[swenhold4]

The beacons are lit! Gondor calls for aid!!

I'm nearly about ready to admit defeat. OR at least that buying this machine was a mistake.

For the past couple years off and on I've been trying to work with a 1983 Traub TND360 that I believe was actually upgraded with a 1985/6 control. I bought it with the hopes that I could work it into production and start my own business venture from the ground up. In fact, my employer has been kind enough to let me store/work on it in the shop I run with the agreement that if/when I get it going I can use it to provide parts/service on equipment here.

The good news is I've been able to get it from a DOA state, to being ready to turn the drives on. I've needed a few new push buttons, a varistor replaced on the main power supply, a replacement operator panel, a new LED screen retrofit, a replacement servo drive power supply, a new timing belt on the x axis, and a couple broken wire repairs.
Even with all that, plus scooping up NOS cards off ebay, I'm still not into it as much as I'd have been for an operating machine.

The bad news is now I'm really hitting my head against the wall and am not finding any help anywhere. I think my contact at Index reached his patience limit because he finally just went through and sent me "everything [he] has about the TX-8". Mitsubishi's level of assistance was to recommend that I go onto their knowledge base at MEAU.com and register so I can find the manuals...that I had been working from for the past 6 months -_-
IF the TRA30/60 servo drives are in velocity loop I can dial in the velocity loop gain (VR1) and VR7 to the point where there's no motor rotation, verified by eye AND the resolver count on the control readout.
In position loop though, VR2 "has been adjusted at the factory. Do not touch this setting." -_- thanks Meldas, big help. Other settings are very obvious to not have changed as they have some resin still on the pot. Anyway as soon as I turn the drives on the motors are powered to what appears to be rapid speed and the control immediately faults out with error# 022 or 023 (whichever comes first). This is of course, a droop error, that if my german is worth anything equates to excessive following error caused by either a mechanical jam, a drive fault, a motor fault, or a control fault. Again, big help in the manual.. IF the motor leads are not connected to the drives everything is fine (DUH). I did manage to find a really kickin deal on a TRA31 drive, so I tried that out and the only result there is now I can't lockout the position loop to setup the speed loop -_- That, plus again this being on BOTH axes, causes me to believe the drives are probably fine enough that they are not causing this particular problem.

I can upload the pages from the manuals I'm working from when I get back to work on Monday, but for now I was just hoping if someone could verify or correct my logic going forward.
In the TRA30/60 & TRA31/41/61 maintenance manuals there is direction on which check pins are for what signals. They even provide neat little diagrams of the desired waveforms, but with little exception though they fail to say exactly what the signal voltage/amplitude should be and for what duration. I'm sure the techs from the 80's could just know, but I wasn't even born then so there's so many things that this "could" be that I'm really unsure as to whether I'm picking up a good signal, or a weak signal, or a plain old junk signal.
I think it's pretty obvious that the motors themselves are okay, because they can work in either direction at any speed when operated under pure velocity loop. Of course the shortfall here is there's no stopping it from running the carriages right off the ballscrews.
In the Mitsubishi L1/M0 manuals which very clearly operates under similar control principles there's even a guide on how to adjust the drive amps. Unfortunately they seem to rely on the machine being operational because the procedure specifies setting a certain jog rate for the appropriate axis...something I wish I could do at the moment.

I'm very nervous the control card itself is bad, but I suppose it's not impossible for both pickup units to be faulty?
So I put my scope on the test pins and believe I found some signals, but again I'm not comfortable enough to say any of them are "okay" or even if I'm 100% confident in my o-scope skills.
I was able to find a 4.7/8/9 V signal off of TCP1 for both drives, the velocity input to the drive from the NC. good?
The resolver excitation circuitry is there on both sin/cos waves for both axes (doesn't look like a sin or cos wave, but more like a sloped square wave @ 2.5ish Vpk-pk, and I see a resolver output that's a definite wave signal at around 1.5Vpk-pk.
What's throwing me is the tacho feedback signal. The books say the Sanyo RST series is supposed to send 2V per 1000 rpm. Who knows what max rpm is though? There doesn't seem to be ANY technical info purely about these tachs that's survived. In any case, my scope's automatic signal finder comes back with so much noise on TCP2 that I don't know what's a legitimate signal and what's not. If I raise the trigger up past 1.75V I do end up getting a 2V +/- 200mV stable signal that looks like a dying wave form and what causes me to question the tach's is that this is whether or not the motors are spinning. But how can a tach that's not spinning generate voltage?

The triangulation wave form is okay.
Continuity along the cabling for the tach's is okay.
Continuity for the cabling between ZP1 on the drives and the CNAx/J4 connections on the control are all okay.
Parameters are freshly loaded from the tape reader and all of that appears to be okay.

