Acorn 10V analog output wonky (Resolved, and not what you thought)

[kb58]

I'm using an Acorn board to run a router, commanded spindle speed via a VFD with the 0-10V analog output.

By itself, commanding the VFD to a spindle speed via its front panel-mounted pot works fine. It was then reconfigured to instead take the speed command from its analog input. Varying that input with a bench supply varies motor speed as expected.

When connected only to a voltmeter, the Acorn 0-10V analog signal appears good. Commanding 20,000 rpm outputs a stable 9.17V, and other speed commands and voltage outputs line up. The problem: When the Acorn analog output is connected to the analog input on the VFD, the signal gets severely loaded down, to 3.0-3.6V, and is very unstable, jumping around within that range. The VFD tries following the unstable command and I can hear it varying up and down in speed.

I've triple-checked everything and don't see what's wrong. The most obvious is that I miswired it, but since when I command a speed, the VFD does respond, it must be going to the right connections. Also, using a variable bench supply to command a speed works fine, which suggests that the VFD is working properly. It's only when the Acorn is connected does the signal get loaded down.

Looking in the Acorn spec sheet, it says the analog output is capable of 1ma minimum and 10ma maximum. I put a 10K resistor on the output to confirm that and it worked fine.

In the long shot chance that someone has had this issue before, this is happening with a Simphoenix E550 VFD.

Thanks

[kb58]

I tried leaving the analog connected but disconnected the FWD line. Surprise, the voltage remained stable at 9.17V, so this may be an EMI issue...

[kb58]

Sure enough, grounding the shield (only) at the control box completely solved it. I admit that I had gotten lax about EMI and it bit me. Strange symptom though, driving down the commanded signal so much. Anyway, thanks for being good listeners, and maybe this will help others.

[cnckeith]

thanks for reporting back.

[Muzzer]

These Chinese VFDs always seem to skip over any requirement for an EMC filter on the AC input, yet they are not magically able to avoid generating significant switching noise during operation. As a matter of good practice, VFDs should be fitted with a decent quality filter, rather than try to figure out and fix problems after they have occurred. Some of the more industrial products either include an EMC filter in the VFD or suggest the part number of a filter that might enable them to meet industrial (Class A) conducted RFI in a typical installation.

[eng199]

Strange symptom though, driving down the commanded signal so much.

The output was probably oscillating. A sine wave would likely display on an oscilloscope. A meter on VDC will just show the average. Once the output is out of control (continuous oscillation) it is hard to say what the DC average will be. Measuring the output on VAC and getting voltage is another way to determine something isn't right.

[Tlempicke]

I had the exact same problem and it was cured by buying a piece of good quality shielded coax rather than the Radio Shack stuff I was using.

[kb58]

Yeah there are a couple things I didn't mention:
1. The VFD user manual does show a filter - but they don't give any specs, making ordering one something of a gamble.
2. During testing, I wondered what would happen if the shield was initially left ungrounded, only grounding it after the spindle speed anomaly started. Interestingly, grounding the shield after the instability started didn't fix the problem. It's as if the VFD was chasing its tail, where once the EMI-induced feedback starts, it's strong enough to get through the shielding, grounded or not.

I'll look into adding an input filter as insurance.

[jarmachines]

I have a Astro CNC Knee mill, induction motor whit 1750 rpm, can I used my Teco VFD to control spindle rpm that is low 0 whit gear and hi 5000 rpm

[kb58]

I have a Astro CNC Knee mill, induction motor whit 1750 rpm, can I used my Teco VFD to control spindle rpm that is low 0 whit gear and hi 5000 rpm

Sounds like the makings of a separate thread!