0-10V analog output voltage not constant (Resolved)

[HoY]

So I got my acorn. Upgrading from an ESS with a C82 BOB using mach 4.
Transferred all of my wiring from one unit to the other, got everything functioning properly (so far, still some functionality i have not setup in regards to ATC)
I just wired the vfd analog input exactly the same as I had it wired to the ESS/C82 on the 0-10V output, and it works- sort of.
The voltage seems to fluctuate by up to 0.5 volts. and since I have a breaking resistor on my vfd, every time the voltage drops the brake kicks in and slows the spindle down to the new perceived rpm demand from the software.

I am not including any wizard screenshots, as none of the settings there are relevant to the spindle except that its max speed is 24,000 and min is 0.
I dont know enough about the software YET to know what to look at to troubleshoot this any further.

For the record, the spindle/VFD performed nominally while on the ESS/C82.

I tested it again once more before I submitted this post, just to double check to make sure it is set to not use an encoder, as my spindle does not have one.
I also ran the test with the wires disconnected from the VFD. The voltage remained constant. it only fluctuates while it is connected to the VFD.

Link to the VFD (2.5kw spindle, 4kw vfd)

https://www.solidrop.net/product/sunfar ... motor.html

pictures of the vfd, though I believe they are irrelevant solely because it works as intended on the ESS

https://imgur.com/a/TFJ3Ey2

video of the problem

https://youtu.be/XReYt98WfWs

Thank you for any help you are able to give, and if you need any other information just ask, I can include pretty much anything.

[Gary Campbell]

18 ga twisted pair is recommended connection wire for the 0-10 analog circuit. Is that what you are using? Is it shielded?

[HoY]

18 ga twisted pair is recommended connection wire for the 0-10 analog circuit. Is that what you are using? Is it shielded?

No, I am using Cat 6 solid, i believe it is 22 gauge. Not shielded, but as I said, I did not have this issue with the other board. I have some 18/4 I can try tomorrow to narrow it down, but voltage sag of 0.5V over 6' of wire? it seems like a little bit of a stretch especially since the sag is not constant, but its one more thing to check

[HoY]

18 ga twisted pair is recommended connection wire for the 0-10 analog circuit. Is that what you are using? Is it shielded?

I called it a stretch, but I just did it quickly instead of waiting for morning, and it seems to have worked! unfortunately I'll need to run two of them instead of the single cat 6, but it works anyway. the other board must supply more current to its 0-10v outlet, so the sag was lower.
Also double checked, the cat6 is 23 gauge not 22. Ran fine for a full year on that.

[cncsnw]

I know of no reason you should have to use 18ga. Every installation I know of (Oak, All-in-one DC, GPIO4D and predecessors; not Acorn) uses 22ga or 24ga shielded.

Voltage drop in a wire is proportional to distance and to current flow. An analog reference circuit like this probably never carries more than 5mA. Over the space of two or three feet, there should not be any significant voltage drop with 24ga or even 30ga wire.

Was the wire you just used shielded (as well as twisted)? If so, where did you land the shield drain wire?

[martyscncgarage]

Indeed. I have had occasional issues when the shield was terminated to ground. Most cases grounding one end of the drain wire works fine, in some cases, I had to lift the drain wire from ground. I always run all ground and drain wires back to one common ground point, where the earth ground enters the cabinet.

[HoY]

Was the wire you just used shielded (as well as twisted)? If so, where did you land the shield drain wire?

Yes it is shielded, it’s 4 wire twisted 18g, with foil shield and shield wire. I ended the shield wire with the other shield wires coming from the other devices that are hooked directly to the acorn, like the probe, tool setter, and 3 homing switches. These are not connected to ground yet, as the whole setup is on the bench and not inside of its cabinet. Once in the cabinet they will share the same post as the stepper motors shield as well, and then be grounded from there.

[eng199]

The issue is not voltage drop. Most likely there is an oscillation caused by some sort of noise. This would show up on an oscilloscope, but is averaged out with a meter or the vfd input.

The electrical noise that triggered the oscillation can be very hard to identify. Changing the wire routing or component locations can be enough to change how a noise source interacts with a signal. The difference between the C82 and ACORN output circuits could be a factor. I wouldn't conclude that the ACORN circuit is inferior just from this case. The roles could easily reverse in a different system.

[HoY]

The issue is not voltage drop. Most likely there is an oscillation caused by some sort of noise. This would show up on an oscilloscope, but is averaged out with a meter or the vfd input.

The electrical noise that triggered the oscillation can be very hard to identify. Changing the wire routing or component locations can be enough to change how a noise source interacts with a signal. The difference between the C82 and ACORN output circuits could be a factor. I wouldn't conclude that the ACORN circuit is inferior just from this case. The roles could easily reverse in a different system.

I agree with every point. I also didn't mean to sound like I thought the C82 was a better board, they both have their strengths. I just meant that they are different.

Before I switched to the heavier gauge wire I tried routing the Cat6 across the bench so it wouldn't be near anything else conductive until it got all the way to the vfd, with no change in the issue.

[martyscncgarage]

Did you get this resolved?

I would disconnect the analog wires from Acorn and do this test.
Go into the Wizard, set the max spindle RPM to 1000
Set the min RPM to 0

Connect ONLY a DC volt meter to Acorn Spindle Analog outputs.
Go into MDI
Command M3S100 do you get 1 volt?
Command M3S500 do you get 5 Volts?
Command MCS1000 do you get 10 Volts?

If so, get a 2 conductor cable, wire it from Acorn spindle analog outputs DIRECTLY to you VFD Analog Input Terminals, bypassing your ethernet cable.

Test your VFD again (don't forget to set your min/max spindle speeds back correctly in the Wizard. Write the settings AND RESTART CNC12! You must restart CNC12 for the spindle speed settings to take effect and Acorn to output the correct voltages based on the new speed set.

Let us know how you make out.
Marty