Beavis Board - Boutique Tube Screamer Issues

[digitalzombie]

I'm trying to build the basic Boutique Tube Screamer from the Beavis Board projects since I'm unfamiliar IC-based projects, and even following simple directions I've hit a roadblock.

Here's what I'm following:


I swear I have everything wired up correctly, yet my DRIVE control seem to be working in reverse, and only have a very mild effect. The volume control works fine albeit only getting slightly above unity fully cranked.

It might be the chip? I only have NJM4558's from Tayda, but as far as I can tell from the data sheet those should work. Like I mentioned though, I'm completely inexperienced with IC-based drives, so I wasn't sure where to look, or even where to probe! I at least took voltages, but I'm not sure what to look for:
9.45Vin
1. 4.77
2. 4.77
3. 4.54
4. 0.25~0.3 (fluctuating. Shouldn't this be 0.0?)
5. 4.77
6. 4.9
7. 4.89
8. 9.39

Tried a second 4558.
1. 4.75
2. 4.75
3. 4.52
4. 0.057
5. 4.73
6. 4.73
7. 4.73
8. 9.37

Here's my board.

[Phoenix]

For a start, that schematic is missing a Vref bypass capacitor, so your AC gain is going to be lower than intended, as set by R4 in series with the parallel combination of R1 and R2, or 54k7, which will give you a maximum gain of 11 instead of the design goal of 117. See AMZ Son of Screamer for the missing capacitor. With the values of R1 and R2 you're using, 10uF would be fine. Otherwise, it's likely you've wired your gain pot backwards if that's how it is currently functioning.

BTW, if you include width=800 inside the first img tag bracket, people can actually view your whole image.

[electrosonic]

i think the drive pot is wired wrong in the diagram. When turned fully clockwise drive will be at the minimum value.

Andrew.

[LightSoundGeometry]

if you include width=800 inside the first img tag bracket, people can actually view your whole image.

nice to knw as my images are usually large but i want people to see it all ..will do this though

[duck_arse]

+1 on the width=800 - please.

+1 on the pot is drawn wired backwards.

+1 on my addition, which is now 4. @digizom - when you are doing voltage measures, short the red and black probes together, see what it reads. 0V00 is good. if you then probe around your circuit earths and get something else .....

4. 0.25~0.3 (fluctuating. Shouldn't this be 0.0?)

.... power off, switch to the ohms range, and start probing those points again. (short the probes on ohms range again, so you know what zero ohms looks like on your meter.)

[digitalzombie]

Thanks guys. I got distracted for a couple weeks but I'm coming back to this.

For the drive pot, we're all in agreement that it's wired backward in the diagram which is most evident in the fact that it works in the reverse fashion that it should, but honestly if I were going off just the schematic then I'd probably wire it up the same way Dano did in the diagram. Looking at a pot on a schem displayed horizontally I'd naturally assuming the corresponding lugs from left to right were 3-2-1. Is there some other rule about that? How did you all look at the diagram and know it's wired backward?

For a start, that schematic is missing a Vref bypass capacitor, so your AC gain is going to be lower than intended, as set by R4 in series with the parallel combination of R1 and R2, or 54k7, which will give you a maximum gain of 11 instead of the design goal of 117. See AMZ Son of Screamer for the missing capacitor. With the values of R1 and R2 you're using, 10uF would be fine. Otherwise, it's likely you've wired your gain pot backwards if that's how it is currently functioning.

Is this the bypass capacitor you're referring to?

Are you suggesting that since I'm using 100k resistors for my voltage divider instead of the 10k's on the SoS circuit that I should use a 10uf Cap instead of the 47uf? I wonder why any of this is relevant.

[Phoenix]

For the drive pot, we're all in agreement that it's wired backward in the diagram which is most evident in the fact that it works in the reverse fashion that it should, but honestly if I were going off just the schematic then I'd probably wire it up the same way Dano did in the diagram. Looking at a pot on a schem displayed horizontally I'd naturally assuming the corresponding lugs from left to right were 3-2-1. Is there some other rule about that? How did you all look at the diagram and know it's wired backward?

There is no rule like that. Better schematics will indicate which side of the potentiometer is clockwise so that it is explicit, but a lot of schematics leave it off because the direction is implied by the operation of the circuit. If you understand how to calculate the gain of a non-inverting op amp, then it's obvious which manner the pot must be wired - but this is not so helpful for begginers.

Are you suggesting that since I'm using 100k resistors for my voltage divider instead of the 10k's on the SoS circuit that I should use a 10uf Cap instead of the 47uf? I wonder why any of this is relevant.

Yes, that is the cap I was talking about. The Vref must be AC ground, or fully bypassed in order for you to get the expected gain. So we need to calculate the half-boost frequency of the Vref point which is R9 and C8 in the schematic you posted. It is calculated as F=1/(2PiRC) where Pi is 3.14, R is R9 and C is C8. You want the frequency to be at least an octave lower than any audio frequency you want to pass. For example, say you want to pass 20Hz, you should bypass to at least 10Hz.

With 10k and 47uF, the frequency is 0.34Hz, so almost 6 octaves below 20Hz. We can also re-arrange the formula so that we can pick a frequency to calculate the required capacitor value by C=1/(2PiRF), so with 100k instead of 10k, we can see that the required capacitor for the same half-boost point also goes down by an order of magnitude, or 4.7uF.
I just recommended 10uF because it's a common value and is probably cheaper than a 4.7uF, it gives a half-boost frequency of 0.16Hz.

You can use any arbitrarily large capacitor, you just need to make sure it's not too small.

For more reading on choosing op amp biasing networks, check out Designing Bias Supply (Vbias or Vb) Networks for Effects at Geofex.

[digitalzombie]

Awesome, Phoenix. Thanks so much!