PDA

View Full Version : It's a frustrating day in the neighborhood, won't you be mine?



Dan Hintz
11-08-2009, 6:43 PM
Just thinking of the happy-go-lucky ways of Mister Rogers and how he really should be here right now.

The left half-inch or so of my table doesn't engrave with as much power as the rest... very frustrating.

Lee DeRaud
11-08-2009, 8:03 PM
The left half-inch or so of my table doesn't engrave with as much power as the rest... very frustrating.Three questions:

1. Do you have the same problem on the righthand 1/2"?

2. Is the problem speed-dependent? I.e. if you drop both speed and power by half, do you get the same symptoms?

3. Does the PLS have "overscan", i.e. can the carraige move left of 0 and right of max-X to give it time to ramp up speed for the next scan line?

(I think you can see where I'm going with this...)

Eric Seest
11-08-2009, 8:52 PM
I have the same problem on my PLS6.60, when I found this was a problem over a year ago, I asked ULS service about it and this was their answer: The laser does not have enough time to activate full power as it travels that close to the edge of the table. So they told me that in order to fix this place engraving areas 1" away from the left side of the table. So what I have done is us the ruler off of my combination square to space everything to the right 1". Hope this helps. Unless someone else has a better answer.

Brian Jacobs
11-08-2009, 11:49 PM
Same problem with my VersaLaser. I'm loosing about 3/8". Tough when you want to rely on the guides in setting up multiple items. Didn't think about testing it by dropping speed and power. Only problem with that is the proportional increase in time. Gotta check to see if I can re-locate the guides tomorrow.

Lee DeRaud
11-09-2009, 12:59 AM
Gotta check to see if I can re-locate the guides tomorrow.Rather than butcher the table (you'd have to cut down the X-axis ruler), it's probably easier and more accurate to just cut a piece of 1/4" MDF exactly 1" wide to use as a spacer.

Dan Hintz
11-09-2009, 7:28 AM
Yeah, just realized I haven't complained about anything going wrong in a while, so it was time ;)

I'm, running at 100S to cause it, as everyone suspected, so I have two choices (neither I'm overly fond of): 1) Slow down, or 2) back off of the origin.

For those that have run across this in the past (or for newbies that will in the future), keep it in mind if you're doing power matrices near the edge of the table... your power selection could be skewed too high if you base your decision upon certain squares.

Scott Shepherd
11-09-2009, 9:05 AM
That's one reason we'll seriously consider a Trotec in the future. It engraves edge to edge, 100% speed. I'm not sure who the idiot, and I use that term on purpose because I think it fits like a glove, was that said "Hey, let's make a machine that has 24" of travel, but then let's limit it's ability to actually work on the entire table to 22".

Whoever made that call should be fired. If you need "extra time" then make your rails longer. But don't you dare say you have a 24" machine.

Lee DeRaud
11-09-2009, 11:33 AM
Whoever made that call should be fired. If you need "extra time" then make your rails longer. But don't you dare say you have a 24" machine.You won't have to. If you provide 1" of overtravel on each end, the marketing department will immediately label it as a 26" machine...and the cycle starts over.

Dan Hintz
11-09-2009, 11:52 AM
I suppose if I wanted to get snippy about it I would agree with you, Steve, but I tried to take a more pragmatic look at things. My car may do 150mph and pull 0.9 Gs on the stickpad... but I never expect it to do both at the same time.

The machine works admirably for most of what I do, even if I wish I was made aware of the limitations ahead of time. When I'm marking a fuel rail, for example, the mark is several inches away from the edge, so I see no issue. If I make a tile that requires edge-to-edge engraving, I can either shift the tile and image or be aware that my settings will need tweaking for a slower run... if I ever find myself doing a lot of that type of adjustment, a different machine may be more appropriate and I'll reevaluate my equipment.

For the time being, I'm quite content with the machine (though I have some real annoyances with the rotary... possibly due to poor operator understanding).

Scott Shepherd
11-09-2009, 12:14 PM
I agree Dan, you learn to work around it. However, when purchasing a 24" travel machine, it would be a VERY safe assumption, in my opinion, that you could do 2 12" tiles at a time for a mural. It would significantly reduce the run time by doing 2 at a time. However, no such luck. Want to do 2 tiles at a time? Won't be doing it on this machine. Had I known that up front, I might have bought a different size machine or even a different brand.

No where in the advertising does it say *24" travel only applies to vector cutting.

