Centroid & MPG Problems

[Marty_Escarcega]

First a little history. A friend bought the newest Centroid retrofit, cabinet and all. The control is the single board identical to the Ajax board. They installed the included prewired servo motors and encoders. The lathe is a Bridgeport EZ Path. Right now the conversion is about complete however, they are wanting to use the original Bridgeport encoder handwheels on the carriage. It is strange, when you gradually turn the X axis, there is no movement of the cross slide UNTIL you turn it RAPIDLY, then it will move! Z axis will not move with the Z axis handwheel. I am not familiar with Centroid or AJAX controls but would like to help. Some direction and diagnostic screen and setup screen locations for the MPG encoders would be helpful. Thanks in advance.

[txcncman]

What did Centroid tell you while you were talking to them on the telephone about this problem? 814-353-9256

[Marty_Escarcega]

Originally Posted by txcncman What did Centroid tell you while you were talking to them on the telephone about this problem? 814-353-9256

I am not at the machine, I hope to be tomorrow. Local tech has been out a couple of times, doesn't seem to be much help. I was HOPING from some pointers from the group so I have some ideas before I get there.

[txcncman]

Did you look on the Centroid site? CNC Milling machine handwhee.

[Marty_Escarcega]

Originally Posted by txcncman Did you look on the Centroid site? CNC Milling machine handwhee.

Different animal. The Bridgeport EZ Path CNC lathe has hand wheels on the apron placed similarly where they would be on a manual machine. The difference is there are encoders on each handwheel that move carriage and cross slide via the servos.

The weird thing is you can gradually turn the X axis wheel no movement, give it more of a sudden turn and it will move the cross slide.

[txcncman]

Originally Posted by Marty_Escarcega The weird thing is you can gradually turn the X axis wheel no movement, give it more of a sudden turn and it will move the cross slide.

And that is probably the number one reason to NOT use the original parts of the machine. I have operated and taught Centroid controls before and using the keyboard jog keys is not difficult to learn (if someone wants to learn).

Reason I posted the link was not to tell you what hand wheel to use. It was just to show that Centroid is capable of using a hand wheel for MPG. I really bet if you phone Centroid, they will attempt to assist this installation and even customize it for your wishes.

[cncsnw]

In the mean time, if you want to move forward with troubleshooting the existing installation, I would first recommend checking whether you are getting reliable counts from the handwheel encoders.

You presumably have the handwheel encoders connected to two of the six encoder inputs on the All-in-one DC unit: probably either #3 and #4, or #4 and #5 (depending on where you have the spindle encoder). Let's assume here that the handwheels are on #4 and #5 (and the spindle encoder is on #3).

I am guessing that you need to map the inputs to axis numbers, at least so that you can get the CNC11 software to show you the counts on its PID configuration screen. To map All-in-one encoder inputs #4 and #5 to axes #4 and #5, you would set Machine Parameter 311 to 4.0, and Machine Parameter 312 to 5.0. See page 14-29 of the T-Series operator's manual.

You should then be able to monitor counts on those encoder inputs, by watching the PID Configuration screen. You should see 4x as many counts per revolution of the handwheel as the encoder has lines. E.g. a 1024-line encoder should give 4096 counts per revolution.

You should see consistent counts on the PID Configuration display regardless of the speed or direction you turn the wheel, and regardless of whether it is working as an MPG or not. This is just the raw encoder input data.

If you don't see consistent counts on the PID screen, then there is an electronic or connection problem.

If you do see consistent counts on the PID screen, then there is a software or configuration problem.

Machine Parameters 348 to 353 should be set to appropriate values to enable encoder inputs as MPG handwheels. See page 14-30 of the T-Series operator's manual. In addition, the PLC program needs to include appropriate logic to enable MPG's #1 and #2 in response to the jog panel key; map them to their respective axes; and allow the jog increment to change in response to the jog panel increment keys. The default/standard PLC programs are written for the single-wheel MPG pendant mentioned above, with its own axis select switch and increment select knob. Therefore a special PLC program is required to support dual handwheels like this. You can find out the name of the PLC program by looking at the F7/Utility -> F8/Options screen. You can offload a copy of the PLC program (among many other things) using the F7/Utility -> F7/Report feature.

Let us know what you find out.

