A1 CV Unipolar - Output-Voltage

e8abac78cdf65c56

I am not sure what the aspected Output-Range is for Unipolar - Setup
I have -12 GN +12 as Power-Rails.

The Range which i get is 0 - 8.3V
IN MIOS: AOUT_PinSet(1, 0) and AOUT_PinSet(1, 65535)

Trimpot Turn Left = Higher Voltage…. but 8.3V is the limit (AOUTPINSET 52062 to 65535 is the range where nothing changes anymore… rail clipped? op amps? resistor changes?)

I have 2 ident A1 CV Modules here (booth in Unipolar mode) > and on booth on all channels i have this max of 8.3V

is there something else to adapt for unipular-set except to not use 1R34, 2R35, 3R36, 4R37, 5R38, 6R39, 7R40, 8R41

TRANSMUTE ST O74IY (MZ334135) as opamps
OCTAL: ST O72C (MZ139) opamps

mostly mouser parts, and some Reichelt parts…

17703227d9536da1

Gain should be approximately 5.1 when the pot is in center position (ca. Rfeedback/Rinput).
superDAC reference voltage is 2.048 (do you reach this voltage at max DAC output? Maybe it doesn’t like a 16-bit number as the DAC is only 12-bit?)
Therefore the full-scale output voltage should be approximately 10.4V, which should be in the common-mode range a TL072 can swing at +/-12V.

e8abac78cdf65c56

2,045 i measure directly on the super DACs CV-OUT (when set to max level…) so its not a software or SuperDAC thing (i guess) I noticed that on a point on the Trimmer, it stops to encrease the voltage (turn to left), but there are a couple of turns left to get to the end position of the potentiometer.

17703227d9536da1

You could try to measure the output voltage on pin 8/14 of the op amp, which should be about that maximum (-10.4V). The second op amp stage is a unity-gain inverter with an optional offset of +5V, therefore giving you the bipolar mode.

You can measure the total resistance between the op amp inputs and outputs (might change a bit with other resistors connected in circuit).

e8abac78cdf65c56

see measurments:

e8abac78cdf65c56

anybody see the problem? - i am not into analog circuits, i only can measure.

17703227d9536da1

Blue circuit seems to be working as intended.
Green-brown circuit doesn’t work as intended, but there are not enough voltage/resistance measurements nor pictures of the board provided.

e8abac78cdf65c56

nope a missunderstanding, this are all Measurments of Channel 1, i only spreadet it in the picture.

the blue and darkgreen-brownisch wires are just orientation helps for me (aka “this part is here”)

The Board is fully stuffed except that 1R34, 2R35… “Unipolar to Bipolar exclude Resistors”.

e8abac78cdf65c56

more detail my cam cant give…

17703227d9536da1

Components seem to be correct, same for op amp orientation.
Were the op amps and trim pots obtained from a good source? I have heard of fake op amps…
I can only suggest to try touching up the soldering points and perhaps replacing the op amps.

I just unsoldered 1R34 from my otherwise complete module and I get the expected ₁₀.4V max and proper 1V/octave. I don’t see why it should make a difference if all channels were unipolar.

e8abac78cdf65c56

OP-Amp: orginal directly from mouser, this type:
https://www.mouser.at/ProductDetail/511-TL074IYDT
(the one from the BOM)

on booth Boards, (i have 2× 8CV Boards) with ident Parts, i have nearly ident max Voltages

strange!

e8abac78cdf65c56

when i Encrease 1R42, 2R43… i should have more amplification?
So for first test reasons take a 100K Potentiometer, replace it with 1R42.

let me think… first i need a Sinus LFO on DAC, then a osci on the output of this Module.
Then encrease the Resistance of the Poti, to the point when the Sinus is not a sinus anymore? clipping > Square…

So i can find the right Resistor? > would that be the right way?

17703227d9536da1

The second op amp stage should simply invert the voltage coming into the inverting input (-10.4V) if the resistors are the same value. No idea why the gain is low, but if it is really the same for 16 channels, it points to some systematic error.

As the value of the feedback resistor increases, so should the gain.

https://www.electronics-tutorials.ws/opamp/opamp_2.html

e8abac78cdf65c56

i will try that out in a few days… maybe the idea comes over night,..

…ok this is a unity gain inverting state…
and if 1R26 and 1R42 match (i will desolder and measure them) we should make out of -10,4 > +10,4V i understand.

but there are extra parts in there:
C22 - which should i think do not much with DC or LFO Signal (For what is it ?) as it blocks? and make then a resistance on AC and higher Frequencys?

