Fadal won't cold start.

This one has me perplexed.

I can jog all axis. All voltages are good. Changed the power supply
just to make sure. When it tries to cold start, the axis do their jiggle
and then it just sits and thinks, and then tells me "problems positioning
slides to zero".

The thing is I can move the axis off of zero, and they all come back to zero,
and then nothing.


Timeline.

It did this once back in december. ONCE, the second CS, and it was fine.

Then in January, it happened again, it took about 3 or 4 CS's to get her
happy.

Then the next time I went to fire it up. It took several hours before it
would CS. During that time, I obviously shut it down completely a bunch
of times, and re-seated all the boards and everything else. Still was
being pissy, then just all of a sudden worked.

I hadn't shut it down since then, just park her and e-stop her overnights
and weekends, but we had a power blip this weekend. So now I need to
fire her up again, and she is not cooperating.

I called ITS tech support and they didn't know. Guy was supposed to call
me back and never did. I probably should try again..

Once it does CS, it runs perfect.


Any ideas???

94 VMC15.

What the heck??? I figured ITS had you on their speed dial in the event they couldn't figure something out on a Fadal.

What do you mean it comes back to Zero? Do the axis move back to home every time you use the CS then the error shows up? Are their any limit switches/sensors, that could be a little funky, involved on finding the home position?

tmt said:

What the heck??? I figured ITS had you on their speed dial in the event they couldn't figure something out on a Fadal.

What do you mean it comes back to Zero? Do the axis move back to home every time you use the CS then the error shows up? Are their any limit switches/sensors, that could be a little funky, involved on finding the home position?

Click to expand...

Let me try and explain better. Before I shut her down, I always send her home. A CS just makes
her jiggle a tiny little bit, no crazy movement.

So now, I power up, and I can jog all over the place. If I CS with an axis off by less than half a rev, it comes right back to where it should be when I CS. If I go to Y=0.050, it comes back to zero, physically verified, as does the Z and X, so its finding the proper point on the resolver.

Then it just sits there thinking. (*X*X*X*X*X*X*X*X*X*X*X) then after a minute or so "Problem positioning slides to zero"

I did find that error in the manual, and it has something to do with glass scales, which I do NOT have. And its not a parameter thing, you just hook them up.


So, now. I'm thinking something is confused at the axis controller card. I'm assuming (and I don't know) that the glass scales probably plug into either the J5 or J6 blue plugs. Pulling the cards and checking them. And then I'm going to pull a card from another machine and swap it in, and see what happens..

I don't think its the resolver(s) since it comes back to the proper point and stops.

I'm just confused at this point.

Bobw said:

I did find that error in the manual, and it has something to do with glass scales, which I do NOT have. And its not a parameter thing, you just hook them up.


So, now. I'm thinking something is confused at the axis controller card. I'm assuming (and I don't know) that the glass scales probably plug into either the J5 or J6 blue plugs. Pulling the cards and checking them. And then I'm going to pull a card from another machine and swap it in, and see what happens..

I don't think its the resolver(s) since it comes back to the proper point and stops.

I'm just confused at this point.

Click to expand...

It sounds like you are on the right track, I am assuming that the only way you can move the axis is by the hand wheel. Since you have access to other cards that's what I would do, swap them out to see if that solves your problem. If that doesn't work you might have a resolver problem. Although they are robust and a analog design, their circuitry is fined tuned in order for them to function properly. Everything wears out over time, hell everyday I get older a circuit or two of mine don't fire like they use too!

sounds like when a computer start to crap out
like having trouble reading data.

the swapping for known good cards is probably the right idea

had a surge protector that was quitting that made it necessary to
shut the blade switch off and hit the green button a couple times to get it to start

I love ITS but when Dan tells me 'let me call you back' that actually means "I am never going to call you back" lol, I don't think I have ever received a phonecall back from him unfortunately, good to know it's not just me I guess?

I am under the impression that the cold start does not actually rely on the resolvers, I believe it uses the 'once per rotation' index pulse from the tachometer to set the cold start position...

Might be that you have a bad tachometer in one servo, I believe everything else would still work fine in this case...

EDIT: The above might be a wild goose chase... I don't know where I heard that the CS procedure uses the tach and I can't find anything in the docs to back that up, so take it with a large grain of salt!

aarongough said:

I love ITS but when Dan tells me 'let me call you back' that actually means "I am never going to call you back" lol, I don't think I have ever received a phonecall back from him unfortunately, good to know it's not just me I guess?

