AC spindle motor / drive problem

[Andy Fritz]

Hi guys,

I have a lathe with a 15kw / 18.5 kw Fanuc AC spindle motor without nameplate from -85.
The AC spindle drive is Fanuc A06B-6044-H742. This drive also powers a Fanuc Spindle motor with 2.2 kw / 3.7 kw power which drives the live tool spindle.

There are a few different problems which may be related as they were discovered in the same week.

1. With main spindle engaged in high gear and 30 rpm the spindle motor gives one high pitched noise per revolution. Torque should not be fantastic at 30 rpm on high gear but I can stop the spindle with my hand. At higher rpm the program stops with ALARM 2 on spindle drive after 1-3 minutes.

There is no noise turning the spindle by hand. If belts are disconnected the motor shaft spins smooth by hand with no noise.
With spindle disconnected I figured out that 1170 rpm in high gear from MDI gives 1508 rpm on the motor shaft. The motor have the same little high pitched noise for every revolution but it is or course hard to hear at this rpm. Other than that the motor is running smooth without alarms. During this test I noted the following:
Frequency is 1400- 1600 hz (fluctuating on cheap DMM)
Voltage is about 125 between all legs but current is not even: U=17 amp, V=6 amp and W=21 amp. Is this normal or should current be the same for all legs?

Motor resistance between windings is 0.1 ohm including motor wires between all legs. Megger tests are very strange. Only motor is U - G = 300 meg, V - G 276 meg and W-G 388 meg ohm. Motor wires disconnected from everything is 700 - 1000 meg ohm. When testing wire and motor but disconnected from drive I get around 120 megohm on all legs. I should ad that we have 32 degree C and 85 - 90 RH here and we are half a click from the ocean so the air is damp and salty. With megger ground clip on frame and the positive test lead in the air 1/2" from the frame I get 400 - 500 megohm. Anyway the meg readings are about the same as they were a year ago.

I don't have a nameplate on the motor, but found a picture of one on the net for a Model 15 motor which has the same power rating.
Volt 200 3 ph, 4 pole induction. RPM 1500 / 4500, amp min. 60/70, 15 kw /18.5 kw. As I need to figure out if I have a motor or drive problem (or both) I hooked the motor to the incoming 220 VAC 3 phase at 50 hz for 1 minutes and got about 1570 rpm and a steady U - 103, V - 105 and W - 106 amps.
I dont know how to interpret the Fanuc nameplate though. Some say amps are during no load, others say full load. I tried to find out if JEC 37 differs from NEMA but got nowhere. Does anyone know if the current draw is abnormally high or if it is ok?
Maybe the nameplate amps are not applicable for running at 50 hz from wall power?

2. When the spindle drive is running the smaller live tool motor everything is ok up to 1400 rpm. Over this an oscillation kicks in which pulls down the rpm slightly in a pulse of one per every 6 revolution of the spindle. There is no high / low gear on this spindle. The oscillation is pushing the Fanuc 10T load meter to 180% and after 1 minute I get the Alarm 2 on the spindle drive.
I get 60 meg ohm on all legs including wires to the live tool spindle motor. I have not checked this before so don't know if it is lower than before.
The winding to winding resistance is 1.4 ohm between all windings.

Again I just want to know which parts to focus on or send away for repair.

Since Alarm 2 is speed deviation and I get the same alarm for main and live tool spindle does anyone know if there is one common feed back circuit on the drive which handles both? If so I wonder if the feed back and speed deviation threshold can be adjusted individually for each encoder?
I should ad that each spindles have position coders hooked to the spindle drive.

3. Lastly the spindle orientation before engaging the C-axis rack is sometimes not working. It is much slower and the spindle hunts around before deciding where to stop. This never happened before.

Any help is greatly appreciated!

Andy

Similar Threads:

• Need Help!- 320X motor voltage problem, motor is unpowered by drive
• Which motor and drive for g0704 spindle motor control
• Fuji FSD-22DN spindle drive dc motor problem
• spindle motor/drive
• Need Help!- Fanuc O-Tc AC Spindle Motor Drive Problem

[tc429]

Just a guess, but:
I'd pop the fan/mag sensor cover and feel if the bearing is shot...the mag sensor is sensitive to gap, especially as they age and the magnets get weaker...if the rear bearing fails, you might feel the shaft wobble(loosen belt so shaft isnt bound) and if it does, the gap change will really screw with the motor...Ive seen chips stuck in the belt cause the shaft wobble every rev of the belt as the chip causes belt tension to vary... if the rear bearing fails, rust from fretting corrosion or even metal bits from the bearings can get caught in the teeth of the sensor ring, but usually youll get a klunk every rev.

