What's new
What's new

Fanuc Repair Gurus - Brain teaser of the week... 21iT tale of woe..

Dan from Oakland

Titanium
Joined
Sep 15, 2005
Location
Oakland, CA
Not to derail the thread, but our shop is within 75 feet of a major North-South highway and the black tire gunk is no joke- I've worried about it for years on control boards etc, but so far no problems. A few machines have the boards contaminated with this stuff- nasty fine dust- its on the roof- turns to mud with rain, on the floor etc. Ya have to see it to believe it, apparently Thermite has seen it too.

Dan
 

Milacron

Super Moderator
Joined
Dec 15, 2000
Location
SC, USA
I will speak with Harrison techs on Monday to try and get encoder pin outs, and Fanuc tech no. ____ to see if I can get the pin outs of the drive encoder cable socket. I'm also going to go ahead and order a replacement servo card. Even though I may not actually need one I figure the $100 to $150 blown on returning it will be worth it just to eliminate that from the mix of possibilities. But if I do need it, all the better, case closed and the failure rate will jump from .001% to .002%.
 

Dave K

Diamond
Joined
Mar 21, 2004
Location
Waukesha, WI
Isn't there a way for you to send your card to them, and have it tested, rather than buy one and take the chance that wasn't the problem?
I thought I recalled a friend of mine saying you could send board and cards to fanuc to have them tested to see if they're bad.
I'm no wiz at this stuff, so I'm just throwing this out there.
 

Milacron

Super Moderator
Joined
Dec 15, 2000
Location
SC, USA
Isn't there a way for you to send your card to them, and have it tested, rather than buy one and take the chance that wasn't the problem?
I thought I recalled a friend of mine saying you could send board and cards to fanuc to have them tested to see if they're bad.
I'm no wiz at this stuff, so I'm just throwing this out there.
Sure....I can ship my card to them and they will test it. The bill for that service is $400.

(not a typo)
 

Bruce Griffing

Titanium
Joined
Jan 1, 2003
Location
Temple, Texas
Fanuc encoders use a query/response system where the servo asks for a reading and the encoder sends a serial response. It is very hard to check them out of circuit without some type of test jig. Even with that, you need an oscilloscope. Further, the whole timing scheme is secret. The easiest test is to put one into a known good system.
 

hitandmiss

Titanium
Joined
Jul 31, 2010
Location
Rochester, NY USA
The encoders are absolute, the backup batteries are attached to the Fanuc servo drive units, the encoder cables connect to the servo drive units. Re encoder troubleshooting, both axis ceased to function at exactly the same moment thus it would seem highly unlikely that both encoders went bad simutaneously.

*Another cute thing is that only 5 of the 10 wires are different colors. The other 5 are black. In checking for continuity, for the black wires I took a red Sharpie and marked tiny lines on each black wire so that I knew which wire was 2 marks and which was 4 marks, for example. But if any of those black wires carry an important voltage or signal it would be pretty difficult to know which was which.

Which begs the question, why would a cable manufacturer even make a cable like that in the first place ? There are more than 6 colors in existance after all, so if one can make a cable with 6 colors, why not go ahead and make it with 10 colors ??? (or colors with stripes or lines or whatever to differentiate one wire from the other)

The cable color code is a common one for differential signals. Once all the colors have been used with the black, then red is used with all colors except black, then the next color is used with all except red and black.
Those black wires are paired with a non black wire, as in red/black twisted pair. The black wires carry the inverse of the signal in the colored wire of the pair.

One way to muck things up is by installing a battery that has the polarity reversed on the connector.
Another way is to swap the encoder connectors at the amp, the X encoder connected to the Z amp, and Z encoder to the X amp.

I have a customer that has a lathe like yours, I called him a few min. ago, he will check Monday to see if it is the same model with a 21i control.

Bill
 

Milacron

Super Moderator
Joined
Dec 15, 2000
Location
SC, USA
I have a customer that has a lathe like yours, I called him a few min. ago, he will check Monday to see if it is the same model with a 21i control.
Thanks. Most likely he has a Harrison 330 Alpha S+ as there are only three other 1330S machines in the USA. Regardless, if it is 1999 or newer it will have a 21i control. Actually even the larger CNC Harrisons might have the same setup, control wise, as long as they are 1999 or newer.
 

jCandlish

Titanium
Joined
Jun 1, 2011
Location
Oberaargau, Swizerland
Further, the whole timing scheme is secret. The easiest test is to put one into a known good system.

Jon Elson of Pico Systems reverse engineered the Fanuc serial protocol and offers converter cards for people wanting to retrofit the older controls.

fanquadsml.jpg


connection diagram with pinouts ==> Fanuc Serial Encoder Converter Connector Pinout

A little bit about the protocol:

It produces a 77-bit string of bits. There is a wide position readout that includes a field for
number of turns (signed) and a field for shaft position, such as 32768 or 65536 counts
(unsigned). It ends with a 5-bit CRC code to verify data quality.
If the encoder is an "incremental" encoder, it has no indication of position, and the
shaft angle value jumps suddenly to zero when it passes the index position for
the first time after power on. So, what they do is provide a backup battery in
the controller, and have brakes in the motor that assure the encoder cannot move
by even one count while powered off. The battery does not power the encoder, it
only powers a CMOS memory that holds the position count while the rest of the
encoder is not powered. This type of encoder could be used to derive commutation
info, but it would need a brake and backup battery, and if the memory was lost,
it would have to be hand-cranked one rev to pass the index pulse.

