Hey ch,

That's an interesting issue, especially since KLANGMASCHINE works perfectly in your Intellijel case.

Both FM voices use separate but identical microcontrollers, so I'm wondering whether the right-hand oscillator might have trouble starting up with the Arturia power supply.

Could you try a few things?

1. Test the individual outputs: Is the right FM voice completely silent at its own OUT jack, or is it only missing from the MIX OUT? Please use the same patch cable and mixer input for both tests.

2. Test KLANGMASCHINE alone: If possible, disconnect the other modules from one of your Arturia racks and see whether the problem persists. This would help rule out a power supply load issue.

3. Try another power cable: Have you used the same ribbon cable in both cases? If so, please try another one and a different power header in the Arturia rack.

4. Check the startup behaviour: Does the right oscillator ever start working after switching the Arturia rack off and on again?
   

KLANGMASCHINE generates its own 5V supply from the +12V rail, so differences in the power supply or its startup behaviour might be relevant.

Let me know what you find, and we can narrow down the cause.

Cheers,

Stefan

Hey Brian, The area around CAL_SHAPE_2A definitely looks damaged. Good to know you've already checked the power header for shorts.

CAL_SHAPE_2A is a 10k trimmer connected between +12V and -12V, so it should not get hot under normal operating conditions.

I'd suggest checking a few more things:

1. Check the resistance across the two outer terminals of CAL_SHAPE_2A. It should be approximately 10k. Compare it with CAL_SHAPE_1A, which uses an identical trimmer arrangement.

2. I'd also recommend thoroughly cleaning the PCB with isopropyl alcohol and a soft brush to remove any flux residue or solder debris, especially around the power header and CAL_SHAPE_2A. Avoid getting IPA inside the pots, sliders, and trimmers, and let everything dry completely before further testing.

3. I also noticed something suspicious around D16/D17. It looks like a scratch or possibly an unintended solder connection. Could you inspect that area closely and check for damaged traces or shorts with a multimeter?
   

Since CAL_SHAPE_2A appears to have suffered significant heat damage, it will most likely need replacing. The PCB underneath should also be inspected for damaged traces or carbonized material. After desoldering, check for shorts again! I might be a connection underneath the trimmer.

Please don't power up the module again until we've identified the cause.

Cheers,

Stefan

I actually figured out the issue. The schottky diode on the negative voltage rail was not fully soldered. Fixed that and now everything works perfectly. Weird things happen when you don't have negative voltage

Hey Anthony, There is no fixed turning direction. It depends on the current offset. The goal is to get the output as close to 0V as possible when the envelope is inactive.

Use a multimeter to measure the OUT voltage and carefully turn the corresponding trimmer in whichever direction brings it closer to 0V.

I recommend a small, non-conductive flat-head trimmer screwdriver to avoid damaging the trimmer or PCB. Cheers,

Stefan

Hey Jeremy,

That’s quite an old one, so some of the links and components may be outdated, discontinued, or no longer in stock.

Cheers,

Stefan

Hi Rich,

the part you need is LF398S8#PBF, the 8-pin SOIC version. Farnell UK, RS Components and DigiKey UK all list this part, although current stock can change.

Best,

Stefan

Hey Nys, I would start with the VCA/main board:

• Reflow suspicious solder joints, start with the ICs.

• Check the board connections with the headers

• Check the supply rails: +12 V / -12 V at TL074 and LM13700 (U2), and the +10 V reference around LM4040 (IC6) and U1 (Pin 1).

• With no audio connected, measure VCA_OUT DC (lower pad of R31) voltage. It should be close to 0 V. If it is strongly positive, negative or near a rail, check U1B, R31, R32 and LM13700 (U2), especialy for solder bridges or bad joints.

• Apply a gate and measure CV_TO_VCA (left pad of R15). The voltage should clearly change with the envelope. If it does, continue through U1C → Q2 → LM13700 pin 1. This is the VCA control-current path.

The constant yellow LED is especially interesting because its driver monitors VCA_OUT. So I would measure the DC voltage at VCA_OUT first. If you send me that voltage plus the voltage at CV_TO_VCA with and without a gate, we can narrow it down quickly. Thanks,

Stefan

Thank you for the schematic. Instanteously realized that I fundamentally misunderstood the PW jack is an output not a CV input, and the Pulse Width pot affects that output.

Did not RTFM to see it has three outputs. ugh.