Component C29 is backwards on all TB-303's... (more inside)

[antto]

i already have raw recordings from my VCO, but more are welcome
btw, record an "step" signal thru your device if you can, so i can approximate the DC-filter frequency and "undo" it, this way i'll restore the waveforms from the recordings ;]

Just posted some! Not sure if Audacity can do the step signal you are talking about, and I'm not sure what that is anyway . Let me know what it is, and I'll see if I can figure out a way to get it done. I'm afraid I may have to dig up a newer audio interface for my Mac, it's a 6-year-old dual G5. Great machine, but the on-board audio in is a bit limited by newer standards. You may want to check the signal polarity when you download the file, I haven't seen a DAC yet, at any price, that didn't invert the signal on playback, and very few DACs allow you to flip polarity. Either it'll need to be done in software, or through and inverting hardware buffer, or just switch the black and red connections to your speakers .

when recording with a PC, your audio card usually has a DC-filter (usually a 1st order HP filter at some low-ish frequency)
so the recorded waveforms from the x0x will not look right, especially the lower freqs
but there is a way to fix this, a "reverse 1st order HP filter" .. i used this with my own recordings at it works pretty good!
the only thing is: i need to know the exact frequency of your DC-filter, in my card it's around 16Hz
the "step signal" thing:
you could use some square wave for this.. even the x0x DinSync clock or RUN signal would work
record this and it'll be very clear what the frequency of your DC-filter is, because these signals have "flat" portions.. if you use the clock - be sure to use some very slow tempo ;]

[Brassteacher]

when recording with a PC, your audio card usually has a DC-filter (usually a 1st order HP filter at some low-ish frequency)
so the recorded waveforms from the x0x will not look right, especially the lower freqs
but there is a way to fix this, a "reverse 1st order HP filter" .. i used this with my own recordings at it works pretty good!
the only thing is: i need to know the exact frequency of your DC-filter, in my card it's around 16Hz
the "step signal" thing:
you could use some square wave for this.. even the x0x DinSync clock or RUN signal would work
record this and it'll be very clear what the frequency of your DC-filter is, because these signals have "flat" portions.. if you use the clock - be sure to use some very slow tempo ;]

Ah, Macs don't have audio cards, it's built onto the motherboard. That's why I used the 2µf cap, I wasn't going to take any chances when dealing with a motherboard. I was tempted to open the case and trace the input jack to the first resistor and cap I could find, but the processor heatsinks were in the way, and I didn't feel like shutting the machine down. I'll run the LFO of my SH-09 through the scope to see how clean its square wave is, and try to get a recording straight off the LFO test point. Luckily I already have it opened up for calibration. As an aside, if the TB-303 calibration instructions are as poorly translated and poorly written as the calibration instructions for the SH-09, there are a LOT of screwed-up TB-303s out there!

[antto]

my audio device is also integrated into the motherboard

[Computer Controlled]

So last night i decided to put this in backwards, and what do ya know, it made a difference in the resonance. still not quite there in regards to the 303's resonance, but much closer than it was before.

[Altitude]

So last night i decided to put this in backwards, and what do ya know, it made a difference in the resonance. still not quite there in regards to the 303's resonance, but much closer than it was before.

try using it right side up but with a 33K resistor across it

[Computer Controlled]

Will this throw it in self oscillation?

[antto]

So last night i decided to put this in backwards, and what do ya know, it made a difference in the resonance. still not quite there in regards to the 303's resonance, but much closer than it was before.

try using it right side up but with a 33K resistor across it

"right side up" you mean like per the x0x schematic (original) or reversed (as in 303) ?
and a 33K resistor in parallel with C29 ?

[Altitude]

exactly like the schematic but with a 33k across the + and - of C29 like don suggested. This gave me results almost identical to Brian's tests..

[Computer Controlled]

I'll have to dig around and see if I have a 33k resistor laying around.

[Altitude]

anything close. I actually use a 100k in mine. The higher the resistance, the less it will knock down the peak

[Computer Controlled]

I don't have any, but it looks like they have some at Rat Shack. How much of a difference would it make to keep the cap reversed and throw the resistor across it? I probably should have socketed it.

[Altitude]

I don't have any, but it looks like they have some at Rat Shack. How much of a difference would it make to keep the cap reversed and throw the resistor across it? I probably should have socketed it.

Huge difference. I have a modern cap in mine and when I reversed it, there was very little change. When I switched it back to stock and added a resistor. the difference was dramatic. If you just want to try this, just solder 3 10Ks in series

[Computer Controlled]

Well, i got a 22k and a 10k to make the 33k. Put the cap back to it's original position and put the resistors across C29. It sounds dirty and well, not good. Almost like there's noise being introduced into it now.

[Computer Controlled]

banned, i think i might have killed C29. I reversed it again and took the resistors out. It sound even WORSE now =o] Now i gotta find another capacitor!

[Computer Controlled]

And now it works fine...

No, it's still iffy. It gets all ugly for a few seconds when i let it sit for a bit, then calms down.