Bad Cat Hot Cat questions

[Kipper4]

Schematic
http://www.sabrotone.com/?attachment_id=246

I want to have a go at breadboarding this but dont want the Edge or level pots so

Do I just omit everything between C6 and R10? or are there some bits I cant omit

[Transmogrifox]

It depends what parts of the emulation you want to omit.  The best bet would be to breadboard it with Edge and Level populated as trimmers.  Dial in the sound you want, measure the part, then put in an equivalent resistor value in the final product.

If you omit everything between C6 and R10 you will miss some potential tone shaping and clipping that happens in there.  If you breadboard the whole thing you can bypass this stuff with wires and see if you like the sound with it omitted.

Your question about whether you can omit things needs to be qualified. 

Can you omit stuff?  Yes, certainly.
Can you omit stuff and still end up with a circuit that sounds the same as the circuit with both pots maxed out?  Probably not, but it may be that stage with "EDGE" and "LEVEL" doesn't shape the tone enough to make a real practical difference, and maybe it's ok to just cut out that whole chunk of stuff.

The best way I can think is to breadboard it and then start removing things you don't want and then decide if it significantly changed the overall sound.

[Kipper4]

Thanks Mate
I've been playing but I still have more to do to get to where I want to be.
Theres always tommorow.

[Kipper4]

I've bread boarded the whole thing and It's quite nice.
However I've decided to cut down on the pots and yes The edge (presence) is ok but I'd rather have it on a switch so I guess I just put a 1k resistor in place of the pot and a swith between R and gnd correct?
I also wish to remove the level pot
So I guess I just omit the B1M pot and go straight from C9 to R10 and I wont suffer any signal loss correct? or do i need a 1M resistor to gnd?
not having done much futzing with this kind of circuit, I'd rather ask and look a ninny than stop it working and start from scratch.
I replaced the tone stack with a AMZ BMP stack. I think this is working well as a stack to my ears, but The stacks input resistor is a 39k as opposed to the 10k in the bad cat. I think this (impedance) change is affecting the output of the stack so can I just make R16 smaller to get more oumph? (technical term for more volume)
All advice greatfully recieved
Thanks
Rich

[Transmogrifox]

... so I guess I just put a 1k resistor in place of the pot and a swith between R and gnd correct?...

Yes, that one is pretty straightforward.  You could even switch between two resistor values if you found a sound on the pot you like somewhere in between.

I also wish to remove the level pot
So I guess I just omit the B1M pot and go straight from C9 to R10 and I wont suffer any signal loss correct? or do i need a 1M resistor to gnd?

In this case the cutoff frequency with the 1M is 30Hz.  If you remove it then it goes down to 20 Hz.  In this case I would think it will not make any audible difference if you simply remove the pot and go straight from C9 to R10.

I replaced the tone stack with a AMZ BMP stack. I think this is working well as a stack to my ears, but The stacks input resistor is a 39k as opposed to the 10k in the bad cat. I think this (impedance) change is affecting the output of the stack so can I just make R16 smaller to get more oumph? (technical term for more volume)

Probably where you are experiencing the most volume loss is in the MASTER pot.  If your MASTER is still 10k, then that is where you're getting the drop.

I recommend bumping MASTER pot up to something like 500k or 1M, and then change R21 to something like 2.2M or more.

If you still need more oomph, then change R11 to something much larger -- like several Megs, and R10 to something moderate like 33k or 47k.

[Kipper4]

Thanks
I'll try some of that.
Great idea on the switch.
Thanks for the help
I'll be back

[Kipper4]

It might just be a breadboard problem why I don't have much oomph. Thinking about it. I can't bias the fets very well with trimmers I'm using on the BB

Questions
Changing the voltage divider R10. R11. Won't compromise the biasing of the next fet will it?
How does this work please?

Cheers

[Transmogrifox]

Questions
Changing the voltage divider R10. R11. Won't compromise the biasing of the next fet will it?
How does this work please?

This is all referenced to DC ground, and DC blocked by C9.  The JFET input impedance is very high, so you can put extremely large resistors from gate to bias voltage (in this case ground) without any ill effects on the bias.

Voltage divider R10, R11 just divides the signal level down.  Changing the ratio will effect how much it clips in the next stage, so to do what I suggested will increase "oomph" but also increase the amount of distortion.

My first recommendation would be to increase the MASTER pot value.  If that doesn't give enough output you can always tweak with R22 in the gain recovery stage to get some higher signal level.

Playing with R10/R11 will give more of what I think of "oomph".  This is increasing in signal compression, signal power (not only loudness like R22).

