Ambika SVF card functions as LP filter in all modes

I bought an Ambika SFV card (with chips) all the way back in 2014 from the old MI webstore. Assembled it in 2016 and tested quickly that it produces sound and left it that. This week I had a play with my Ambika and noticed that the filter functions only as a LP filter in all modes. Changing its mode doesn’t mute the card (even for a moment); it just sounds like a LP filter no matter the mode.

Currently I only have two cards installed on my Ambika. The other card (a transistor ladder filter) does react normally when I change the filter mode. When comparing my card to the one which is pictured on the build guide, the only difference I can spot is that my styroflex caps might be installed the other way round, but that shouldn’t matter, right? I’m asking if anyone could point where I should start debugging this problem. I can post pictures of the board if someone wants to take a closer look.

Check pin 9 and 10 of the 4052. Do they change when you select another filter mode?

Using pin 6 as a reference point, I measured the following readings (all in V):

LP
pin9: 1,0
pin10: 0,350

BP
pin9: 1,0
pin10: 0,002

HP
pin9: 0,003
pin10: 0,350

And what if you remove the IC from its socket?

With the IC gone and measuring from the socket the readings are the same.

And what about pins 23 and 24 of the MCU?

Using MCU pin 22 as a reference point, I measured the following readings (all in V coarsely rounded):

LP
pin23: 0
pin24: 0

BP
pin23: -5
pin24: -5

HP
pin23: 5
pin24: 0

Are you sure about this -5V ???

My bad. It appears that I was off by one pin during the last measurements. Here are the new readings:

LP
pin23: 0
pin24: 0

BP
pin23: 5
pin24: 0

HP
pin23: 0
pin24: 5

Were you also off by one pin when you gave the measurements on the 4052 pins?

Just to be sure, I remeasured the 4052 pins, but they remain the same as before. As the wanted pins on the 4052 are right next to the edge of their packaging, it’s more difficult to get them wrong.

Then the problem is with the level shifting circuit achieved by Q1 and Q2. Since both control bits are affected, you probably made the same mistake twice (wrong polarity?).

Could the most likely reason be that I’ve fried both Q1 and Q2 while soldering them? Both R7 and R8 (which are connected between VCC and MOSFET drains) are seeing 8V on their VCC leg, so the only conclusion I can come to is that the MOSFETs are faulty and are not pulling the correct voltage, even though their gates are receiving voltages correctly from the MCU. Was there some other possibility you implied by wrong polarity? My knowledge in electronics is elementary at best, so every tip is appreciated!

I don’t know how a component could be fried by mere soldering.

By polarity I meant soldering them backwards. Or maybe you soldered the wrong part there (eg a BJT transistor)?

Looking at the datasheet of these MOSFETs, maximum exposure to the soldering iron is 10 seconds, so it’s unlikely I’ve caused heat damage in this case. However, from the same datasheet I am noticing that it is indeed starting to look likely that I’ve soldered them backwards…

My next question is that should I try to salvage the two existing MOSFETs or are they as good as gone?

No idea, sorry!