[Montabelli]

I am in the process of re outfitting an old motion master cnc to run with vxd xe 160v drives and a vesdepi interface card. I have an old e stop switch which is normally closed. There is a diagram in the manuals to wire a normally open switch but it uses the external relay buss as well, is there a way to wire the normally closed switch and not use the relay position, or will the external relay still work?

I am just about to the point of testing the fanuc motors in the machine. I am starting with mach3 as the controller, I am wondering if it would be best to also get a smooth stepper board. to insure the most consistent timing for the operation of the unit. ultimately it will be a 4 axis unit with the 4th axis lying along the x axis in the configuration of a lathe.

[Xerxes]

You can wire a relay normally regardless of E-stop switch connecting to same terminal.

You may well try without smoothstepper and get it later if you need more speed.

[Montabelli]

I have been sorting out the wiring and all axes work, I am getting motion errors on the x or table axis, mostly, some on the others. It is my guess it weighs about 350#. I have been able to reduce that some with the acceleration rate, currently set at 150mm/sec/sec. Even in jog mode when stopping the move it stops abruptly and and gives me a flashing red, motion error. I am running fanuc model 5 motors, 110 v 5.9nm 6.8 amp peak. I have a 1.5kw power supply, powering 3 motors and 1 extra axis driver card, have to put a new encoder on the 4th axis motor. I hooked up the regenerative resistor from the original controller, 15ohm that helped some, should I perhaps increase the setting for over voltage fault? if it was an encoder, I would be more inclined to think it would not be intermittent at all. also it would not be all three of the operating axes. Is it possible I need more power?
Thanks.

[Xerxes]

Please email me a saved drive configuration file. It may reveal the source of problem.

[Montabelli]

Xerxes, thanks for the quick reply, I think my problem is a hardware issue, I got what seemed like a half pound of plastic shavings out of the ball nut when I disassembled it, once reassembled I tested the motor again and it read 1 encoder pulse over 4 poles, that's just off by 4999, I expect a wiring problem is the issue and could explain the intermittent nature of the problem.

[Xerxes]

If only 1 encoder count is "seen" by drive it is likely that one channel of encoder is not passing thru to drive (probably wiring error).

You can test encoder by looking at GDtool Device status page while turning motor by hand. Encoder counter displays the encoder angle that drive sees.

[Montabelli]

Well I cleaned the contacts in the cannon connector and have had better readings from the encoder it tests ok but i have motion problems with that motor, I suspect the encoder still.

I made a little g code file that only involved moving that motor back and forth at increasing speeds. as the motor increases in speed it starts to make jumps clunks or what ever one might call it at regular intervals. the higher the rpm the bigger and seemingly more sporadic the clunks. it is not driving a load, so there should be no mechanical issues.

i stopped the program at about 900 rpm, not even half speed. the encoder sees normal on visual inspection, I have another used motor on the way, and have an alpha series fanuc that I am adapting a new encoder to. once I have either of these motors available I will repeat the test and see if I get smooth motion.

with no load on the motor I had no following errors, I did zero out the motion fault limit parameter, but still get occasional motion fault errors on the other axies. what kind of issues might be causing the flashing red led after zeroing it out?

[Xerxes]

It sounds like there is bad quality of signal somewhere (probably encoder) causing the jumps. You can try to move motor in GDtool by entering "moveinc somevalue" in Event log. I.e. moveinc 10000. If its still jumpy, then it at least can't be caused by step/dir source.

Also if you can cause the fault while using GDtool, you can see drive status bits for better analysis of fault. There may be some other faults in addition to follow error which prevent clearing faults (some faults are non-clearable, like the ones caused by large fluctuations in 12V logic supply voltage).

[Montabelli]

when we did the moveinc from the shell we noticed that the rotations were inconsistent, so I finished adapting a fanuc alpha series motor with an encoder, and it worked smoothly.

[Montabelli]

I just thought I would give an update as to my progress. I have replaced two of the fanuc encoders and their wiring. the encoders were both problematic and their wiring harnesses were suspect as well. when it came to moving the motors , even unattached I was still having difficulties. Any rapids were filled with extremely harsh interrupts. I decided the problem was probably with either the parallel port or more likely with windows. one of the guys on the artsoft page suggested I try electronic gearing, I cut the resolution by a factor of 5, lowered the kernel speed in mach3 and now, no more kachunks. I have a smooth stepper coming, in two weeks, then I will be able to regain my lost resolution. I did optimize windows prior to my expiroments, but I have had more than my share of windows generated faults over the years, so I think it is wiser to remove the counting from the province of windows and play to it's strengths, timing with out interrupts not being one of them.