[Kipper4]

Thanks for explaining that for me.
I can see it now.
Play time now I'm done working.

[Kipper4]

Funny you should mention this

"If you still need more oomph, then change R11 to something much larger -- like several Megs, and R10 to something moderate like 33k or 47k."

I built it almost verbatum per the schemo and I'm losing too much signal at the gate of Q3.
looks like i need to change that voltage divider too. R7,R8

[Kipper4]

Perhaps I should add this to the mix too.
Harold's notes

http://www.sabrotone.com/?p=259

[Transmogrifox]

You probably don't want to mess with R7 and R8 unless you don't think it's distorting enough.  BMP tone stack does have a lot of attenuation, so a good way to rule out all other gain losses is to just bypass the tonestack and see if the level is where you need it to be.

My suggestions are aimed at ways to compensate the tonestack losses.  These are best addressed by
R22 for gain recovery
R11, R10 for a little more drive.
R7 and R8 would be something to tweak if you want significantly more distortion.

[Kipper4]

Thanks I'll investigate tonight when I get home.

[Kipper4]

Yep Its kinda working now.
I had R10 and R11 reversed. and had used a 3n3 instead of a 360pf. Verbatum my a.r.s.e.
Currantly osscilating, need to do some more work yet. Maybe later when I get back I'll get time.

[Kipper4]

Im using a variation of V2 here for the tone stack.


http://www.muzique.com/lab/tone3.htm

[ashcat_lt]

Questions
Changing the voltage divider R10. R11. Won't compromise the biasing of the next fet will it?
How does this work please?

Look at the symbol for the JFET.  See the arrow point "into" it, from the gate to the source?  Looks a lot like the schematic symbol for a diode, because it is.  Just like a diode, this one will conduct when the arrow is pointing from higher voltage toward lower.  But we don't want this diode to conduct.  All of the interesting action of the transistor comes in the area before the gate is enough more positive than the source to get over the forward voltage.  So, we want to keep the gate negative with respect to the source, but we don't have any negative voltages to bias that signal with.  We can't lower the (quiescent) voltage  of the gate below 0, so instead we raise the source above 0.  How do we do that?

Look for the voltage divider.  TRM4, Q4 and R12 form a voltage divider.  Actually, there's a couple there, but we're talking about the voltage at the source, right?  The transistor acts like a resistor whose value is dependent on the voltage between the gate and the source, and that value plus whatever TRM4 is set to will be the "top resistor" for our divider, and R12 all by itself is the "bottom".  Only one of these values is adjustable at all.  Bias would be set by adjusting TRM4 so that when the gate is at 0V, the source floats up to about half the supply, or more importantly so that when the audio signal swings the gate as far positive as we expect it to go (big strum) it's still negative compared to the source.

Of course, the actual output comes off the voltage divider that has TRM4 as the "top resistor" and the transistor plus R12 as the "bottom", so that changing TRM4 will have an impact on the overall gain of the stage as well.

As was said above, the gate swings around 0V, and the R10/R11 divider has no impact on that.

[Kipper4]

Thanks Ashcat
heres my voltages
looks way off to me
       D       G      S
Q1  4.2     0       0.001
Q2  4.2     0.26   0
Q3  4.8     0.04   4.8
Q4  4.6     0.3     0
Q5  8.9     4.45   4.47
Q6  4.3     0.2     0

except for Q5 i was expecting voltages to fall in the D ~4.5v    S ~0.2-0.5v less than a volt region and gates to be around 0V
 

[ashcat_lt]

Course, I only sort of understand how these things work, but I think I got it right, since nobody's come along to correct me. 

Q1 is a bit different.  It has no resistor between ground and S, so that'll sit at ground.  I'm guessing its either "expecting" a signal whose positive swings don't go over the diode drop, or were deliberately clipping off anything bigger. 

But are you sure you haven't swapped at least some of your Ss and Gs there?  Things would look more correct to me if that were the case. 

Edit - there would still be something wrong with Q3, though.  You talked about "omitting" that whole thing, but what did you actually do there?  Is C5 shorted?

[Kipper4]

Spot on Ashcat.
There was a problem with C5. My layout was so tight I made a wiring mistake.
Still not perfect. Still a whining problem but I think It should be down to a good biasing set up hopefully.

[Kipper4]

What would be the effect of using a switching diode from gate to drain on a jfet.
Any advantages, Disadvantages. and which fet would it work best on if any?
Thanks.

As seen here in the semenov boogie thread
I realise the diodes are only used on the mosfets here.

http://www.diystompboxes.com/smfforum/index.php?topic=110165.0