[Marty_Escarcega]

Originally Posted by cncsnw In the mean time, if you want to move forward with troubleshooting the existing installation, I would first recommend checking whether you are getting reliable counts from the handwheel encoders. You presumably have the handwheel encoders connected to two of the six encoder inputs on the All-in-one DC unit: probably either #3 and #4, or #4 and #5 (depending on where you have the spindle encoder). Let's assume here that the handwheels are on #4 and #5 (and the spindle encoder is on #3). I am guessing that you need to map the inputs to axis numbers, at least so that you can get the CNC11 software to show you the counts on its PID configuration screen. To map All-in-one encoder inputs #4 and #5 to axes #4 and #5, you would set Machine Parameter 311 to 4.0, and Machine Parameter 312 to 5.0. See page 14-29 of the T-Series operator's manual. You should then be able to monitor counts on those encoder inputs, by watching the PID Configuration screen. You should see 4x as many counts per revolution of the handwheel as the encoder has lines. E.g. a 1024-line encoder should give 4096 counts per revolution. You should see consistent counts on the PID Configuration display regardless of the speed or direction you turn the wheel, and regardless of whether it is working as an MPG or not. This is just the raw encoder input data. If you don't see consistent counts on the PID screen, then there is an electronic or connection problem. If you do see consistent counts on the PID screen, then there is a software or configuration problem. Machine Parameters 348 to 353 should be set to appropriate values to enable encoder inputs as MPG handwheels. See page 14-30 of the T-Series operator's manual. In addition, the PLC program needs to include appropriate logic to enable MPG's #1 and #2 in response to the jog panel key; map them to their respective axes; and allow the jog increment to change in response to the jog panel increment keys. The default/standard PLC programs are written for the single-wheel MPG pendant mentioned above, with its own axis select switch and increment select knob. Therefore a special PLC program is required to support dual handwheels like this. You can find out the name of the PLC program by looking at the F7/Utility -> F8/Options screen. You can offload a copy of the PLC program (among many other things) using the F7/Utility -> F7/Report feature. Let us know what you find out.

Thanks for the very useful information. I was already gone and over there before I saw this. Centroid setup the control for the two MPG handwheels. I think encoder input 4 and 5 were setup for this. It was more of a trying to use the old encoders with the new control thing. We found a bad line driver chip on the encoder. Probably happened when they were wiring it up. The one encoder that seemed to work was a little odd with the control. Turn the wheel slowly nothing, turn it quickly and produced choppy movement. Centroid sent little TTL to Differential boards to put between the encoder and the encoder input. Now get this we had the X axis handwheel plugged directly into the control and the TTL to differential board plugged in on Z encoder input with no encoder on it. IT MADE X WORK OK!! Figure that one out? We plugged in the second TTL to differential board on Z and everything was working ok. The encoders themselves are Differential! A, A/, B, B/ +5vdc, GND. Still scratching my head on that one.

Question though, they encoders are 600cpr. We think we found where to input them in the software (please verify if you can) I think the handwheels have 250 graduations. Do you input 250 where it says encoder detent (trying to go from memory here)

Thanks for the assist. Been fun trying to work on this thing.

Marty

[cncsnw]

Now get this we had the X axis handwheel plugged directly into the control and the TTL to differential board plugged in on Z encoder input with no encoder on it. IT MADE X WORK OK!! Figure that one out? We plugged in the second TTL to differential board on Z and everything was working ok. The encoders themselves are Differential! A, A/, B, B/ +5vdc, GND. Still scratching my head on that one.

Sounds like everyone involved with this was a little confused.

If the encoders are differential, then there should be no need for the TTL to differential boards anywhere.

If the encoders really have 600 cycles per revolution, then you would enter 2400 for Parameters 350 and 353.

If the handwheels really have 250 detents, then you would enter 250 for Parameters 349 and 352.

It seems improbable that Bridgeport would have installed 600cpr encoders on 250-position handwheels, since 250 does not go evenly into 600 or 2400. Were these encoders aftermarket replacements?

[Marty_Escarcega]

Originally Posted by cncsnw Sounds like everyone involved with this was a little confused. If the encoders are differential, then there should be no need for the TTL to differential boards anywhere. If the encoders really have 600 cycles per revolution, then you would enter 2400 for Parameters 350 and 353. If the handwheels really have 250 detents, then you would enter 250 for Parameters 349 and 352. It seems improbable that Bridgeport would have installed 600cpr encoders on 250-position handwheels, since 250 does not go evenly into 600 or 2400. Were these encoders aftermarket replacements?

The encoders appeared to be the originals. I should double check on the handwheels to see how they were graduated. They might have been 200.
Thanks for the info on the parameters.

Marty

[Marty_Escarcega]

Originally Posted by Marty_Escarcega The encoders appeared to be the originals. I should double check on the handwheels to see how they were graduated. They might have been 200. Thanks for the info on the parameters. Marty

Found the handwheels were 240 divisions. SO that works, 600cprx4 =2400