Fadal 4020a spindle shuddering, fluctuating speeds and going backwards to commanded direction

[static99899]

So the 1010-6 card with the ele-0001 eproms came in and I promptly installed it. Spindle turns wrong way so I switched T1 & T2 outputs from spindle drive. Spindle turns correct way and speed but now it throws an "error 8 motor overload". Attempt spindle orient, spins correct direction and speed but the orientation arm never tries to engage and it eventually gives an "error 31". DS/DI to check for orientation magnet detection and the bit changes when the magnet passes so the control sees it. I did notice that if I move it faster than about 15 rpm by hand, it doesn't change the bit(or it happens too fast for me to see it)

Saw someone having weird issues online due to a dead or low voltage CPU board battery. I ordered and changed the battery. No change in issues.

Decided to put the original 1010-4 spindle card back in, and now it will spindle orient again but now it is giving one of two errors when I try to run the spindle. Either "error 8 motor overload" or the newest error is "Axis controller does not respond to NC"! AAAAGGGGGGHHHHHH!!!!!!!

Not sure what to do next, I'm waiting to hear back from Dan at ITS(Itold him all of this above)

Just so we are all kept up on what's been replaced, here are the ONLY things I HAVEN'T replaced:

Spindle motor
Delta/WYE contactors
CPU board (1400-4), program module (1400-4) & computer interface board (1030)
backplane (1060-0B)
power distribution board in spindle cabinet (1101-0A)
Wiring to and from all components

Once I do figure this out, I'll have enough spare parts to build another machine lol.

 

[triumph406]

If BOBW read this he'd likely suggest checking the 12v power supply.

 

[snowshooze]

I bought a used 1996 Fadal 4020a with the 15hp 7500rpm rigid tap spindle. As the title says, if I command an M19, sometimes it orients and most of the time it does not. If I command S1000 M3, it turns CCW at about 55 rpm and has the spindle load meter pegged at 150%. It never errors out, so I hit the spindle stop button and sometimes it stops, sometimes is shudders CW and CCW until it stops and sometimes it continues to spin at about 30 rpm.

Those are the symptoms and now I will list what I have checked or replaced; in the order I replaced/checked them. Please note, none of these have made any change at all:

1) verified voltages incoming were correct, and balanced within 5 volts. This was checked at the transformer, and at the spindle inverter.
2) verified and added addition machine ground to earth ground rod
3) verified DC voltages coming from power supply in control cabinet.
4) Checked encoder cabling for shorts with meter, none found
5) unplugged encoder, it runs at the correct rpm for 3 seconds the says "motor overload" and shuts the spindle off (so it sees the encoder feedback is missing when unplugged)
6) since I don't have an oscilloscope, I replaced the US digital encoder
7) swapped 1010-4 spindle card for a known good 1010-4 spindle card
8) Bought a battery box from cncpros to directly test spindle drive, drive failed the tests due to still having the issue after bypassing control cards, so...
9) bought a brand new Glentek spindle inverter, and the problem persists! (mode 1; 15hp 7.5krpm sensored)

So I did try to operate the spindle drive in a mode that doesn't require the encoder input and it will work perfectly in this mode. It spins at the speed given by the battery box and turns the correct direction. On this setting, it's the wrong rpm for my machine(15hp 10krpm VFD), so I can't leave it there and I'd like to retain my rigid tap feature, so this is not a good permanent option.

Neither the original Baldor drive nor the new Glentek show an error of any kind. Both display the wrong rpm and direction of the actual motion.

I'm running out of ideas and money to throw at it in parts so if anyone has a suggestion, I'd loved to get another perspective before calling the local tech out to the tune of $150/hr.

Edit: Between 6 & 7, I did replace with a brand new hall sensor for the orient and it does show it works in the DI page.

If you are anywhere near Chatsworth, call Dave Madden.
It'll take him 10 minutes to sort it out.
He did that for me, and I flew him to Alaska..
M&M CNC.

 

[triumph406]

Are you checking the voltage output from the spindle card to the spindle drive? That would tell you if the correct voltage value is being output.

What happens if you remove the encoder and run open loop?

 

[static99899]

If BOBW read this he'd likely suggest checking the 12v power supply.