If the encoder is an "absolute" encoder, it also has a low-resolution field that
reports absolute position with 1024 counts/motor quadrant immediately at power on.
This could be used to derive commutation info without requiring a battery, and
the standard homing routine could pick up the index position. The encoder reader
would need to detect the appearance of the "homed" bit and prevent a servo trip at
that point due to the shift in count.


I guess Fanuc is hanging the fiber-optic transceiver after all that.

So there are two versions of the 21i-T control. 21i-TA and 21i-TB. Which one do you have? Why not post a picture of the control?
 

metaltech

Hot Rolled
Joined
Feb 7, 2007
Location
midcoast U.S.A.
Don, I've had to replace these type boards a couple of times, either the Axis Control or the Servo Interface modules mounted on the MAIN (I'm talking 16/18/21, not the i series, as I previously mentioned, but I suspect they are similar). They're pretty damn reliable, but anything can fail, of course. I would send it to TIE to be tested. I believe what you want to ask for is "Clean and Test". Last time for me was $150. If bad, they'll put the test fee toward a repair or exchange, or they can just send the bad part back.
 

Milacron

Super Moderator
Joined
Dec 15, 2000
Location
SC, USA
Don, I've had to replace these type boards a couple of times, either the Axis Control or the Servo Interface modules mounted on the MAIN (I'm talking 16/18/21, not the i series, as I previously mentioned, but I suspect they are similar). They're pretty damn reliable, but anything can fail, of course. I would send it to TIE to be tested. I believe what you want to ask for is "Clean and Test". Last time for me was $150. If bad, they'll put the test fee toward a repair or exchange, or they can just send the bad part back.
Ok I called TIE and $150 it is...except they charge less than Fanuc for restocking fee so it would be $50 cheaper to buy one and return it if the card wasn't the issue. I did talk to their tech and he emailed some instructions for temporarily changing parameters so the control only looks for one axis per bootup. That, plus moving the fiber optic cable (and dummy plug) around will confirm both axis not communicating. Or maybe, just maybe...one axis is communicating just fine but the other isn't, and is somehow mucking up the other axis when both are being looked for in bootup. If both axis are not communicating when viewed separately, then bad servo card is about all that is left for a culprit.
 

Milacron

Super Moderator
Joined
Dec 15, 2000
Location
SC, USA
I did talk to their tech and he emailed some instructions for temporarily changing parameters so the control only looks for one axis per bootup. That, plus moving the fiber optic cable (and dummy plug) around will confirm both axis not communicating.
Didn't work for some reason. I changed parameter 9900 from 2 axis to 1 axis but with one axis disconnected I then get 5136 alarm "number of amps too small". Just to see what would happen I changed 9900 to a zero and still got "number of amps too small"..... how can number of amps be smaller than zero ?? :cryin:

Methinks either I must have yet another parameter telling the control how many amps (drives) I have, or a bad servo card is causing some crazy alarms.

(on edit) Now TIE is telling me to order the servo card. Apparently they have concluded my 9900 experience indicates a servo card gone insane.
 

metaltech

Hot Rolled
Joined
Feb 7, 2007
Location
midcoast U.S.A.
I'm waiting with baited breath (WTH does that expression mean, anyway?) to see if it's that card. I've been suspecting that's been the problem. I hope it is, since that's relatively easy and inexpensive.
 

Milacron

Super Moderator
Joined
Dec 15, 2000
Location
SC, USA
I'm waiting with baited breath (WTH does that expression mean, anyway?) to see if it's that card. I've been suspecting that's been the problem. I hope it is, since that's relatively easy and inexpensive.
That's been the main suspect with me after I determined the fiber optic cables were ok, but the machine is so late (to me anyway), pristine and low hour, and the supposed failure rate of that card is so low, it's just been hard to wrap my head around the fact that the servo card could really go bad.

MT, can you edit your location entry in profile as "midcoast" is nearly meaningless.
 

Milacron

Super Moderator
Joined
Dec 15, 2000
Location
SC, USA
I'm waiting with baited breath (WTH does that expression mean, anyway?) to see if it's that card. I've been suspecting that's been the problem. I hope it is, since that's relatively easy and inexpensive.
So, I ordered a brand spankin new servo card today. UPS ground from TN (assuming they don't in reality have it drop shipped from Fanuc in IL)...so maybe I will have it by Thursday or Friday. After all this testing, what a cringer it will be to boot up the machine after that new card is installed.

I know the screen will still be flashing "alarm" as I wlll still have the 300 alarms. But that will be one nervous index finger aiming at the "message" button to see if the 301 has finally been banished. :ill: And if 301's are gone, the next cringer will be turning the handwheels to see if motion and position display have returned to the axis and monitor.
 








 
Top