason to spend the extra money with servos over stepper motors. ULS uses stepper motors and I don't think you'll hear any real complaints about speed (same speed as Epilog). Save the money and put it towards a western machine when the time is right...

[Tim Drinkwater]

you can get galvo machines with Reci style (glass DC) tubes, how they perform i have no idea, probably ok as the galvos are so fast that they can skip between non-marking areas then a slight delay can be programed at the start of engraving to compensate for the response time of the laser, it will slow the marking process down but you would not notice it unless you had a high speed machine to compare against side by side.
So a galvo doesn't shut the laser off it just moves fast enough between unmarked areas that it doesn't leave a mark?


If i was you Tim and you want to see how things go first dont bother with servos, more expensive, more complex, and a pain if they go wrong, stepper motors are simple, very reliable and do a pretty ok job.
Also maybe look at a fibre galvo, as i suspect it might be better for what you are trying to do.

I thought about galvo but liked the idea of versatility that the gantry style could add. I guess if I acquire enough work on the chinese machine I could also look at a galvo, should be even faster than the Trotec.


I see no reason to spend the extra money with servos over stepper motors. ULS uses stepper motors and I don't think you'll hear any real complaints about speed (same speed as Epilog). Save the money and put it towards a western machine when the time is right...

I had forgotten that ULS is getting it done with steppers. I guess I just assumed that servos would be more robust and reliable. Very good point about saving the upgrade money for a better machine as that is what I'd prefer to have anyways.

Thank you for the feedback thus far everyone. Very good food for thought.

[matthew knott]

A galvo still shuts the laser off in exactly the same way but the enhanced control and massive difference in acceleration times is what makes the difference, it would just enable you to tune the slow response of a glass laser to the motion system to mask any problems. Any decent system would use a 'western' laser source.

Also in my experience steppers are very robust and reliable and easy to fix/replace if things go wrong. What versitity firm the gantry system do you prefer? We mark 1000's of anodised parts a week and never use our gantry laser as we find it restrictive

[Rodne Gold]

I run both types , IMO it does not warrant the extra expense of servos.
I liked the servos in my CnC machines and lasers as it can go back to a job if you stopped it and moved the heads away, but my stepper shenuis do the same.
Closed loop servo is a double edged sword on our CnC's.. it actually limits the speed when doing things fast. It actually has worse cuts when doing ultrafast work as the head overshoots due to inertia and then the encoder corrects etc...
The chinese lasers heads and lens assemblies are fairly heavy and massive compared to others and you could have the same problems.

[Scott Shepherd]

Closed loop servo is a double edged sword on our CnC's.. it actually limits the speed when doing things fast. It actually has worse cuts when doing ultrafast work as the head overshoots due to inertia and then the encoder corrects etc...

That must be an issue with your brand(s) of CNC. CNC milling machines have been using closed loop servos for decades and they do ultra high speed machining of aerospace parts with the highest precision and no overshooting. It's not a function of the servo and closed loop system. That system works just fine on CNC milling machines.

[Tim Drinkwater]

Also in my experience steppers are very robust and reliable and easy to fix/replace if things go wrong. What versitity firm the gantry system do you prefer? We mark 1000's of anodised parts a week and never use our gantry laser as we find it restrictive

I'm glad to hear that you feel the steppers are reliable. Part of my concern with steppers is randomly loosing a part because of a missed step something. My parts aren't horribly expensive but scraping parts on the last operation is always painful.

The ability to cut sheet goods is the extra versatility that I'm after. At this point I kind of see the chinese gantry machines as a general purpose machine, not the best at any one task but can get it done. My general plan at this point is to get the chinese machine as a starting point then buy more appropriate equipment that fits the work load.


I run both types , IMO it does not warrant the extra expense of servos.
I liked the servos in my CnC machines and lasers as it can go back to a job if you stopped it and moved the heads away, but my stepper shenuis do the same.
Closed loop servo is a double edged sword on our CnC's.. it actually limits the speed when doing things fast. It actually has worse cuts when doing ultrafast work as the head overshoots due to inertia and then the encoder corrects etc...
The chinese lasers heads and lens assemblies are fairly heavy and massive compared to others and you could have the same problems.

Sounds like it's a servo tuning issue, they aren't breaking soon enough or hard enough. On my industrial machining equipment those settings are adjustable. Interesting to know that the chinese heads are heavy, that is something I could likely modify...

Looks like the vote is pretty unanimous for steppers. In the machine tool world steppers are shunned I guess that is what made me concerned.

[Joe Hillmann]

I don't do much anodized engraving but I would think your speed would be more limited to what gives you a nice clean mark rather than how fast the machine can actually go. Maybe someone who does a lot of anodized can chime in.

Steppers aren't nearly as much of a problem in lasers as they are in other machine tools. There are only a few times the steppers will be off, if you run the head of the laser into something, if the belt or cogs on the motors aren't tight enough, or if you try and accelerate / deaccelerate it too fast. It is very easy to prevent any of those from happening.

[Mikah Barnett]

I bought my G.Weike specifically and primarily for marking our anodized aluminum products. So far, it's been great for reliability and consistency once you learn to use the wonky LaserCut software.

I don't have any experience with servo machines, but the steppers in mine are performing well in everything we've attempted thus far. I will say I wish I had a little more control over laser on/off timing/process, but with the stepper controller I've got it's just not in the cards. And, so far, I've never had an issue I couldn't work out one way or another.

As for scrapping parts - we've had to do a few and we re-etched, tumbled in walnut shells and rouge, re-anodized and were back to new. Not ideal, but better (at least in our case) than scrapping the part.

[Dan Hintz]

art of my concern with steppers is randomly loosing a part because of a missed step something.
A stepper missing steps is no different than a servo's optical feedback mechanism encountering dust bunnies (and how many Epilog machine threads have we seen over the years that freaked out in the middle of a run and were cured by a good house cleaning of the optical strip?). Keep the optical path clean for servos and make sure your stepper motors are not being driven too close to their resonance points under load. It's a different problem for each, but the bad result is the same in both cases. Easily solved in both cases, too.

I don't do much anodized engraving but I would think your speed would be more limited to what gives you a nice clean mark rather than how fast the machine can actually go. Maybe someone who does a lot of anodized can chime in.

On my 60W ULS, I'm usually around the 20-25P range... if I could find a machine that was 4 times as fast, I'd still be good, so speed of the gantry is the limiting factor. Trotec's are twice the ULS speed and you're still way down on power. We need faster gantries :D

[Rodne Gold]

Im using a Tekcel - a rolls royce in terms of CnC routers..the speeds I am talking about are really fast ... taking .2mm curved engraving cuts in engravers brass with a high freq spindle at max speed of 300-400mm per sec , the engraving is no where nearly as good as at 1/2 that speed

[Joe Hillmann]

A stepper missing steps is no different than a servo's optical feedback mechanism encountering dust bunnies (and how many Epilog machine threads have we seen over the years that freaked out in the middle of a run and were cured by a good house cleaning of the optical strip?). Keep the optical path clean for servos and make sure your stepper motors are not being driven too close to their resonance points under load. It's a different problem for each, but the bad result is the same in both cases. Easily solved in both cases, too.


On my 60W ULS, I'm usually around the 20-25P range... if I could find a machine that was 4 times as fast, I'd still be good, so speed of the gantry is the limiting factor. Trotec's are twice the ULS speed and you're still way down on power. We need faster gantries :D


I stand corrected. Ignore what I said earlier about speed.

[Brian Robison]

Not to go off subject too far, but 100w to engrave anodized aluminum?
Way overkill.