I bought a cheap x0xb0x but failed. She had multiple bugg, at least 25 dry bad welds, even the sequencer did not work ... Here i managed to make it work, but there remains a problem with the filter.
The resonnance potentiometer acts as if it were a volume knob, the lowest it completely mutes. I had never seen that ... It strange
VCF failed
Moderators: altitude, adafruit_support_bill, adafruit, phono, hamburgers
Please be positive and constructive with your questions and comments.
- antto
- Posts: 1636
- Joined: Thu Apr 15, 2010 3:21 pm
Re: VCF failed
so when the resonance knob is zero - there's silence
when it's at max, does it sound like the signal is being filtered? is there actual resonance then, can you hear the cutoff frequency being modulated by the envelopes, cutoff knob, etc..?
that's what i suspect
in that case, it would mean that the actual filter is most-likely okay
the TB-303 filter needs a negative feedback gain of 17 to reach self-oscillation, that's crazy high (compared to 4 for a transistor ladder by the book)
with a negative gain of 17, the passband gain at the output of the filter drops TONs
so the roland engineres have added a compensation, another output which only gets amplified when there's resonance
with this, they've compensated it to loose about as much gain as a transistor ladder (not fully compensated)
the output of the filter is from the negative leg of C14
the compensation output comes directly from the whiper on the resonance pot
these two get mixed right before they enter the VCA
the normal output goes via C21 and R121, the compensation goes via C20 and R122
in your case, if what i said above is correct - i suspect there's something wrong with the path of the normal filter output, perhaps after C14, the signal doesn't reach the VCA, so check C21, R121 (their values, and if they're soldered properly)
another possibility is, a short between the joint of C14-C21 and ground (or any other voltage source) - it won't draw current because it's decoupled, but it will mute the audio signal at that point
when it's at max, does it sound like the signal is being filtered? is there actual resonance then, can you hear the cutoff frequency being modulated by the envelopes, cutoff knob, etc..?
that's what i suspect
in that case, it would mean that the actual filter is most-likely okay
the TB-303 filter needs a negative feedback gain of 17 to reach self-oscillation, that's crazy high (compared to 4 for a transistor ladder by the book)
with a negative gain of 17, the passband gain at the output of the filter drops TONs
so the roland engineres have added a compensation, another output which only gets amplified when there's resonance
with this, they've compensated it to loose about as much gain as a transistor ladder (not fully compensated)
the output of the filter is from the negative leg of C14
the compensation output comes directly from the whiper on the resonance pot
these two get mixed right before they enter the VCA
the normal output goes via C21 and R121, the compensation goes via C20 and R122
in your case, if what i said above is correct - i suspect there's something wrong with the path of the normal filter output, perhaps after C14, the signal doesn't reach the VCA, so check C21, R121 (their values, and if they're soldered properly)
another possibility is, a short between the joint of C14-C21 and ground (or any other voltage source) - it won't draw current because it's decoupled, but it will mute the audio signal at that point
- antto
- Posts: 1636
- Joined: Thu Apr 15, 2010 3:21 pm
Re: VCF failed
if you have a scope, follow the audio signal from the output of the filter
on both sides of C14, then on both sides of C21, then R121, see where it vanishes
the signal at the other side of R121 (where it connects with R122) will be very weak and mixed with the other signal from the resonance gain compensation, so have that in mind
if you don't have a scope - you can try the cheap headphones probe as mentioned in the build manual
on both sides of C14, then on both sides of C21, then R121, see where it vanishes
the signal at the other side of R121 (where it connects with R122) will be very weak and mixed with the other signal from the resonance gain compensation, so have that in mind
if you don't have a scope - you can try the cheap headphones probe as mentioned in the build manual
- diplowave
- Posts: 103
- Joined: Wed Aug 21, 2013 9:49 am
Re: VCF failed
Antto thank you very much! I looked and there was actually a bad weld that i had not done again for R121, it took me 5 minutes. The guy who built this x0xb0x was really bad, it took me about 20 hours to get to the repair totally ... I had never seen such a rotten. I thought i would not get out, whenever i found a bug , there was still another.
Now it works.
Now it works.
Please be positive and constructive with your questions and comments.