I am under the impression that the cold start does not actually rely on the resolvers, I believe it uses the 'once per rotation' index pulse from the tachometer to set the cold start position...

Might be that you have a bad tachometer in one servo, I believe everything else would still work fine in this case...

EDIT: The above might be a wild goose chase... I don't know where I heard that the CS procedure uses the tach and I can't find anything in the docs to back that up, so take it with a large grain of salt!

Click to expand...

I was going to say resolvers also but I also thought I heard they had nothing to do with it. tach might be the problem like brushs in them.
however it didnt specify which axis in the error. so it might be something else?
wonder if you can see the following error with out doing a CS. that might help to see what it loosk like?

I don't know if this is useful to you or not, because the symptoms are different, but I'll toss it out there anyway.

On my 2000 4020HT (with Multi-Processor control), if I attempt to start that machine with an unpowered PC hooked up to the RS232 port, it won't cold start. It'll go through the process and it appears to work, but then instead of prompting me to hit start to return the the home position (I've got the home position set to Y10 to make it easier to set up jobs), it'll just reset the control and again prompt me to do the cold start all over again.

First time that ever happened I was in a panic. But I finally figured it out, it doesn't like an unpowered PC hooked up to the RS232 port during cold start. Power up the PC and it's fine.

Problem solved, for now anyways.

As for the location pulse, that comes from the resolver. That's how you can change your
CS position, by rotating the resolver.

Its been a while since I've messed with a tachometer, but there is nothing there
that could give you a location, all it does is generate a voltage based on RPMs
and direction. I think it was 14volts per 1000 rpms, and I think its the only
thing in the entire machine that uses -12v.

I will add my tachometer wisdom here for those that don't know yet. NEVER take
the magnets out, over time the glue loses its stick and the magnet will come
out in pieces, and you can put it back in, but it will never work right ever again.

I learned this the hard way, I had to buy to buy two new complete tachogenerators, when
all I needed was one of the little windy things in the middle.

On the RS232, it was one of the first things I unplugged, not because I'm a genius, but
I remember reading about people having problems with that before, and not just on a Fadal.

So, I swapped out the X,Y,Z axis control cards. No difference, and then I figured I would
just pull the A axis card and see what happened. Fired right up.

I'm not shutting it down, I've got parts to make..

WHY did the A axis card cause a problem? Was it actually the A axis card, or did
I just luck out and she was feeling nice at the moment? I don't know. I'm sure
I'll find out at some point in the future.

Should have pulled the A card earlier, but..


I'll add that I also checked the battery and it was fine. My next thing was to
try zeroing the memory, I have no idea why that was my next move, but it was.

Bobw said:

Problem solved, for now anyways.

I'll add that I also checked the battery and it was fine. My next thing was to
try zeroing the memory, I have no idea why that was my next move, but it was.

Click to expand...

Wow!! I had thought about the A and or B axis but, figured why would that make any difference if they are not plugged in to the actual 4th and or 5th axis. The 4th and or 5th are designed to be removed when not needed. Go figure!

I think zeroing out the memory might have been the probable cure to your ill's. You would think on boot-up that it would automatically clear the memory then do a memory check but, that type of thing is above my pay grade and capabilities. Anyway who really knows what computers really think!

Glad to hear the Fadal Master got it working!! When you get the courage up to turn it off again, grab yourself a good stool/chair and a bottle of Whiskey or your favorite Beverage of Choice, just in case it want's to be contrary again. The Whiskey won't make you any smarter or better looking but, if you get lucky, it might make you feel that you are.

aarongough said:

I love ITS but when Dan tells me 'let me call you back' that actually means "I am never going to call you back" lol, I don't think I have ever received a phonecall back from him unfortunately, good to know it's not just me I guess?

Click to expand...

Put me in that group as well.

Bob, thanks for updating this and documenting what you do, as always.

There are no "pulses" generated by a resolver. There is a "null" point in the analog voltage between the 2 stator windings when the rotor is at a specific position. That "null" position is what a resolver based position control system uses for determining a one revolution point. The Fadal "CS jiggle" is the system moving the motor back and forth while looking at the voltages of each stator winding to determine the position at which they are closest to null. That's where it sets home. A bearing going out in a resolver can eff up that search for null because the rotor can kind of "flop" around a bit and so when the servo jiggles back and forth there is an ever changing point at which the analog voltages are null.