At its core is something is telling the motors to rotate very fast right away, but the control is able to both see that and say hey no that's wrong, fault out the error code, and shut down the drives.

I'm to the point where I'd be willing to hire an expert, as long as I could trust they knew what they were doing, because so far I've been able to reverse engineer farther than any of the tech help can at both Index & Traub. I'd also need to have funds to be able to do that though...so here we are.
Bonus prize is that whoever can help with this also might like to help me learn how to megohm a motor so I can verify the spindle motor is still good

In all seriousness though, I've put a lot of time an effort into this and I'd like to give it my best shot before I part it out/scrap or consider retrofitting a newer control so whoever can help, I'd very much appreciate it. This entire venture may be relying on whoever you are.
Help me Obiwan Kenobi, you're our only hope!

[swenhold4]

OK.
After a couple days of reviewing and reviewing and etc etc after that very cathartic beginning post ..
It seems the first thing I need to look at is verifying the control voltage coming from the control (DUH!! That's embarrassing).
Attached is the best I could find of the circuitry principles and my interpretation of them.

Does anyone know what the UL63 (ie TL-1 ie Traub version of Mitsubishi L1 ie TX8) card # LX61 should be sending to the ZP-1 connector for analog voltage at rest? Right now of particular interest would be Z, but I'd guess its the same either way.
I get that it's all basic arrithmetic: Output C = input R minus actual position X, but what is input R officially supposed to be? The manual speed input is a 24V scale, with rest being 0V. So does that mean the control should be outputting 0v control voltage at rest? And if that's the case, then 0 - 0 should be 0, ie shouldn't the tach output also be 0?
The tach output does appear to be 0V at rest when measured with a multimeter, but then I can also find a 5v signal with the scope. Could something be backfeeding somewhere? A ground issue perhaps?

I don't know.. I guess for starters, again, does anyone know what the Analog Control Voltage going to the servo should be AT REST?
Thanks

[swenhold4]

Ok so something finally makes sense:
Both axes are being fed from J4 (Pin 1 for X & Pin 11 for Z) what appears to be rapid motion voltage at rest.
In the pictures Channel 1 (yellow) is from the back of the L61 board, and then channel 2 (green) is from check point 2 of the TRA board.
I'm 99% certain I've been able to determine that the tach's are not providing extraneous voltage either.

Now the question is why is the NC telling the motors to go somewhere very quickly....

[swenhold4]

I was able to find in the L1 manual the confirmation that +/- 10V is in fact the speed control voltage scale.

Now on to reverse engineering the LX61 control card..

[swenhold4]

RESOLVED:
The problem ended up being in the LX-61B main control card. I reverse engineered probably 30% of the logic circuit for command voltage and it dawned on me that I had a short between 12V positive & negative. After ruling out most of the components by simple reasoning, I was able to determine that one of the 2 main capacitors was internally shorted with no external sign of failure. My reasoning was that all of the downline components were simply reacting to the voltage they were seeing, and sure enough it ended up being a shorted tantalum capacitor, #C58.
I'm going to hang on to my dwg file of what I'd worked out so far for a circuit diagram, but for now this issue is closed.

[andrzej1964]

I have a problem with the Traub TND360 machine, it has lost its parameters. The Traub System TX8 Control Unit Meldas UL63 lathe has a black screen after the battery runs out. Do you have a system recovery procedure? Thank you in advance for your help.

[swenhold4]

I have a problem with the Traub TND360 machine, it has lost its parameters. The Traub System TX8 Control Unit Meldas UL63 lathe has a black screen after the battery runs out. Do you have a system recovery procedure? Thank you in advance for your help.

I'm sorry no, I don't have a recovery procedure. All I have is the process to reload parameters from a punched tape roll.
If all you have is a black screen I can tell you this, you have more than a parameters issue. Right now I'm researching some memory errors that I am getting in the check menu page, and I tried swapping the LX64C card with a couple I picked up on eBay and neither of them let the controller boot up.
I know it was trying to, because my screen didn't go blue complaining of a lack of output signal. All they gave was a blank screen and none of the operators panel buttons would work. So I put the original LX64C card back in the UL63 and it booted up fine enough. Long story short you could have a problem in your controller memory is what I'm saying, beyond what I could help with.
OR your screen is junk. I replaced the old CRT tube that had such corroded boards and terminals that it was much easier to fit an LCD screen from a guy online who specifically deals with that sort of thing.
Sorry I couldn't be more of a help but that's what I can tell you. You're going to have to dig into one of those routes because the control should load up even with a parameter loss. It just won't be properly programmed with the correct information.