The never-ending saga of the Hurco that's "ready to make parts"

Re: Code issue, please help. This stopped being fun a few days ago.

he's right, if the cards are the same for x y and z, swap the cards and see if the problem moves axes with the card swap.
 
Re: Code issue, please help. This stopped being fun a few days ago.

remember, the only reason the motor will move like that is if it 'thinks' the motor is out of position. Encoder or encoder feedback problem? Check the brushes and commutator of the encoder if it's of that type.

Is the encoder on the ballscrew, or on the motor?

Can you verify the 'following error' via the control? See if it's bouncing.
 
Re: Code issue, please help. This stopped being fun a few days ago.

Pulled the gib out and checked it and all seems well. I didn't expect to find a problem there because I can move the table (by hand) through the complete range with no real effort and no noticeable change in feel. Good to eliminate it as a culprit though.

Not sure why I did this but I pulled the motor out to check the encoder. Connections seem fine. I put an ohm meter on each connection and wiggled the wires and got no interruption. If there is a short, it's not in there. I'll check the main wires to the cabinet before I button it all back up. I realize I should have just switched the cards first but this was knawing at me for some reason.

When the machine jumps position, the readout on the screen follows it so I don't believe there's a problem with the encoder. It seems to be tracking just fine.

Anyway, here's a shot of the encoder since I had it apart. Not much info for these machine on the interwebs so I figured why not. It's a Datametrics S-9992b.

100_0903.JPG





On the first page of this thread I posted this picture and said it was the encoder. Obviously I was wrong. Anyone know what this is? It's on the Z axis.



attachment.php?attachmentid=81432&d=1407250150&thumb=1.jpg

100_0903.JPG
 
Re: Code issue, please help. This stopped being fun a few days ago.

Since you have the motor out anyway, pull the encoder off of the motor and rotate the shaft by hand. If it feels rough at any spot, then that may be the problem. Sometimes the bearing gets sticky and will cause what you are seeing.

In some encoders, the actual encoder is mounted on a spring plate inside the housing and if the bearings are sticky, it will bounce in certain spots. On some of the machines that I used to work on, just a small ''tick'' noise as the encoder is rotated was enough to fowl up the whole system. You had to have the encoder right up against you ear to hear this.

It seems that the Datametrics encoders are not available, it would be easy to replace the them with another brand, and the good news is that they are reasonably inexpensive. If it comes to the point of replacing encoders I'll be happy to recommend something if you like. All of the needed information should be on the data plate on the encoder.

I suspect the z-axis picture of what you thought was the encoder, is actually a lubrication tube, protected by a flex conduit.

By the time you get done with this, you are going to be a CNC trouble shooting expert.
 
Re: Code issue, please help. This stopped being fun a few days ago.

JimDawson said:
By the time you get done with this, you are going to be a CNC trouble shooting expert.
Click to expand...

:lmao:


Not exactly my goal but I suspect so.




I'll pull that off tonight and check it out. I'll also check the coupling to be sure there's no defect in it either.



I was trolling the interwebs earlier and stumbled across an old thread and someone was having a problem with one of their axis; not the same problem, but a problem none the less. It was pointed out to them that the cables from the cabinet to the servos have the same plugs and can be interchanged to see if the problem is inside the cabinet or outside. I suspect this could also help determine if the problem is in the cable too. If I switch the cables and the problem moves with the cable then I can switch the boards to see if the problem follows the board. This should narrow it down a bit further. I was wondering how I was going to tell if the cable had issues without actually probing each pin and a corresponding connection in the cabinet.

This is way more fun than you guys can imagine :nuts:.
 
Re: Code issue, please help. This stopped being fun a few days ago.

One more thing. When I pulled the cover off the wire housing on the side off the motor (where the plug goes) it was about 1/2 full of oil. The motor and encoder was completely dry and I pulled out the brushes and they were all dry but apparently oil seeped into the plug housing. Could this have an adverse affect? How should I get that oil all dried out to put it back together?
 
Re: Code issue, please help. This stopped being fun a few days ago.

The best way to clean the oil out is using 1,1,1triclorethane, but I don't think you can't buy that any more. So, find some electrical contact cleaner in a spray can, probably from your local industrial or electrical supply, or maybe an auto parts store, a NAPA store might be the best bet here. The good news is that you live in a heavy industrial area, so it should be pretty easy to find. If you can find a store that carries CRC products, they will have just what you need.

As long as the oil is not in to the works, it is probably not causing a problem.
 
Re: Code issue, please help. This stopped being fun a few days ago.

swap the boards. swap x and y, and see if the problem moves to x. it's very easy, common troubleshooting method. Also, check the board backs for hot spots (darkening), blown resistors or whatnot, and clean the board and drive contacts and terminals and make sure there is no corrosion.

but swap the boards, this will tell you if the y card is bad or needs adjustment.
 
Re: Code issue, please help. This stopped being fun a few days ago.

While I had the Y motor out I decided to check it out. I cleaned the contacts, pulled the brushes, checked and wiped them off and grounded the shielded cable inside the motor (approx 6" of cable) put it all back together and checked it.

Same problem.




Just swapped the boards and the problem is still on the Y axis.




What next?
 
Re: Code issue, please help. This stopped being fun a few days ago.

I would say you have eliminated everything but the encoder. Without a scope, I would swap encoders and see if the problem follows the encoder.
 
You must log in or register to reply here.
Back
Top