Printable View
Hi I am having problems with a hitachi seiki 3ne with fanuc 6t control. the machine suddenly started to mis position / miscount on the x axis
I thought this might be an encoder issue so I changed it for a used one, and the problem still persists,
even with the servo disconnected mechanically when I jog the axis using mpg it will not return to the original position.
my question is : could this be a set up problem or u.s. servo control card thanks in advance for any help
-check carbon bruch in tachogenerator.
-check vcm and tsa signal.
Best regard.
Enikma_II
thanks for the reply I have had the servo apart and cleaned the comm and brushes etc but I don't think it has a tacho generator fitted as it only has the 4 main brushes it is fitted with a 2500 line
pulse coder directly on the servo, I am not sure how the system works, but the pulse coder seems to be the only method of positioning/speed i.e no linear scale or other tacho,
regards Clive
-change pulse coder X and Z axis
Best regard.
Enikma_II
first thing I'd do is smell the x axis pulsecoder plug... might sound silly, but if its corroded/burnt between the blue insert layers it will definitely stink...usually though, they are pretty good at detecting missing pulses from shorted connectors or broken pins... if you think you can get the old amphenol shell apart without breaking the 30 yr old wires off, might look in the back of the plug for the shield- be sure its insulated. they use a ground clamp to ground the shield in the cabinet, but the rest of the shield needs to be insulated from ground- else things like brush dust in the motor(assume this has yellow or black cap servos) will cause ground fault currents to circulate thru the shield...a noisy encoder usually trips out, but not always.
second thing id try, is take out two of the big gold LSI chips and swap them...x is on the left. swap the first two chips, if it goes away, bet the LSI is damaged. buy a RC robot motherboard - can probably get one for near nothing as they are about beyond obsolete- but there are 6 of those wonderful 7103 or whatever chips on there...handy if you ever need to add a analog spindle to a zero-c too :) the numbers i think might vary between controls, but have never seen a issue with interchanging...
suggest checking the 5v supply too- it needs to be 5.0 and noise free...as the caps age, they can get noisy... again, if you can find a old RC robot control for parts, they have the B002 power supplies also- although at their ages now, would suggest replacing all the caps
while thinking about the old sixes, thought id mention another kinda common failure to the old 6 (or zeroA) is the little 455 op amps right next to the axis LSIs, they wont cause lost counts, but will cause scratchy output on the velocity command. back before we mothballed all our 6B-2s had to swap quite a few 455 op amps, and quite a few of the blocking diodes across the three big reed relay coils...if a 6 ever shows a 5 volt short on the master, get a low-ohm meter and probe around for the lowest point...odds are it will be one of those relay diodes.
aside from the problems above, the 6 was like a old ford truck...just cant hardly kill them.
well, didnt mention the sticky reeds on the 'connection unit' (i/o card), as most folks still running a 6 know on occasion they might need to strum a plastic screwdriver handle across the 'chips' to dislodge the sticky contacts... I really like the old 6B2... we just pulled the last one out last year, think it was bought in 84-85, beat to death, still making product.
the 6B2, the 6050 amp and the Alpha22 motors I think were the best Fanuc ever made, durability wise. I havent ran a alpha22 on a 6050 yet, but probably will soon...we are using them on old 6066 robot drives with good results... theres so much copper in a alpha ( just lift up a A12/A22 and a similar size old 10S/20S- the alphas are heavy ) they make a ton of torque- pretty sure they would do fine on a 185 volt drive...