Seems like such a waste to make such a solid, well built machine and then hose it up by making the travel run out near the end. Just seems stupid to me and a very poor decision by someone, somewhere.

I don't agree with your car analogy. Tiles are 12" wide. Material is sold in 24" sheets. These are industry standards, not some obscure dimension. That's actually my one issue with Trotec. Their machine is metric, so their table sizes are very odd.

Lee DeRaud
11-09-2009, 12:39 PM
However, when purchasing a 24" travel machine, it would be a VERY safe assumption, in my opinion, that you could do 2 12" tiles at a time for a mural. It would significantly reduce the run time by doing 2 at a time.Only if you think a 7% reduction is "significant". Assuming 1" overtravel is required, a single tile is effectively 14" wide (28" total), and the two-tile set is effectively 26" wide...26/28 = 0.928. That's worst-case: a smaller overtravel yields even less savings.

A better analogy than the car example would be printers: I expect to pay a premium for a printer that does full-page bleeds. (In the case of laser printers, that feature is pretty much unavailable, for technical reasons not unlike the ones constraining your laser.)

Dan Hintz
11-09-2009, 1:12 PM
As I think about this problem, why is the power dropping at the edges? If there is no control of the power based upon speed, then the engraving at the edges should be overblown (same power, slower speed). To me, this says the driver is reducing power as the carriage slows down in an attempt to keep power level versus speed equivalent to what's set by the user, but it's reducing power too quickly.

I guarantee that laser can turn on full blast long before the carriage has enough time to move more than a few mils, and as it was I was only running at 15P. I'm not in front of the machine at the moment, but I'm pretty sure there's at least 1" of runoff space, more than enough to get the laser firing.

I have a new test to run, now... two horizontal raster lines starting right at the edge. Hopefully the machine will try to raster one when going to the left and one only while going to the right. If I can get it to do that, and increasing the power etches properly at the edge, then I would tend to blame a bad driver routine. As it is, if I increase the power the etching near the edge gets cleaner (at the expense of blowing out the etching closer to the middle). But that could also be due to more etching going on during the towards-the-edge stroke, so I need to separate the two.

Lee DeRaud
11-09-2009, 1:26 PM
To me, this says the driver is reducing power as the carriage slows down in an attempt to keep power level versus speed equivalent to what's set by the user, but it's reducing power too quickly.Agreed. But it's a no-win situation for the poor bugger coding the driver: the driver may be doing the "right" thing (in theory), but a lot of materials are nonlinear with respect to changes in power/speed.

Mike Null
11-09-2009, 1:27 PM
I suspect that the power rise is based on time rather than speed. That might explain why slower speeds perform better at the left edge.

Lee DeRaud
11-09-2009, 1:36 PM
I suspect that the power rise is based on time rather than speed. That might explain why slower speeds perform better at the left edge.Same thing, pretty much: either the driver actually knows how fast the carraige is moving at any instant, or it's computing that speed based on nominal acceleration curves for the motion system.

At lower speeds, the distance (or time if you prefer) required to get up to commanded speed is reduced, i.e. less overtravel is required or (equivalently) power compensation is required for a shorter interval. (From Scott's point of view, the bed gets wider as the speed is reduced.:p)

Dan Hintz
11-09-2009, 2:18 PM
Mike,

Since the motors are steppers, the programmer is aware of the instantaneous velocity and acceleration (whether he attempts to use that information or not is unknown), and therefore keeping track of power versus time would not make (useful) sense.




Lee,

Agreed, materials are often non-linear, but if they're going to provide for a power ramp, they should also provide a way to tune that ramp (since we can tune just about everything else on this machine).

I won't go so far as to say "I can't believe they didn't do this!" as I've programmed enough stepper drivers in my day to know they can be pretty nasty in certain cases. What I'm proposing is not a walk in the park, for sure, but it's not rocket science, either. I imagine most materials could be controlled using one of two options, either a linear or a logarithmic change. The linear change would be a simple on/off selection for the user (and be a very easy addition to the laser's power control code). The logarithmic change wouldn't be much more than that, just an additional beta (i.e., rate of rise) parameter that would be tuned in like all of the other settings (and would take maybe an extra hour or two of coding compared to the linear version).

Would the extra work really help that many people? Would it add significantly more confusion/parameters with little return? I suppose that depends upon who you ask. If someone knew this shortcoming beforehand but knew they could tune it out, that may prove useful. ULS already has a beginner versus advanced control panel option (i.e., materials versus manual control), so why not make manual control even better? If the user doesn't turn on the "edge" option, the machine would continue to etch the way it is now. I'm just thinking out loud here...

