Hello Guest it is January 28, 2020, 04:41:03 PM

Author Topic: csmio/ip-a + Mach 3 servo problems  (Read 285 times)

0 Members and 1 Guest are viewing this topic.

csmio/ip-a + Mach 3 servo problems
« on: December 17, 2019, 01:16:02 PM »
I decided to upgrade my Bridgeport Interact 1 mk2 to a more modern control and chose a csmio/ip-a controller.
I already have a Mach3 licence so it made sense to go that way.
Things were going to plan up until I wanted to test the three axes.
I have changed the old sinusoidal encoders on all three axes to AMT103 units with differential drivers and they all work fine, giving correct readouts per mm of movement.
The servo +/- 10 volt inputs are fine when unconnected and just sit there dithering slightly with tiny interference voltages. However, when I connect them to the csmio/ip-a and power up, immediately each of the axes makes a bolt for one end of travel and causes an ePID fault warning.
I have played about with all the settings and still nothing. Just full speed movement for a very short time before trip out.
To try another system, I have got a 30 day licence for CS Labs own software but sadly the same thing still happens with that.
I know that others have done similar to me and wonder what I am doing wrong please?

Les
Re: csmio/ip-a + Mach 3 servo problems
« Reply #1 on: December 17, 2019, 05:02:41 PM »
Hi,
sounds to me like you have the encoder reading 'upwards' in one direction whereas the servo rotates 'downwards'

Try reversing the main pos and neg power feeds to the servo.

Craig
My wife left with my best friend...
     and I miss him!
Re: csmio/ip-a + Mach 3 servo problems
« Reply #2 on: December 18, 2019, 11:41:41 AM »
That is one thing that I have not tried.
Unless it would be the same as reversing the encoder direction in the software?
Which only makes the test axis shuttle violently back and forth until it trips.
Re: csmio/ip-a + Mach 3 servo problems
« Reply #3 on: December 18, 2019, 11:49:44 AM »
Hi,
reversing the servo direction OR reversing the encoder direction have the same effect.

If the servo moves back and forth.....then its trying to do the right thing whereas if it moves violently to one
end of an axis without attempting to reverse, that is a symptom of when the ecoder and servo directions are
crossed.

It sounds to me like you have had the correct directions when the servo moved back and forth but
you need to tune it. Try to start with small or no I or D, just P alone. I and D components can very
quickly cause instability.

Craig
My wife left with my best friend...
     and I miss him!
Re: csmio/ip-a + Mach 3 servo problems
« Reply #4 on: December 18, 2019, 12:20:49 PM »
Thanks,
I'll give that a go in the morning.

Les
Re: csmio/ip-a + Mach 3 servo problems
« Reply #5 on: December 19, 2019, 12:49:20 PM »
I went back to basics today. I disconnected the x axis and tried it with a 1.5 volt battery on the analogue input, the cable sheath earth was joined to the machine earth.
It works fine. That led me to think about the connections to the csmio/ip-a board. I tested the ground terminals on the analogue output connector and found they didn't have a circuit to 'real' earth.
So, i disconnected all the cable screens from the terminal block and connected them to machine earth.
As if by magic, all now works fine. Each axis was automatically tuned with the plug in tuner and everything works fine now.

Tomorrow I try to make the spindle work......

Les