MPG jog wheel

[The Blight]

I know TCNC is kind of crappy when it comes to handling jog wheels (MPGs) because it ignores the input from the wheel while moving an axis. What happens is that you turn the jog wheel, TCNC moves the axis, and ignores further input until move is done. If you turn it fast enough, TCNC will see the bits that come before the ones it last read.... hmm.. will clear that one up.

A B
0 0
1 0
1 1
0 1

It starts of at 0 0, and when you turn the wheel, it goes to 1 0. TCNC moves the axis, and meanwhile you turn the wheel, and go by 1 1, 0 1, and come back to 0 0 (if TCNC is done making its move it will register this), so TCNC thinks you are going backwards.

I have a high pulsecount encoder (400ppr), and TCNC counts every single bit changed. When one bit changes, it makes one step (or as many steps as you set it). This I dont really like, so I have made a simple program for a PIC which waits for a full sequence before it sends the signal on to TCNC. So after a full sequence, the PIC sends a 1 0 to TCNC. After the next sequence it sends 1 1 to TCNC..etc. I might also put in a timer into the program so that the PIC wont send out signals faster then what TCNC can handle (this will have to be adjusted out from how big steps you are taking).

So if this actually works, I will be able to make 100 steps with my 400ppr encoder (as it is intended for). This way it will be a bit harder to turn it too fast. Right now I cant even exceed 60rpm. So if I can get up to 3rps I'm happy. Dave has talked about fixing the code so that TCNC checks the input all the time.

If anyone is interested in this, or have any suggestions or experience with jog wheels in TCNC, please comment.

PS. I will have to wait until next week to test the thing. I need a better powersupply and I have to get my breakout board back from work.

[The Blight]

Well I got the power supply today, so I have programmed the chip, and if I have the time, I will test it tonight without a breakout board. So it will just be to see if it does what it's supposed to do. I wont be able to see if it works with TCNC. That will have to wait.

[The Blight]

Got to do a little test, and it seems to be working. I guess this will create a smoother signal (because all bits sent to the parallel port will be of equal length as long as you turn the handwheel at a "constant" rate). I don't know if all the bits from a rotary encoder are of equal length. If not, then I can see how TCNC can have so much trouble reading all of them. Anyway, when I get my breakout board back on mondey, I will get to do some more tests on this.

PS. I don't really mind talking to myself, and I don't even know if anyone is really listening or caring, so I will just keep on posting. Maybe some day someone will find this usefull.

[The Blight]

Got to test it with TCNC today, and it works a lot better then it did without the controller. I will hand out the hex file later. The controller does not need any external parts if you can supply power from the breakout board.

[Bubba]

Just stumbled on your thread, don't know how I missed it before, but YES, I am interested!

also wish Dave would release the update! I have a visit from him last September and based on what he said, thought it would be shortly.

[The Blight]

I will just have to test it on my machine next week before I hand out the code. A simple schematic might be added too.

I hope he starts working on it again. If he does, I will have to ask if he can implement a tool changer macro. Or else I will have to make another chip to take care of that.

[The Blight]

Well my machine is still down, so no testing.. I have also come to the conclusion that I would like something better for my MPG. So I now have started working on a buffer.

I'm making it so that its interacting with a keyboard encoder plugged to the computer. This way you can select 4 different jog increments, and the MPG encoder will know what increment you have selected. So if you turn the MPG encoder too fast, it will just put that signal as the next in line. It can hold about +/-100 signals in stack. So if you turn the jog wheel 1 round really fast (100 steps) it will move the machine 100 steps. This means that when you have stopped turning the jog wheel, the machine might still be moving.

You will have to calibrate the controller so that it knows how long it takes to performe the steps you want it to make. I plan on making it operatable with only 5 buttons.