TM6, 5.33V, and why the TB-303 has so much resonance

[antto]

When i first got my x0xb0x (pre-built by x0xsource) it surprised me that the maximum resonance level wasn't as high as i expected... and i didn't expect it to be as high as josh wink, just high enough that it's "audiable" as a slight oscillation
so i had to do the resonance boost mod (R97) and just reduce the 10K original value to slightly less, ~9.2 was already near-josh-wink level
later, i also did the C29-backwards mod (so now it's the same orientation as on a TB-303, altho i don't think it makes any difference to the feedback level)

but i wonder, what kind of resistor tolerance must there be to reach such resonance level as on many TB-303s
i don't quite believe that's the only reason the 303s have more reso than a strict-spec x0xb0x

so now i know that TM6, more specifically the "5.33V" voltage affects not just the VCO tuning, but also the resonance, and so it's much easier to believe that those TB-303s were just tuned in such a way that gives more resonance

i had my 5.33V tuned to exactly 5.33V before
and now i tested - TM6 gives me a range between 5.15 to 5.54V
the higher the voltage - the more resonance on the filter (and it seems it slightly pushes the filter cutoff frequency, which can be compensated via TM3)
Note: this affects mostly the VCO Tuning, and i suspect also the Square Wave, thus the VCO must be retuned

my x0x, at 5.54V and resonance mod disabled (R97 = 10K) still doesn't give enough resonance, but at least it's slightly more than the original level as when i recieved the x0xb0x
i estimate, i'd need something like ~5.8V to reach a good resonance level

i'm curious, since i know the main difference in the x0xb0x vs the TB-303 is in the power-supply section
What is the range of the "5.33V" on the TB-303s

in any case, i want to know how to make the range of my "5.33V" voltage wider
i wanna see how this affects the square wave, because i suspect higher voltage == more reso AND square wave sounding a bit more like the 303 (since you know all strict-spec x0xb0xes have the same square sound which is slightly different than many 303s)

[aminoacid]

hi antto!

when talking about the filter, the very main difference between the 303 and the xox is in the film caps to decide the character and feedback behavior.

i must say so since i have tried 5 different caps in the ladder, including original transparent green caps which im using now (i have two other brands to try out some day).

i think that LCR differncies in the caps in combination of the circuits tuning is responsible for theese drastic changes.

i think using another cap than the original, one could tune some bias circuits and achieve a similar behavior as in the 303.

i have experimented some with the circuits around q18 and q20. just by selecting differnet hfe on thoose makes sound difference.

q9 and its bias is interesting aswell!

[antto]

correction: C29 reversed (like it is on the TB-303) actually reduces the resonance even more
so there must be something adding more resonance on the TB-303

correction2: it also seems that the 5.33V value affects the VCF Envelope, which cannot be compensated with TM3

aminoacid: which caps exactly? 18 19 24 and 26?
in my x0x those are green with labels:
2A183KT (C18)
2A333K TRACON (the other 3)

i found TB-303 photos, it's very hard to see the labels since the components are very tight there, but i think i saw C24 which looks like mine (green)

[aminoacid]

there are 4 points in the vcf where 5.333v is entering.

when changing the 5.333V you are on all theese places at the same time changing thoose circuits.


i also reveresed my c29. i cant remember actually what happend but i thought it was more funky


im talking about all film caps in the vcf but thoose 18nf and 33nf are the once that sets the character the most.

polystyrene caps sounded very good but they made the resonance to much feedbacking.

thoose original transparent green caps are long gone out of production.


if you are going to experiment further with the circuits, pplease be kind and report!

[antto]

i'm not very familiar with caps, i only distinguish between electrolytic and <non-electrolytic>
so "polystyrene" increase the feedback? IDEAL candidate

imma be visiting the electronics shop, to get a new soldering iron (i ruined the tip on this one) and i'll see if they got such caps there

C29 backwards (as on TB-303) reduces the feedback a little

i see the 5.33V also used in the Envelope section, so maybe it determines the "top" point of the cutoff envelope (when it retriggers) but i'm really not sure

[MMM]

I'd leave the 5.33V as they are because changing this value messes up the external CV as well (the 5.33V were picked because they result in a 1V/oct behaviour; retuning the VCO will NOT change this CV problem).

I'd look at the circuitry surrounding Q18 again; perhaps playing a bit with different values for R99 and R66?

[antto]

good point, i haven't used the CV out, cuz i don't have any other toys that have CV-in

R99 R66 .. hm.. i guess tweaking those would be similar to the reso-boost mod (which is to reduce the value of R97)

i am trying to figure out why TB-303s seem to have slightly more resonance than strict-spec x0xes
the usual suspects are component tolerances and age, i know many x0x builders might have used 1% resistors all over, while the TB-303 was built with cheap components, probably 5% resistors mostly..

my theory was that maybe they were tweaked with high "5.33V" value (via TM6) and thus more reso and more-strange square wave
but as you now pointed out, that would mean their CV-out would be way off, and i guess that's not quite the case? or is it?

ideally, some volunteers can open up their 303s, measure their 5.33V and say a word or two about the resonance level (or record a small sample)

i found some caps, 0.033uF and 0.015uF (they didn't had 0.018) but socketing those caps in the VCF will be a bit tough for me, as the area there is quite crowded.. i'll see

[aminoacid]

tolerance of the resistors are 5% in the 303.

transistor tolerance does some aswell.

polystyrene are abit oldschool nowadays so they can be hard to find in larger capacitance. i got mine from surplus of nebraska, i think they were called that...

0,1uf i had to get from ebay..!

[aminoacid]

oh yea..! they are often bigger.

i had to mount some on the other side of the pcb!

[antto]

the ones i got are yellow-plastic box-shaped with their values printed on the top
their dimensions look identical to the green ones in my x0x, so they gotta fit
i'm more concerned about how i'll manage to solder sockets there.. oh well

[Altitude]

.. i got mine from surplus of nebraska, i think they were called that...

0,1uf i had to get from ebay..!

Surplus Sales of Nebraska, but quite down about that place, dont need everyone raiding it for goodies..

[aminoacid]

Xcuzie Moi

[gurth]

Is there any news about why the 303 has more resonance?

Am checking this topic 6 times a day.....

[antto]

i haven't socketed my caps in the ladder yet, since i am very busy doing other stuff and i need the x0x running "stable" ..
i'm always on the hunt for 303 myths and trying to put them to the test
there's a myth that the resonance on the 303 is somehow "more magical" than any other clone (including the x0xb0x) due to the powersupply
that's what i'm looking forward to test these days (waiting for someone that claims so, to record me a sample from a 303 which demonstrates such magical resonance...)