Paul Brinkmeyer
11-09-2009, 4:45 PM
Dan
what is your issue with the rotary?
Is it postioning in the "Y" by chance?

I do quite a lot of rotary stuff, so if you want let me know and maybe I can help.

AS to the power issue on the edge, maybe that is where ULS measures the tube and give it 60W and you measure in the middle and get that extra. :p

PM if you want.

Mitchell Andrus
11-09-2009, 4:55 PM
You won't have to. If you provide 1" of overtravel on each end, the marketing department will immediately label it as a 26" machine...and the cycle starts over.

That's kinda what LaserPro does. I have a Spirit. In the driver I can select to go into the run-up area to extend the bed from the listed spec width. It's default is the smaller area. When I do this (I almost never raster) the machine adjusts. IIRC they call it "Smart - something or other". I've only needed to get into that extra run-up area a few times and always worked fine.

Dan Hintz
11-09-2009, 7:55 PM
Paul,

ULS measures at the output window of the cartridge. My main frustration with the rotary has always been focus, though I'm positive there are a few shortcuts that I just haven't taken the time to learn yet.

When you scan to the axis of the rotary for focusing, you really want to back it up about 1/2" so the measuring stick can rest on the object. Why can I not adjust X, Y, and Z through the front panel controls? Want to know my (old) stupid, time-consuming way?
1) Set the Z to about where I expect it based upon the object's diameter.
2) Enter X/Y mode, zoom over to where I want to measure (about 1/2" back from the rotary axis, middle of the engraved area), take a Z measurement by eye, then exit the X/Y mode, pushing the carriage back to (0, 0). :(
3) Adjust Z.
4) Goto step 1) until perfect.

Angling the rotary is another chore. For some reason, they put small rubber bumpers/feet on the far end, which means you can't push it up against the rulers... you have to eyeball it from above. Okay, I could take the bumpers off, but they have to be on there for a reason, right? Right? But back to the angling... I would think with the addition of a free-hanging spirit level and visible laser line, I wouldn't have to keep adjusting, testing, and adjusting again. Come to think of it, I may just kludge together something like that and try it out!

Scott Shepherd
11-09-2009, 8:11 PM
Dan, you are wasting your time with the rotary problem. I drew up a detailed drawing explaining the problem about 2 years ago and sent it to ULS. You notice it still hasn't been fixed.

Again, another stupid stupid stupid thing that needs someone with a brain to look at. If you have to focus with the focus tool which works off the front of the lens assembly, then why wouldn't you make the focus point be that point when you plug in the rotary? It focuses with the center of the lens over the center of the part, which means you CANNOT use the focus tool. Apparently no one at ULS has ever actually used the rotary axis in real life, I guess?

Dan Hintz
11-09-2009, 9:02 PM
Steve,

There has to be some way we can get the ear of a ULS rep that can actually get things done. Mike Mackenzie (I believe it was him) said he would try to make that contact for us if we had a list of problems, but either he forgot or the contact never materialized... I had forgotten about that until just now. Between the two of us, I think we could have a serious list of improvements. Why a company would not have an interest in listening to people who have long-term, real-life experience with their product and years of experience designing products for the consumer world is beyond me. But so it is...

Scott Shepherd
11-09-2009, 9:24 PM
It's not a rep problem. I could write a mini series on my experiences in the laser world.....

I wrote a very detailed email report of an account I had with them not too long ago. I waited for several days before I wrote it, as cooler heads prevail. My email was polite and detailed the facts, point by point. I know for a fact the CEO got it. I was told I would be getting a phone call from the CEO personally to discuss it. Needless to say, I didn't wait by the phone. It never rang.

From that email, there were some things that were addressed at the top level and they show signs of improvement, but for the most part, it ain't happening.

ULS went through a major transition not too long ago. Being in manufacturing my entire adult life, I have seen this movie before so I know how it goes. I think they are trying to go from a kick butt small company to a large company and the growing pains are killing them. Their pricing has come down considerably, which means their model is no longer more expensive, low volume, to inexpensive, high volume.

I also know for a fact that some very key people left that organization during the transition. Those losses are apparent to the end users.