if youve never changed a rear bearing before, a note of caution- dont pull the sensor ring off with a puller- its dead soft and will be ruined/stretched oversize... I just pull the sensor itself out of the way, protect the wires somehow from heat, carefully heat the areas between the holes on the ring with a torch, it will pull right off with just twisting/using a glove... if yours has the double sensor(outer one only has one groove for 1 rev pulse), mark the single groove thing relative the shaft so you can put it back in alignment, the 256 tooth ring wont matter, its just pulses for speed/direction... Ive changed rear bearings in the machine before, pretty easy if the fan end is out in the clear... also, measure the shaft position to the ring, they sometimes arent shouldered, so you have to reinstall them to roughly the same axial position- I think give or take 1/16 axial wont hurt much, its not super critical- just the teeth need to be somewhat in the middle of the sensor face...radial clearance is important, I usually set them real close, like a piece of paper thickness- seems to help with older sensors to run the gaps tighter than original... be REAL careful no chips/crap get stuck in the fine teeth of the sensor ring, a chip(or ding in the teeth) will mess it up...

edit- shoulda mentioned, if you have a scope, you can look at PA/PB feedback pins- if you see uneven sinewaves, might just need adjusted, but if you see wild amplitude spikes, Id suspect bearing... and if you pull the rear endbell to swap the bearing, watch out for the wavespring too- dont let it fall inside the motor

[Andy Fritz]

I did check the ring and sensor. Ring looks good and I cleaned the sensor and set the gap to 0.04 mm and checked that the ring did not touch during rotation. Did not help. Will look at pa pb again. There is 0.02 mm radial play in the rear bearing but it is very smooth. Maybe I should try to increase the sensor gap a little?

[nautomation]

fix the sensor and emergency stop , rotate the spindle by hand and see the Spindle Speed count is on screen or not ? in both direction , than confirm .
regards

[Andy Fritz]

Hi,

Yes I get RPM in both directions on the lower right on the screen.
Andy

[Andy Fritz]

But I think the RPM on the screen comes from the spindle position coder and not from the spindle motor pulse generator.
There are actually approx 4 high pitched noises per motor revolution, not 1/rev like I wrote at first. It is not exactly 4 noises per rev though as I marked the pulse generator gear as a visual marker. If the marker coincide with one noise it will sync up with the fifth noise and the 9th, but then slowly drift out of sync, so I dont think it is linked to the 4 poles of the motor. This also shows it is not linked to one area of the pulse generator gear as it drifts. If I remove the spindle V-belts the noise is still there but much less pronounced. I think the motor is not turning at a fixed speed but more of a pulsed speed. The noise seems triggered to either acceleration or deceleration. If I run from MDI at 10 rpm the RPM on the screen fluctuates between 10 and 16 rpm.

I am reading some generic VFD trouble shooting guides from other manufacturers and it does not seem like a good thing when output voltage or current is not even between the legs. Unless Fanuc drives are unique in that unbalance is normal I would think something is wrong with my drive. Specially since running the motor direct from mains at least shows the same current draw on each phase.

If someone could point me to areas of the drive to check it would be great.

Andy

[underthetire]

Usually rpm does come from the motor sensor, not the encoder. Encoder is for ipr mode and threading. This really sounds like a transistor module failing, or the drive circuit for it, but I certainly would not rule out the motor sensor at this point. Someone will have the connection manual so you can scope the sensor feedback.

[Al_The_Man]

In my experience, the spindle speed screen display does usually come from the spindle encoder to register actual spindle rpm, rather than motor rpm.
Often this can be confirmed by turning the spindle by hand, and noting the rpm and if it changes?
Al.

[Andy Fritz]

The spindle RPM comes from the spindle position coder. I just took the belts off and get no rpm running the motor and do get rpm turning the spindle by hand.
I didn't check the live tool spindle at 10 rpm before. Now I did and it is very visible how it also fluctuates in speed with a high pitched noise for every deceleration.

The voltage on the motor is U-V 9V, V-W 20V and U-W 23V. Phase to ground is more even at around 130 - 140V per leg. Current is fluctuating U-6-9 amp, V-0.6-1.6 amp and W-6-12 amp. So just like when running the main spindle current is also unbalanced here with V low.

I haven't touched the live tool motor. What are the chances both main spindle and live tool motors or their pulse generators packing up in the same time?

Andy

[harrimansat]

Spindle driver has one bad current amplifier, do you solve the problem?