Author Topic: Ibanez ST-9 Super Tube Screamer (STL) - any way to get more midrange boost?  (Read 5709 times)

aion

Schematic (note that some of the pots are drawn incorrectly, but everything else is correct)

The Ibanez ST-9 has a sweepable midrange boost that comes before the drive section. (or after it, on the STL - otherwise it's the same circuit) The manual for the ST9 says it's a 25dB gain. The manual for the STL omits the gain from the description.

However, building one of these (and comparing mine against some demo videos to make sure that mine is accurate), there's definitely not 25dB happening. I read someplace else that it's an 8dB boost, which sounds much more along the lines of what I was hearing.

That said: it's a little subtle, and I'm wondering how might a person go about increasing the gain? How did Ibanez calculate 25dB, how did someone else calculate 8dB, are either of them correct, and can I change it so the effect is more drastic?

I did substitute the impossible-to-find 20kC dual pot for a 100kC, and in the process I multiplied the resistor values by 5 (1k5 -> 7k5) and cut the capacitor values by 5 (68n -> 12n, 27n -> 5n6) to maintain the frequencies. It works great. But I also tested it with the original values and a dual 20kA wired backwards and it sounded exactly the same, so as far as I can tell the gain isn't a factor of the actual resistance in the op amp feedback. If that was the case then I'd expect that with more resistance (100k vs 20k) the gain of the mid boost would go up.
Aion Electronics - DIY PCB projects for classic and obscure circuits

ashcat_lt

The height of the peak is determined by the difference between the cap values.  Make them more different, and you get more boost.  Make them different by the same factor and the center frequency stays the same.  Course, the Q gets wider at the same time, but if you want more boost without a wider bandpass, I think you need more poles.  The simulation shows that 6.8nf and 270nf give a little more than 25db, with the sweep covering just about the same range as with the values shown.

J0K3RX

Maybe?
Doesn't matter what you did to get it... If it sounds good, then it is good!

bool

For a "pre-distortion mid-boost" I favored the "other" variation on a T-filter, similar snippet (like ROG use in their new design). Imho a slightly better vibe, but thats only me. It's easier imho to get the in-your-faceness with some preserved "teeth" on top, like a bit better pick/transient projection.

Not to mention being able to use a simple, cheap single pot. Hey, whats not to like ..


aion

The height of the peak is determined by the difference between the cap values.  Make them more different, and you get more boost.  Make them different by the same factor and the center frequency stays the same.  Course, the Q gets wider at the same time, but if you want more boost without a wider bandpass, I think you need more poles.  The simulation shows that 6.8nf and 270nf give a little more than 25db, with the sweep covering just about the same range as with the values shown.

Very cool, thanks. I know the Pearl OD-05 uses 4n7/68n and it's labeled a 15dB boost, so I'll experiment with that combination as well - maybe a halfway point between the ST-9's actual and claimed boost levels without getting the Q too wide.

What's the name of this type of filter? Is it just a bandpass with adjustable center frequency, or is there a more specific name? I'd be interested in reading more about it... this and the Pearl OD-05 have caught my attention lately and they both have the same circuit block.
Aion Electronics - DIY PCB projects for classic and obscure circuits

bool

With the original values, a 120nF/15nF cap combo would get you a very similar frequencies range and a circa 14dB boost. Imho a too high Q doesn't sound that good at lower frequencies and also too much boost (dBs) at lower-mid freqs isn't what I'd call a "good OD tone" ...

Maybe for an "effect"..

Nordskov

Im sorry to hijack this thread for a second.

Im about to order parts for the ST9 and I have an issue.
In this schematic the pots have the sufixes A, B, C and G.
http://elektrotanya.com/PREVIEWS/63463243/23432455/ibanez/ibanez_st9_sch.pdf_1.png

According to the TS9 I guess 500k Drive and 100k Tone are both lin. 100k Level is log.
But what kind is the 20k dual Mid Boost?

Thanks

________

Thread released...   ;)
« Last Edit: March 05, 2015, 05:55:06 AM by Nordskov »

aion

It's a "C" taper, reverse audio. The dual 20kC pot will be impossible to find. Your best options are to get dual 20kA's from Tayda and wire them backwards (the knob will work in reverse), or you can just substitute the surrounding resistors and capacitors and adapt the circuit for a dual 100kC pot which is much easier to find (it's the same value as used in the Univibe). More details in the build docs here:

https://aionelectronics.com/product/cirrus-ibanez-st9-super-tube-screamer/
Aion Electronics - DIY PCB projects for classic and obscure circuits

duck_arse

the st9 tone pot shows as "G", not B. G is I think the 'S' curve, no? see RG's secret life of pots page. and you could use a dual 20kA pot wired backwards, call it "mid cut" instead.
"Decrease the bracken!" - Bobo Gargle

Nordskov

Thanks for a swift reply.

This must be my lucky day.
http://www.aliexpress.com/item/C50K-Potentiometer-Pots-Reverse-Log-Taper-S-Shaft-20mm-Dual-6pins-5-pcs-per-Lot/471531173.html  :D
I have to buy 5 pcs, but hey, its better than fooling around with the wrong type.
Its 50 kOhms but according to Aion its just a tweak.
Would you know how to recalculate the circuit to fit the new value?

Another thing. On the TS9 the bypass is done via transistors. Theres no such circuit here, so I guess its done at the footswitch, right?

EDIT: Theres a lot of solutions going through my mind right now. Would it be possible to buy 2 mono C20k and 1 stereo of the same make and type and assemble a custom stereo C20k?
Is the taper mirrored from top to bottom in a stereo pot, or is it reversed?
« Last Edit: March 06, 2015, 02:45:35 PM by Nordskov »

duck_arse

there is a pair of fets on the right of the hendrik's circuit for the st9 which do the bypassing ("effect/bypass"). there is usually a transistor flip-flop to drive them, it seems to have been left off.

as for the calculations, I should know, but don't. if you read the Keen page on pots (and his page on the tube screamer), you'll see you can taper a higher value lin pot to reverse log, so if aion or someone who knows can't help, you might be better getting a lin pot and tapering it. or just use those C50k as is.
"Decrease the bracken!" - Bobo Gargle