Soul Sonic - Hunny Bunny, sweet overdrive

IvIark · Post by **IvIark** » 17 Jan 2009, 01:29

Futurlec has a LM301N and LM301H, does anyone know the difference and would either do? There's no datasheet attached to them on the site.

soulsonic · Post by **soulsonic** » 17 Jan 2009, 02:28

IvIark wrote:Futurlec has a LM301N and LM301H, does anyone know the difference and would either do? There's no datasheet attached to them on the site.

LM301H is the metal can version. I don't think it's worth spending the extra money on unless you want serious mojo points.

nooneknows · Post by **nooneknows** » 17 Jan 2009, 08:38

soulsonic wrote:You could also try increasing the size of the compensation cap to roll off more highs.

Are you talking about the cap over pin 1 and 8 of the opamp, right? I have to try it
Anyway I think most of the shrill comes from the mosfet clipping, it already happened to me playing around with overdriven Mosfet, my amp is a bit shrill too so the two effects can combine toghether.

Thank you Martin, great design.

ciao
Marcello

soulsonic · Post by **soulsonic** » 17 Jan 2009, 09:29

Yes, the cap between pin 1 & 8, you could try going as high as possibly even 1000pf....

It might not sound good, but it won't hurt anything to try.
You could also maybe try increasing the value of the resistor in series with the gate of the MOSFET (the 1K). Maybe you could try a pot in place of that resistor? I would try a 50K or 100K and see what happens.

The way the gain of the opamp is set emphasizes the highs. The 1k/150nF combination is what boosts the high-frequency gain. You could try lowering the gain here and see what happens. Maybe try a 3.3k/47nF combo...

WhiteKeyHole · Post by **WhiteKeyHole** » 17 Jan 2009, 20:37

I went with the classic dual low-pass filters on the output to cure the shrillness of mine. I ended up with 10K/2.2nf + 12K/3.3nF.

moltenmetalburn · Post by **moltenmetalburn** » 18 Jan 2009, 01:50

I went with the classic dual low-pass filters on the output to cure the shrillness of mine. I ended up with 10K/2.2nf + 12K/3.3nF.

not familiar with this, why two?

nooneknows · Post by **nooneknows** » 18 Jan 2009, 20:43

it's a common trick to tame down the high 'shrill' after a distortion stage, take a look at thisfor example.

moltenmetalburn · Post by **moltenmetalburn** » 18 Jan 2009, 23:38

it's a common trick to tame down the high 'shrill' after a distortion stage, take a look at thisfor example.

Thanks but I was really wondering why it took two rather than one. looking for technical explanation.

soulsonic · Post by **soulsonic** » 19 Jan 2009, 00:52

moltenmetalburn wrote:
it's a common trick to tame down the high 'shrill' after a distortion stage, take a look at thisfor example.
Thanks but I was really wondering why it took two rather than one. looking for technical explanation.

Using two raises the order of the filter and gives a steeper "knee" to the cutoff.
Sometimes using a higher order filter at a higher frequency can sound better than a lower order one at a lower frequency... if that explanation makes any sense...
Let's say you're using a simple single stage one, and you find that you have to have the frequency set pretty low to cut the shrillness out, but you find that the highs get too dead sounding; you may find that using a higher order filter with a steeper knee will give you the shrill removal you desire, but you don't need to set it to as low of a frequency to get the effect, so the highs aren't dulled so much.... that's just one hypothetical example and by no means consider that a hard n' fast rule, but hopefully that will give you an idea of what I'm talking about.

moltenmetalburn · Post by **moltenmetalburn** » 19 Jan 2009, 00:55

Using two raises the order of the filter and gives a steeper "knee" to the cutoff.

awesome thanks soulsonic! I fully understand.

Mr. G. · Post by **Mr. G.** » 20 Jan 2009, 12:33

After recently messing around with the BOR and BSIAB, I was wondering this exact same thing. Thanks for the explanation!

briggs · Post by **briggs** » 20 Jan 2009, 12:38

A single order filter will drop frequencies at 3dB around the corner freq, double it up and it'll drop 'em at 6bB. As soul said, it's a more 'dramatic' roll off of the unwanted frequencies.

moltenmetalburn · Post by **moltenmetalburn** » 20 Jan 2009, 16:04

10K/2.2nf + 12K/3.3nF

This is the combination he said he wound up with, the two sets have different values, what is the reason for this? is this the typical usage with different values or was it his tweaking that led to this? How do you go about calculating the values? (in pairs)

briggs · Post by **briggs** » 20 Jan 2009, 21:26

Ok, the 10k and 2.2nF filter rolls of highs above 7234Hz, the 12k & 3.3nF filter rolls off at 4019Hz; Using the formula F= 1/(2Pi*R*C) with R in ohms and C in Farads. So the freqs above 7234Hz are rolled off at a 6dB knee (due to the overlap in filters) while those above 4019Hz are rolled off at a 3dB knee. Everything below 4019Hz runs through unaffected.......

Or that is how I calculate! And I'm just back from the pub

Fuzzer · Post by **Fuzzer** » 20 Jan 2009, 22:02

Those calculations might be right my friends, but if I remember correctly, you have to buffer the filters if you want them to act as you expect.

moltenmetalburn · Post by **moltenmetalburn** » 20 Jan 2009, 22:33

Ok, the 10k and 2.2nF filter rolls of highs above 7234Hz, the 12k & 3.3nF filter rolls off at 4019Hz; Using the formula F= 1/(2Pi*R*C) with R in ohms and C in Farads. So the freqs above 7234Hz are rolled off at a 6dB knee (due to the overlap in filters) while those above 4019Hz are rolled off at a 3dB knee. Everything below 4019Hz runs through unaffected.......

Or that is how I calculate! And I'm just back from the pub

thanks for the great explanation, And for leading me here, it was you after all that introduced me to this site and I am completely addicted!

nooneknows · Post by **nooneknows** » 20 Jan 2009, 22:40

Fuzzer wrote:Those calculations might be right my friends, but if I remember correctly, you have to buffer the filters if you want them to act as you expect.

It seems you're right, like in the central picture here

Entrant_21 · Post by **Entrant_21** » 21 Jan 2009, 00:18

looks great, I love the RAT style frequency dumping in the op feedback loop. nice touch

Fuzzer · Post by **Fuzzer** » 21 Jan 2009, 04:24

yes man, that's what I was talking about, thank you very much.

RnFR · Post by **RnFR** » 21 Jan 2009, 08:27

nooneknows wrote:
Fuzzer wrote:Those calculations might be right my friends, but if I remember correctly, you have to buffer the filters if you want them to act as you expect.
It seems you're right, like in the central picture here

great link! i'm definitely going to be spending some time there.