Hi,

The waves are a well known problem. It's just a lack of heat. The wire no longer cuts by radiance, but by contact which generates vibrations and hence the waves.

Gilles

Originally Posted by Guy97

Thanks Gilles, I will try to keep it simple, and as chronological as I can recall. Please understand that I am very new to GMFC and to CNC work, and my mate with the setup knows his system as far as using it to cut, but his design or programming experience is also limited.

I can add the .cnc files, but my logical analysis of the events suggest to me a mis-use of software that may have caused a failure in the controllers or the hardware itself. So, let me explain as clearly as I can, the events of a few days ago.

Some three weeks prior I visited my mate for the first time, and we cut a number of sets of cores for the same aircraft. We always used 25mm wide foam blocks at the extremes, to validate the cuts, before committing to the full blocks. All of these cuts went off without incident. Before I left I suggested my mate have a look for any updates, as his registered version of GMFC Pro was around 2007.

When I arrived at my mate's place a few days ago, he informed me he had updated his software. What we didn't realise is that he had downloaded and run the updated installer for GMFC Expert. I opened the program (Expert) from the desktop shortcut, and loaded the project file. There were some obvious differences from the program we had used a few weeks prior, but we both put it down to the "update". Here's where the fun started.

I opened the "cut" menu, and attempted to do a "test" - the wire rose about 80mm on the Y-axis (both 'vertical' sides moved together) and then it began to move along the X-axis (both again). When it got to about 400 or 450mm along the X-axis, the motors "locked-up", began making a loudish buzzing noise, and would not move.

We shut down the power to the wire and motors, then had to kill Expert.

After turning on the wire/motor power supply a couple more times, with the same noise, we discovered that it was Expert, and not Pro that we had used.

I uninstalled Expert, and Pro, and then reinstalled Pro from my mate's original USB stick.

We ran Pro, turned on the wire/motor power, and the motors became responsive again. This was great, but there was a problem lurking.

I still cannot really explain this, but whereas the test cuts we then did looked fine (using the 25mm wide blocks), when we ran the same cut on the full block, weird things started happening. Please now refer to the photos below for some explanation of the weird stuff...



Anyway, after shutting the PC down a couple of times, and after leaving the wire/motor box stay off for a good 30mins, the test cuts actually looked fine, and I tried a full block. That cut seemed to go fine, so I did all the rest, some 18 cuts - and they all looked fine.

Closer inspection after-the-fact reveals that there was still some mis-alignment, causing the LE to be too "pointy", but it is harder to see.

So there it is - I do not know whether there is some sort of 'glitch' going on, or whether a controller has been compromised, or if a stepper or the mechanism it drives has been 'damaged'.

Is there anyone who has seen or heard of this type of error, or thinks they know what's going on here ???

It goes without saying, but I obviously appreciate you reading all this, and look forward to some clarity.

Thanks again, Steve.

You may want to try with the older version.

Originally Posted by gm.cnc

Hi,

The waves are a well known problem. It's just a lack of heat. The wire no longer cuts by radiance, but by contact which generates vibrations and hence the waves.

Gilles

Gilles, that's why I showed the photo overlaying the two cuts. I find it hard to believe that vibrations could generate waves in exactly the same place all the way along the chord - twice in a row.

Have you increased wire tension and/or wire heat?

Worth a try.

Dan

Originally Posted by danielphantom

Have you increased wire tension and/or wire heat?

Worth a try.

Dan

Thanks Dan, yes, we actually increased the tension between the two wavy cuts - we were trying to establish the cause by changing variables. As you see, the cuts were identical. The heat is fine - there is almost no weight on the foam and it doesn't move.

I really think the problem is related to the "lock up" and noise when the wire was run by Expert. I'm just not sure which components were effected - the steppers, the controllers, and/or the worm drive setup.

Originally Posted by Guy97

Gilles, that's why I showed the photo overlaying the two cuts. I find it hard to believe that vibrations could generate waves in exactly the same place all the way along the chord - twice in a row.

In fact that's possible, if the lack of heat (or too much speed) occurs at the same point.

Checking that is easy, just increase the heat a bit.