R50 should do current limiting? or at least protect the circuit? it is directly connect to the 3.5Jacks to the modular, (what case could kill the circuit?)

i cant calculate nor think it but, 1R26 is not equal the combination of 1R42+R50+C22?

i would try: remove R51 + C22 > would that be save, or is the potential dangerous?

17703227d9536da1

There is a filter network on the op amp formed by C22 and the resistors. R50 is within the op amp loop so there shouldn’t be much of a voltage drop. You’re right that it is not exactly “50k” like 1R26, but that’s why you can slightly adjust the gain with VR1 to match it.

Feel free to modify the circuit as you wish. TL07X can survive a short circuit to 0V, so you could remove R50 if you want. I don’t know how much more help I can provide and you have the circuit diagram to work with. Still, if you find the reason why it doesn’t work I am interested in that.

54f37a1928a96a1c

@Phatline - Your measurements - I assume the resistance readings are with the power off? Are the resistance readings the same if you swap the meter leads over (reverse the bias generated by the meter) - this will affect any semiconductors such as the op-amp.
You measure -1.2v on pin 6 of the op-amp CANNOT BE! it’s a ‘virtual earth’ node - the op-amp will do everything in its power to make this match the non-inverting input (which is connected to 0v) CHECK PIN 5 VOLTAGE IS 0v! Then check the voltage between pins 5 and 6 (and 2 and 3 should also be identical at 0v), you should always read 0v difference here as long as the output hasn’t reached a power rail voltage.

I will be starting to build my Transmute8 board tomorrow all being well, I can try making mine all unipolar if it might help but these 50k resistors only add a bias current into the ‘virtual earth’ input on the inverting op-amp.
C22 is just high-frequency stability, to stop the op-amp oscillating (standard configuration) and R50 protects the op-amp output from accidental shorts, current-limiting, if maybe somebody patches two outputs together on the synth.

54f37a1928a96a1c

Some quick calculations around IC1D (and same for IC1C) :-

Gain with Variable resistor at max (50k) = 50k/(130k+100k+50k) = 5.6
Gain with Variable resistor at min (0k) = 50k/(130k+100k+0) = 4.6
So if you can set an input from the DAC (AOUT_1) at +1.0v then you should see between -4.6v and -5.6v at the output of IC1D, pin 14 regardless of where VR1 is set. Anything else is wrong, so that measurement looks OK (-10.4v/2.045v= a gain of 5.09) and the fault must be AFTER that.
IC1A could be adapted by altering 1R26 as you suggest to provide gain but this would be hiding a fault!
If 1R26 was the wrong value (within reason, say 1k to 1Meg) then making 1R42 match it would correct the amp’s gain BUT this does NOT help with that mysterious -1.22v on the input to IC1B! Something is wrong with pin 5 or the OpAmp.

54f37a1928a96a1c

FWIW - An observation I made yesterday …
I have already built D1 to D3 and Eureciever boards with no problems but when I started on this PCB, the Transmute8 I find it is much more difficult to get the solder to whet the PCB pad! Both sides of this PCB are giving me grief - the pads are gold plate and nice and shiny and I only opened the plastic bag a week ago so there is NO WAY it could be corrosion - I wonder if there is some sort of lacquer or other manufacturing residue on this board? Even applying flux pen to the pad doesn’t always help.
I am using 60/40 tin/lead solder with rosin flux and an iron temp of 300C (I have used this since the 1970’s!)
I mention it here because a dry joint could be the cause of @Phatline’s problem.

I still have some way to go with this PCB so will be experimenting a bit during its assembly - if there is some surface coating then I will be using an abrasive pencil to clean each pad.

Work on this PCB will continue tomorrow 😀

17703227d9536da1

Interesting! I think these boards are from 2019, so I wouldn’t rule out some surface oxidation, but normally they are vacuum packed right away. The tip with cleaning or slight abrasion is a good one. I would have expected that ENIG boards are better, as I’ve also seen issues with older boards but they were normally HASL.

A dry joint or missing 0V (ground) connection could explain some things? The fact that all 16 channels don’t work would make individual joint failures unlikely I think.

Have fun with the build! It’s certainly one of the denser PCBs that we do.

e8abac78cdf65c56

i too noticed that it was bad to solder, so i tinned all ic pads before, and removed/aka make the pads flat again with Solder-Remove-Wire, at the end i put on the IC and then it was better solder-able. I then too noticed that it was the same with the Capacitors…. here i thought it was more a thermal mass thing (groundplane)…. but after realiseing that i already heating the pad with my wella station at 550°C and a few seconds… i then too startet to thinning the C and R pads….

that said, i already took care to get a contact… a have time on weekend for further investigations…-