Topics

1

General Mach Discussion / head stopping up

« on: February 09, 2012, 04:57:09 PM »

we have a recurring problem with our waterjet, we are sure it's not anything to do with Mach3, but this is about the only CNC forum I have really gotten any help from, and I'm hoping some of you have some suggestions or even maybe have had something like this happen before

the problem is that the nozzle frequently clogs when we're cutting ring shaped parts out of "haysite?" fiberglass (not sure how it's spelled)

it RARELY clogs on anything else we cut, but it's not a matter of the head getting too close, since it was about a 1/4" away from the highest point of the material on the last job, and still clogged more often than it ever has, even on this haysite stuff

the only guesses I haven't been able to disprove completely are...
1) the abrasive is getting wet and clumping, or otherwise getting some larger debris, when we put it into the hopper
    but...we've devoted 2-3x the attention to that step when on this job, as compared to normal

2) the haysite gets inside the head somehow, when the water stream stops, since the stuff does tend to bubble up, as the dust forms bubbles, sometimes up around the nozzle
    but...I distinctly remember at least a couple times watching it as it stopped, no bubbles even close to it, and it still clogged

3) the nozzle itself siezes up due to heat, and the orifice closes
    but...wouldn't heat make it EXPAND, not contract?


I'm relatively sure yall are going to want our .xml file, and maybe a code sample, so I'll do what I can to get them posted, but not in a position to do so at the moment...hopefully tonight if I can remember

2

General Mach Discussion / Feed Hold bugs

« on: January 09, 2012, 11:43:52 AM »

I've been trying to research Mach3 revision history/bugfixes, but the only thing I found so far was a large text file I saw as painful to search through...can anyone direct me to any bookmarks that may help? or, better yet, maybe someone has had this happen before, and can tell me what's going on?

Our problem with FeedHold pops its ugly head in one of two cases...1) any time we pause on an arc, and 2) any time we try to do anything in between pausing and resuming operation

1) it kind of seems to forget it was on a G2 or G3 line, and instead of resuming as it should, it goes straight to the end of the arc as if it defaults back to G1...this also applies if we pause in between cuts (on a G00, when the head is turned off) and instead of moving quickly, it moves at the current FeedRate

now that I post it, I wonder if it would go straight to the end of a second arc, since our CAM software doesn't include G codes on many lines...something for me to try, I guess (would it be important?)

2) we tend to get very erratic results when we try to shift the origin while its paused (namely, shifting upwards so it doesn't hit the material)

3

General Mach Discussion / problem immediately after REF ALL HOME

« on: December 04, 2011, 05:57:31 PM »

Someone at my job recently felt the need to reinstall Mach3, trying to fix a very minor problem, and now I wish I'd have put up a much greater resistance to doing so. In any case, what's done is done, and it's my job to fix it.

There are a few terms that I use here, that I'm not sure I'm using correctly, and can't seem to find defined elsewhere. I'm going to define them so hopefully using terms incorrectly wont be a problem, as well as making correction clearer.
--machine home - where all 3 axes are 0 if you turn on machine coordinates
--DRO - coordinate numbers (can be either machine coords or work offset coords?)

I know that in Config>homing/limits, I can modify the distances between machine home and the point it moves to after referencing, but I don't know how to modify what machine home should be in relation to the home switches. This is essentially all I *think* I need to know, but in case it's important, I'll detail the rest of the situation better below.



Normally, when we hit REF ALL HOME, the machine finds its home switches, and proceeds to move .125 in the positive direction on the X and Y axes, and doesn't move the Z axis at all.

Now, however, not only does it move the X and Y more than an inch, it drops the Z by about 6 inches, I think it was!!! It started out at just 2 inches, but has somehow decided it needs to drop it further, and ended up at 6 by the time we gave up trying to fix it last night. Another half an inch, and even without the focusing tube on, it will ram the head into the table! I think my heart still skips a beat every time! I'm very glad we had the sense to remove the focusing tube early on, or it would have either snapped or rammed down HARD into the table.

Not entirely sure on all the specs of our machine, nor of much of the terminology, but I do know we have servos, not steppers. Also, fairly sure the X and Y are on rack and pinions, while the Z is on a screw.

Also, our home switches double as our limit switches. I'm sure this is not a particularly good idea, but it hasn't been much of a problem before, and I don't think it has anything to do with the problem I'm posting about. Figured better safe than sorry though, so I included it.

Normal operation still seems fine, after referencing.

4

Mach3 and G-Rex / Feed hold acts quirky

« on: March 24, 2010, 07:57:59 PM »

I can see from a short search that there are quite a few known problems with Feed Hold, with most but not all relating to subroutines.

Now, I don't know what a subroutine is, and so, can't say for certain we don't use them, but, having said that, we have random and varying problems with the feature. I would submit a code sample, but I have yet to reproduce any of the instances we've had, despite several attempts. I can, however, describe what happened, and detail a bit about our setup.

We have a waterjet from EZ Router, a pretty standard PC, I think it's a Dell with XP Home? If more specific details are required, I can get back to you on that.

In the first case, we were cutting out a the shape of Texas, and used Feed Hold as it got to the bottom tip, if I remember correctly, and left it, assuming like an idiot that it would work right (haha), but when I came back, I found that while the part was fine, it cut an addiitonal line from the tip, grazing and passing right on by the tip to the northwest, continuing on for a few inches.

In the second case, we were cutting a horseshoe, and hit Feed Hold about half way through the run, and when we restarted it, the head rammed down into the part and through it, breaking the stone in half (thankfully it was positioned in between the grating!)

In both cases, we paused it either to change the speed or raise the head/re-zero the Z axis, to adjust cut height.