O_C Phaserville 1.7 on Plum Audio OCP - Attenuverters and voltage outputs are off

[GabrielCrecan]

Hi,
I'm running o_C-phazerville-v1.7_phz+VOR-6d32547c.hex on a Plum Audio OCP and have just calibrated the module, but when using any applet, the values I get from the outputs are off by ±0.1v to ±0.5v. Is this expected due to hardware limitations ?
Another issue is related to the attenuators; If using the 'Scope' applet and setting an attentuator to full CW position, the output voltage is 4.92 (and -3.25 CCW to +4.87 CW when bi/uni button is depressed, so in Biplolar mode)

Any help in understanding this would be appreciated,
Thank you

[djphazer]

It kinda sounds like it's not properly calibrated... When calibrating the CV inputs with 1V and 3V, make sure the attenuators are open all the way. Make sure they are closed for the 0V point.

Double-check the Bipolar Vbias calibration step to ensure proper DAC offset for the ±5V mode (Hemisphere default)

All that being said, there is a known issue that can cause a slight discrepancy between what you measure in the calibration routine and in normal operation, but in my experience, it's only about ±0.01V

I plan to overhaul the calibration routine eventually to make it more intuitive and potentially more accurate.

[GabrielCrecan]

Thank you for the reply.
Just a quick clarification:

• 'make sure attenuators are open all the way. Make sure they are closed for the 0V point.'

You mean they need to be set to Unipolar ?
=>
all the way closed = 0v,
all the way open = 5v

I presume that when they are in bipolar mode, fully closed means -5v and fully open +5v

[djphazer]

Ah yeah, Unipolar mode. I was just throwing out a couple of ideas... I usually leave the attenuators in Unipolar mode because that's the only way I can be sure they're set to 0 volts (full CCW)

Are you using a precision multimeter to calibrate the DAC? Also, is this a DIY build? or assembled by Plum Audio?

[GabrielCrecan]

Got it, thank you.
I'm using the Disting EX's 'Show input voltages' feature from Settings -> Calibration.
Unit is assembled by Plum Audio.

Interestingly, when calibrating OCP and reaching the 0 V bipolar and asym steps, the 4 outputs have slightly different values A and C are around -0.5, while B and D are close to 0

[djphazer]

Did you find a resolution to this?
I was wondering if things work properly in Unipolar (0-10V) range. If calibration is good for Unipolar, but outputs are inconsistent in Asym or Bipolar, I'd suspect a problem with the Vbias circuit.

[GabrielCrecan]

Hey, thanks for following up !

No, definitely several things are wrong. Did the following tests on the outputs (after re-calibrating and checking the that calibration was not lost):

1. AttenOff / Scope
   Attenuverters: Unipolar & fully CCW
   Output range:
   5v to +5v: A=-0.05, B=-0.01, C=-0.03, D=0.00
   3v to +7v: A=-0.03, B=-0.01, C=-0.02, D=0.00
   0v to +10v: All outputs = 0v

2. AttenOff / Scope
   Attenuverters: Unipolar & fully CW
   Output range:
   5v to +5v: A=4.94, B=4.98, C=4.97, D=5.00
   3v to +7v: A=5.02, B=5.05, C=5.01, D=5.07
   0v to +10v: A=5.05, B=5.05, C=5.02, D=5.07

Test cases were done using 'AttenOff' (with the default settings) in Hemisphere
(Double checked using 'Scope' and got same output values)

Interestingly, got slightly different values while testing the same cases in with the Calibr8or app (also with the default settings):

1. Calibr8or
   Attenuverters: Unipolar & fully CCW
   Output Range:
   5v to +5v: same (A=-0.05, B=-0.01, C=-0.03, D=0.00)
   3v to +7v: same (A=-0.03, B=-0.01, C=-0.02, D=0.00)
   0v to +10v: same (All outputs = 0v)

2. Calibr8or
   Attenuverters: Unipolar & fully CW
   Output Range:
   5v to +5v: same (A=4.94, B=4.98, C=4.97, D=5.00)
   3v to +7v: A=5.01, B=5.07, C=4.98, D=5.09
   0v to +10v: A=5.05, B=5.08, C=5.00, D=5.08

I'm not even going to dive into what happens with the attenuverters set to Bipolar mode, but will mention that the 'Scope' applet displays:

Left Hemisphere:
CCW: -3.30v
CW: +5.02v

Right Hemisphere:
CCW: -3.32v
CW: +4.97v CW

I understand some could see this as nitpicking, but honestly it's driving me nuts.
Perhaps there's no point in seeking more precision than the above values ?

[djphazer]

Hey, sorry for such delayed response... the variance does seem to be more than I'd expect, but if we're just nitpicking the max voltage that's normalled to the attenuverters, I wouldn't be surprised if it's not exactly 5V. Looks like some of mine max out at 5.02V... unless my ADC calibration is a little off.

I would verify ADC and DAC calibration independently, using Scope with external voltage source input, and Voltage for outputs to measure.