Gilles

Originally Posted by Guy97

Thanks Dan, yes, we actually increased the tension between the two wavy cuts - we were trying to establish the cause by changing variables. As you see, the cuts were identical. The heat is fine - there is almost no weight on the foam and it doesn't move.

I really think the problem is related to the "lock up" and noise when the wire was run by Expert. I'm just not sure which components were effected - the steppers, the controllers, and/or the worm drive setup.

- The PRO and EXPERT rely on the same code for moving motors. If you were using the same table configuration, this can't be a software issue.

- If you want to be sure that the controller is not faulty, do displacements under "zero axis" and measure the result to be sure that no step is lost.

- Increasing the wire tension does not solve the issue, if you start cutting by contact, that's wrong. Increasing the wire tension may reduce the waves, but you will always drag the foam which alters the trajectory.

Gilles

My sugestion is to check the power suply for wire heating, if drops the voltage regarding the owerheating. I have 24V and 8A swither for that, just toughts.
I dont think the tension is the problem, I don't have anny tension on wire.
Second, the bad connection power wire to heating wire can also be a problem. I don't have anny clamps or similar for that. Just pill of abouth 2cm of the wire and wrap it. When problems ocure cut it of and make new.

Originally Posted by gm.cnc

- The PRO and EXPERT rely on the same code for moving motors. If you were using the same table configuration, this can't be a software issue.

I think that's exactly where the problem started Gilles. The PRO was all set up for the table and because we didn't cotton on to the fact that it was Expert that we were running, we didn't even check the table setting. It could have been set, or defaulted, to anything. And there could be any number of other settings that we assumed were ok, that could have caused the motors to stall and buzz. In the end, it's just a question of what damage was done.

Originally Posted by gm.cnc

-

- If you want to be sure that the controller is not faulty, do displacements under "zero axis" and measure the result to be sure that no step is lost.

Will do, thanks for the tip. It may take a few days to organise to go up to my mate's place again, but I will definitely let you guys know how it goes. Thanks for all the input. Steve.

Originally Posted by Guy97

I think that's exactly where the problem started Gilles. The PRO was all set up for the table and because we didn't cotton on to the fact that it was Expert that we were running, we didn't even check the table setting. It could have been set, or defaulted, to anything. And there could be any number of other settings that we assumed were ok, that could have caused the motors to stall and buzz. In the end, it's just a question of what damage was done.

If you run the Expert after the PRO, the table configuration is translated. I am using both for development purpose. I don't this the problem is there.

I any case, check the table configuration.

Gilles

Another thought that might not solve the problem but.....
I would degrease those leadscrews and clean them up.
Then use them dry or really investigate what grease to use, depending of material of nut/screw (there is a science behind that too).
All that dirt and maybe "dried up" grease and some small misalignment combined might make the motors miss steps.
If that would happen every half turn or something, stuff might be getting wavy too.
But i dont think one would be so "lucky" that it happens on 2 axis at the same time to make THOSE waves thas is same left and right side.
Still. those steppers might have a touch time at the moment.

/Ulf

Originally Posted by methuso

Another thought that might not solve the problem but.....
I would degrease those leadscrews and clean them up.
Then use them dry or really investigate what grease to use, depending of material of nut/screw (there is a science behind that too).
All that dirt and maybe "dried up" grease and some small misalignment combined might make the motors miss steps.
If that would happen every half turn or something, stuff might be getting wavy too.
But i dont think one would be so "lucky" that it happens on 2 axis at the same time to make THOSE waves thas is same left and right side.
Still. those steppers might have a touch time at the moment.

/Ulf

That's right Ulf, and I've got to say one of the most puzzling things was the early start of the underside cut on the second run. The profile clearly starts well before re-entering the foam block, yet the waves were in the exact same location - refer to the 9th photo in Post #1710. Thanks for the reply, Steve.

Originally Posted by Guy97

That's right Ulf, and I've got to say one of the most puzzling things was the early start of the underside cut on the second run. The profile clearly starts well before re-entering the foam block, yet the waves were in the exact same location - refer to the 9th photo in Post #1710. Thanks for the reply, Steve.

BTW, I just saw that you are using version 3.0.83 which is way too old. Please update to the last one.

Gilles

One more thing to check. Look and see if the lash (slop) in the drive screw vs drive nuts has increased for some reason. I had one go bad and it did funny things.....