Z Axis Troubles

[camaro_dan67]

Ok here goes. I posted about problems with this before I'll start at the beginning.

The machine is a Kiwa Excel center-4 built in 1984. The control is Fanuc 3m model c. The z servo motor is Fanuc Dc 10m.

I was running a simple program the day this started. Countersinking some parts nothing special. I went to shut the machine off after running a while and I always had to handwheel the z down on a prop board so the head doesn't drop when the machine is off. I was handwheeling the z at a slow increment and it jumped down about a half inch. It got alarms but I don't remember exactly what they were. I reset the alarm and ran the program over. It was ok running in auto. In handwheel it dropped in the z again. I think at this point the watchdog alarm led lit on the motherboard. I turned the machine off and when I went to restart later it never came back on. The fanuc manual says if you get a watchdog alarm to replace the main board. I then bought a used motherboard off of ebay. I transferred everthing in sockets to the ebay board. After putting all the boards back in I re-entered all my parameters. After that the machine started with only a servo error 34 ( z feedback cable is disconnected). I did successfully start the machine up with no error 1 time and homed it. While it was sitting with the spindle running I heard click got the servo 34 alarm again.I then took the z motor off and took it apart to clean it out and found oil from the gearbox below it inside the motor. I cleaned the whole motor with QD electronic cleaner let it dry and put it back together. After putting the motor back on I discovered if I warm the motherboard around the z encoder plug it will come up with no servo 34 error. Now when I try to start the servos the z tries to slam in the minus direction into my prop board before the servos shut off. The z drive then throws an OVL alarm on the front of the drive. I have tried another drive since this started happening with no change. I'm trying to figure out what to do next.

 

[Ianagos]

Wow that’s a rabbit hole of issues.

I’d check your boards for any components that look burnt or damaged. Then also check your z axis cable for any damage at all. I’ve seen wires look fine but be broken Internally due to fatigue. Something else I’ve done in the past is let a machine stay powered up for 30 plus minutes before I power on the controller.

Now it is also a fanuc so I’m sure there are plenty of people here that can help you better than I can. I’m not a big fanuc guy pretty terrible controls in my opinion

 

[Vancbiker]

Did you ever scope the encoder signals? Is there 5V and a good ground at the encoder? Have you metered the encoder cable under a load? Broken wires will sometimes look ok with a multimeter because the current applied is so low but will not carry enough current to actually work.

Having to prop the axis at power off means a bad brake or broken counterweight chain.

The run away into the prop can be a result of a bad encoder, bad cable/connections, bad drive, bad mainboard, or if the motor has a tachogenerator it is bad.

 

[camaro_dan67]

I tried to scope the signals at the plug on the motherboard all I found was power no signals. I think I will try like was said looking for ground and signals at the encoder itself. What signals should be present at the encoder with the control on and the e-stop in?

 

[doug8cat]

I have same sort problem in cinci lancer 2000 with acaramitic 2100 control and Kollmorgan serovs. Z ill drop out and drp 1/2". Spoke with Fives and sre looking to it. Is your problem intermittant? I.e. every 4-5 days?

 

[Vancbiker]

I tried to scope the signals at the plug on the motherboard all I found was power no signals. I think I will try like was said looking for ground and signals at the encoder itself. What signals should be present at the encoder with the control on and the e-stop in?

Fanuc Encoders are a 5V square wave with 50% duty cycle. The frequency will depend on how fast it is turning. No movement, no signal. With the machine in e-stop, turn the motor by hand to see the signals. If the motor is difficult to remove, remove the encoder ( be sure to mark it and the motor shaft so you get it back in the same position) and rotate it independent of the motor.

 

[camaro_dan67]

I got to test the encoder tonight. The model number is A860-0300-T001 2000p.
I took the motor off the machine and took the cap off the encoder. There was 5 volts where the wires connect to the board. Here's the rest.

B- had a + square wave that increased when the motor went faster.

B had a + square wave that increased when the motor went faster.

Z Had a single + pulse per rev.

Z - Had a single - pulse per rev.

A- had a + square wave that increased when the motor went faster.

A had a + square wave that increased when the motor went faster.

I didnt get to test the cable. Would it be best to test the same signals at the other end of the cable and then if that checks out follow those signals as far into the board as I can?

 

[Vancbiker]

Good! Now check for A and B + and - at the mainboard connection.

 

[camaro_dan67]

I have not had a chance to scope the signals at the board, maybe later today. I was wondering if the motor runaway condition could be caused by the encoder being 180 degrees off in relation to the motor. I thought I marked both of them but I may have put it in wrong.

 

[Vancbiker]

180 off on that encoder type won’t cause a runaway. A missing tach signal if that’s how the machine is configured can cause that. If the encoder signal is missing it can cause runaway but usually limited to one rev of the screw if parameters are set correctly. If jumpers on the mainboard are set to use a tach signal but the servo is not equipped, that can cause runaway.

 

[camaro_dan67]

I checked the signals where the M11 plug goes into the motherboard they are the same as at the encoder. I also warmed the board and checked a few of the same signals while the green and red leds were both lit. I think they are labled led 2 and 3. I saw no difference in the signals either way. I suspect there is some communication trace or chip on the board that is not carrying the signal through. I do have another motherboard on the way so I might just send this one back for a refund.

 

[camaro_dan67]

I recieved the other motherboard today. Before I install it does anyone think with the problems I've been having that the motor or encoder could be messing up the communications chips on the board? I would hate to install a working board only to have the other things connected to it make it junk.