I did replace it with a brand new one from CNCPROS, no change

 

[static99899]

If you are anywhere near Chatsworth, call Dave Madden.
It'll take him 10 minutes to sort it out.
He did that for me, and I flew him to Alaska..
M&M CNC.

Unfortunately I'm in Bowie, Texas. My closest Fadal service that I know of is Eaglemachine in Mansfield and their reputation for quick diagnostics is not great but may have to use them eventually.

 

[static99899]

Are you checking the voltage output from the spindle card to the spindle drive? That would tell you if the correct voltage value is being output.

What happens if you remove the encoder and run open loop?

I have not check the output but I doubt 3 different cards would all output the wrong voltage.

I haven't tried this again since getting the 1010-6 card. I'll have to try that again and see but I have replaced the encoder, and both of the encoder cables so not sure what could be wrong there.

 

[Chimpwithastick]

This may be completely useless, but how does the spindle Orient Lock? I had a Matsuura that was doing some pretty screwy things with accelerating to commanded speed, tapping, and tool changes. It ended up being the orientation lock proximity sensor that was failing and preventing FIN signals from sending. Just throwing it out there, sometimes its something "random."

 

[snowshooze]

Unfortunately I'm in Bowie, Texas. My closest Fadal service that I know of is Eaglemachine in Mansfield and their reputation for quick diagnostics is not great but may have to use them eventually.

You call Dave.
The danged guy is for real.
He tried like heck to diagnose over the phone and just tell me how to fix mine.
No charge, Too much time.
Finally I begged him to come to Alaska. Airfare, R&B, refreshments.
I arranged for a three day stay at a very nice place.
( I figured we had to have time if parts were needed )
He warned me he would probably fix it in ten minutes.
He was wrong. He had it fixed in 5 minutes.
Then he fixed my Anilam Control, in about 10 minutes.
He'd never seen one before.
I don't remember much after that aside of being drunk and fishing...
I was so pleased. It put me on my feet.
( He moved a jumper on a board....)
1-818-974-3992 M&M
I am still best of friends with him and we yap it up on the phone all the time.
Hey, he travels all over still. Other countries even.
And you'd do my buddy a favor, it's slow in Cali right now.
Take him on a hunting trip after he has your machine up in 10 minutes....

 

[triumph406]

I have not check the output but I doubt 3 different cards would all output the wrong voltage.

I haven't tried this again since getting the 1010-6 card. I'll have to try that again and see but I have replaced the encoder, and both of the encoder cables so not sure what could be wrong there.

If you disconnect the encoder and run the drive open loop, and the problem persists, then it isn't the encoder. If it stops, then you have an encoder problem. Certainly worth a try.

also might want to check the fault line from the drive to the card.

 

[Lenco Machine]

Hey this sounds like the exact same problem we had a few weeks ago with our 4020. Our motor was bad so we ordered a rebuilt one. Put it in and was getting a very jerky, kinda twitching start and was going WAY too slow. We jacked with it for a few days, swapped some leads and got all the same symptoms you described. We called Eagle Machine and talked to Sam and he walked me thru some things over the phone and had me swap some more wires and we had it working in less than an hour. The encoder was probably on backwards. You should give them a call.

 

[static99899]

Ok everyone, I think it's fixed. it's been running for a week with no more issues so far.

So if you look back on all of my previous posts, you know I've been chasing this for a while and at great expense. The devil is in the details as they say....

One of the first things I replaced was the hall sensor for spindle orient. I had bought it new off of ebay and the part number advertised was correct so I installed it, did the diagnostic screen, the machine saw the sensor but it didn't fix my issues, so I moved on.

While cleaning up after a recent trial and error session, I found the box the sensor came in, and as I sometimes do, checked to make sure there was nothing I "needed to keep" in there and found a slip of paper under a flap of cardboard that showed how to install the sensor in the tool atc turret. So I found this chart at ITS : https://itscnc.com/proximity-switches

And guess what! The damn senor was the wrong one!
So I ordered the correct honeywell sensor from ITS (ele-2500), which side by side looked nearly identical to what I had, installed it and the machine has been running ever since.

If anything changes, I'll let yall know. On the bright side, the machine runs and I have a LOT of spare parts for it now lol. Ugh, so frustrating...