When I was at a ULS sponsored advanced training class recently, I asked why the drivers weren't available online. I was told the reason and now it's easy to see why obvious little things never get fixed. I feel for them. My guess is they have a small handful of good people still there that are working their butts off and seeing no results. Been there, done that. It's just the typical stuff manufacturers end up going through in growth mode, when they don't listen to the people that got them there.

Just my opinion.

George Brown
11-09-2009, 10:49 PM
That's one reason we'll seriously consider a Trotec in the future. It engraves edge to edge, 100% speed. I'm not sure who the idiot, and I use that term on purpose because I think it fits like a glove, was that said "Hey, let's make a machine that has 24" of travel, but then let's limit it's ability to actually work on the entire table to 22".

Whoever made that call should be fired. If you need "extra time" then make your rails longer. But don't you dare say you have a 24" machine.

You're absolutely right Scott. I have been working with machines all my life, from computers to software to machines of multiple types for many more moons than I care to admit. It is extremely rare to find something designed well. I my opinion, it is because the ones who design it don't really spend time using what they've designed. It may look good on their cad program or in code on the computer screen, but that is a far cry from actually putting your hands on it and using the darn thing.

I have found two significant flaws in the drivers of the ULS machine I have, one of them a very big problem, but they don't seem to care. Granted there are some nice design features of the ULS machines, but the hardware is only as good as the drivers that run it. Poor drivers can ruin a great piece of hardware really fast.

Mike Mackenzie
11-09-2009, 11:00 PM
Steve,

You hit the nail directly on the head!!

I have tried for years for what I think are simple changes.

We keep trying, we keep sending in reports and explaining what is going on in the field and what is getting asked for.

Every sales meeting the discussion comes up about what we would like to see / have.

Dan if you would like to send me a list of issues, additions, problems, & or changes I would be more than happy to send it along to people who will read them.

I do not want you or anyone to think that I can / get the changes asked for done but I will get it to the management for you.

Scott Shepherd
11-10-2009, 8:57 AM
Here's a simple one Mike. I asked if it would be possible to put a help button on the control panel that linked to a file that told what all the features do. Keeping in mind, my machine is 2 years old, so the driver I started with, isn't the driver we use now. So there are a lot of additional things on the UCP now. However, I have no clue what they do and no way to find out.

Great example- "Frame Rasters". What the heck is that? (I know because I asked here and I think it was you that answered).

So, as a user, how am I supposed to know how or when to use "frame rasters". You might as well not put it on there if you won't tell me what it does or how to use it.

The answer I got was "all changes are in the change log that's included with the driver".

You are kidding me right? Okay, so let's say I just bought a machine today. The driver has "Frame Rasters" in it. Now what? I have no change log. Also, what if I'm the person that installs the drivers and I have other people that actually use the machine. They never see the change log.

Are you serious? With all the skill and technology available today, we can't get a freakin' HELP button installed that tells us what the buttons actually do? I've written many a programs, and I can say with certainty that we're talking about a no-brainer. The instruction manual already exists in digital form. So we're talking about putting a "HELP" button up and linking it to a already existing file, and that's too complicated?

Dan Hintz
11-10-2009, 11:42 AM
I whole-heartedly second what Steve says (there's quite a bit on the driver I don't know what to do with, and even some of the supposedly old stuff isn't in the manual).

But aside from that, Mike, I think I would have a tough time (at this point in the game, anyway) handing my suggestions over to anyone other than the CEO or a programmer. No offense is intended towards you, I'm glad you're around to help people, but handing suggestions to reps has amounted to dumping them straight into a black hole (and through no fault of the rep, I'm sure).

At least as a beta tester for C0rel, I see the results of my suggestions within a few weeks, having the major ones listed as recognized and repaired on their newsletter with the following release. With ULS, where does it go? Does it ever get fixed? Does anyone even look at it? I know some of the things Steve has suggested were great ideas, and I'm pretty confident some of the things I found were useful, but why should I spend time writing up a consistent way to repeat a bug (as I'm told to do by ULS) only to have it summarily ignored? Am I being told to waste my precious time just so I'll feel like I'm helping and leave them alone? Someone send me the email address for the CEO and I'd be happy to put my thoughts in his hands.

Lee DeRaud
11-10-2009, 12:50 PM
Does ULS do its own software development?

If they're outsourcing the drivers etc, it helps explain the long update cycles. (Notice I said "explain", not "justify".)

Scott Shepherd
11-10-2009, 1:52 PM
Does ULS do its own software development?



Yes, I believe they do.