I still have a long way to go on this protect, I have to figure out how to get the complete interface hooked up between mach 3 and the vesdepi card, first, so I can reset the drives from mach 3 and then all the limit switches and home switches etc then I have to get my spindle mounted and fly cut the table. then I can finally start to learn how to get the optimum performance from it.

Im having fun... is that allowed?

[Xerxes]

Thanks for update. I hope your performance expectations are exceeded once you get machine up and running. Remember to have fun :-)

[Montabelli]

It has been a while.. couple weeks for the smooth stepper to arrive which it did as I left town for a couple weeks then a week to get every thing settled at work and the changes done in the machine. the machine is back to full resolution and the motion is smooth as silk.

I also tried the tempest planner (developmental from Art ) and ran some tool paths in the air with very smooth motion.

I am still getting drive faults if I break into a tool path move which is not a big deal so long as I can reset the drive . Is there a way to have the reset button in mach3 reset a faulted drive? or must I cycle the power to the vsd drives? I am using a switching power supply for the logic in my machine so I can put a remote off on switch on the front of the panel to cycle power when faults occur, I'm just concerned that I am missing something on the setup. is there a more complete explanation of how the functions of the vsdepi card are set through the config.drc file?

[Xerxes]

I believe you can quite easily tune drive not to fault in the first place.

However, when VSDEPI & Mach3 are configured according to VSDEPI manual, un-toggling the blinking e-stop button in Mach will clear also drive faults.

[Montabelli]

OK, so I'm an idiot... I went back over my original set up and found my logic fault. it was the e-stop set up. every thing is behaving properly now. I knew it would be something simple, that at one time I had misconstrued.

I still have a few things to do to get this thing ready to go into production. spindle, lube system, limits and home switches. once I get all that arranged, I'll shoot some video of the machine under power. thanks for all your help.

[Montabelli]

[nomedia="http://www.youtube.com/watch?v=sxx51KNqxgc"]YouTube- VID 00000-20100409-1054.3GP[/nomedia]

here is a preliminary video of the motion master, I have it running pretty slow, and the flashing light in the cabinet is the driver for the 4th axis which I have to build yet, I still need to find a way to reduce jerk in the motion, and still be able to run smoothly through a lot of short segment code in 4 axis motion but two d works great.

[Xerxes]

Nice video! Thank you.

How fast are the rapids?

[Montabelli]

I think the rapids are set at about 8000mm/min, I haven't really pushed it too hard yet I was cutting the pocket in the video at 2500mm/min, I have cut air at 4000mm/min in 2d tool paths but I am going to need fairly fast machining of 3d and 4 axis production cutting pretty soon.

I have ordered an analog output interface that I am going to implement because I need better smoothing than is currently available in 4x in Mach3. and I haven't been able to come up with a way to make it happen with mastercam and helical tool paths. I have worked with techn isel controllers for a long time and I have also worked with the old DOS version of wincnc either on costs me about the same money, but I have my operator set with the techno and this way I have only one interface on all my machines except my wood lathe. I will probably move mach3 to that

[Montabelli]

Xerxes,
Can I run an analog interface through the vsdepi card ? it will already provide a single source for relays, enable switching etc, I would assume that it should work easily enough to wire the analog control through pins 15 & 16. What do you think?

[Xerxes]

Xerxes,
Can I run an analog interface through the vsdepi card ? it will already provide a single source for relays, enable switching etc, I would assume that it should work easily enough to wire the analog control through pins 15 & 16. What do you think?

You probably need to forget VSDEPI for analog control and also configure I/O pins of drive differently from GDtool.

This is because VSDEPI uses analog input pin as drive disable/enable control.

[Montabelli]

[nomedia="http://www.youtube.com/watch?v=trSPcHrWbRA"]YouTube- motionmmach3tempst.3GP[/nomedia]

this is a video of some actual production of parts, probably too long, running motion master at 3.5 m/min rapids 8m/min 15 deg ramp entrance pcd tool 18000 rpm cutting 3/4" mdf using mach3 tempest planner, smooth stepper and of course vsd xe drives running fanuc ac servos with encoder count only (no hall effect), 1k steps /mm resolution.
Xerxes, your products are amazing!

Thanks