Need help tuning bass response in Guyatone TO-2 please! (theory inquiries as well)

[Audiotexan]

Let me start by saying, the root of this goes back a ways (Ampage ala 2002 )for the schematic that I believe originated there (thanks again to the fine folks like Robert Strand, Erik Robertson and BoyHowdy among others)!
Aside: I don't get the Korg reference on the schem, unless Westbury was a sub-brand for them at the time. I've never seen a Korg labelled unit like this. Feel free to shed knowledge/light on that as well if anyone has it!
Also have to give a nod to both BugMeNot and 4Shared, as I was unable to source the resulting schem otherwise!

Putting it up here for not only reference, but others who might need it in times of trouble in the future.


On to the goal at hand. I've owned this pedal for nigh on 15 years (or better), and have always loved the basic character of the tone. However, it was always 'lackluster' in the bass department. And from what I've gleened on a few other forums, it's a concensus, and appears to be inherent in the design (as opposed to a defective/drifting component). Bottom line, after playing with it for a bit, it usually goes back into the closet in favor of another pedal with fuller range/usability. I'm tired of that happening since this is truly a great little pedal. It just needs to be tweaked.

Side note: It always struck me odd that they used a dual-gang 1M pot for the tone stack, as opposed to the more 'normal' 50k or 100k...they did the same in the Guyatone/Nady TD-1 (Westbury W-20) FWIW.

Anyways, the treble feels quite responsive, and am content with the amount of control that it provides (quite a long range actually!). However, given that, the Bass control is almost the opposite. Very limited in the overall frequency range, and very little difference in applied effect, though noting it's range/effect changes with regard to the gain setting.

Since changing only the bass pot is not an option, that area only leaves C4 as an "easy access" tuning point.
I was also considering changing C1 in the NFB loop. Also, what about the C3 bypass...?
Curious as well if C5/C6 may be a limiting factor.

Wondering what input you guys might have about a preferred point(s) of attack.

Also, any theory insights are welcome. I still have a lot to learn about tube theory! For instance, why such a large shift in bass response with dependancy on the gain setting (while fully understanding that R5/RV1B/C3 in essence form a tank ckt. It just doesn't seem like it would be as large a factor as it is.)

Or maybe, why didn't they move the bass control nearer the output like the treble?? Or both in the middle -between stage 1 and stage 2.

PS: please bear in mind my "deficiency" for certain terminology interpretation. If you say use a smaller/larger cap at loc. xx, please accompany it with a suggested value so I don't 'shift the definition' internally.

[52 Bill]

I think I may still have one of these pedals somewhere or it may be a different model. I should look for it when I get a chance.

The two tone controls here are not what I would call a tone stack. They are both simple forms of cut controls. The controls only have an effect on the signal when they are turned down. The treble control is just like a tone control in a guitar that reduces the treble response by shunting higher frequencies to ground through C7. The bass control reduces the bass response by forcing the signal to pass through the small value of C4.

So the most bass that the circuit can pass is when C4 is shunted by the bass pot and is effectively out of circuit. Changing the value of C4 will change the range of the bass control.

[Audiotexan]

I'm probably overthinking things as usual...but that's why I posted. =)

I think I may still have one of these pedals somewhere or it may be a different model. I should look for it when I get a chance.

Thank you sir! I've A/B'd the drawing with mine, and it seems pretty dead on, however mine has different factory wiring colors.

The two tone controls here are not what I would call a tone stack.

I've always associated it with tone controls (in a general sense).

So more like a Baxandall is what most tech's would consider a 'stack'? Or like a 5F6-A, where they're actually 'stacked' (dependant/inter-related on the the prior control's output)??

If so, looks like I need to be a bit less liberal with my terminology then. *chuckles*

They are both simple forms of cut controls. The controls only have an effect on the signal when they are turned down. The treble control is just like a tone control in a guitar that reduces the treble response by shunting higher frequencies to ground through C7.

*nods* Yeah, got that with no problem. It just strikes me as an odd way to do the controls. One 1 way, and one another.

The bass control reduces the bass response by forcing the signal to pass through the small value of C4.

Right...but what I was wondering is was this done because they're riding 'the edge' design-wise, and didn't want to over-saturate the input to v1b? Hence relying on the clipping diodes for distortion while keeping the tube section 'clean'...or am I as usual just overthinking things and seeing a zebra when the horse is staring me in the face? (I know Enzo...I know. ;p) It's a bad habit. Again, my sig is more for my benefit than others. lol

So the most bass that the circuit can pass is when C4 is shunted by the bass pot and is effectively out of circuit. Changing the value of C4 will change the range of the bass control.

Much appreciated! Will target that before the others.

I assume C1 also acts as a 'bleeder', or is it just for voltage stability on the bias?
If the latter, why wouldn't they also have it on v1b?
(If my term. bleeder is mis-applied, please LMK! As I've always considered it a 'bleeder' whether it's draining frequency off, or voltage off. I'd like to be on the same page as most here in my usage though. As knowledge here far outweighs my own.)

[52 Bill]

The only time I hear the tone stack reference is when it is the typical stack of interrelated controls. You can use the term any way that you want, but in order to better understand what you are looking for/talking about I was just trying to be clear.

These controls are the simplest versions of tone controls that you can find. I have no idea as to why they were chosen and could only assume that the designer thought that they were the best sound/cost wise.

As for C1, that is called a cathode bypass cap. It attaches to the tube cathode and bypasses the cathode resistor. By this we mean it creates a low impedance path around the resistor for the ac signal, which increases the ac gain of the stage. The value of the cap will change the frequency response of the gain boost. Some amps will use this as a treble boost or a fat boost feature. If you study early schematics, look at the Tweed Bassman circuit. Each channel has a different cathode cap value on the first input stage.

I think that there must be a frequency response calculator out there on the interwebs to figure out what value will boost which range of frequencies. Somebody here will probably have a link for it.

[Audiotexan]

The only time I hear the tone stack reference is when it is the typical stack of interrelated controls.

Ok, will try to remember to apply that term to only that 'circumstance' then!

You can use the term any way that you want, but in order to better understand what you are looking for/talking about I was just trying to be clear.

No offence taken. And that's a prime example of why/where my terms need to be adjusted. That's why I asked for and appreciate your explanation! I can translate internally, but it's more important for me to be able to try to use the terms the same way others here use them, so I can try to at least "try" to put a question/thought out in a manner that others understand without having to decipher my 'issues'.

These controls are the simplest versions of tone controls that you can find. I have no idea as to why they were chosen and could only assume that the designer thought that they were the best sound/cost wise.

I'm sure. It just seems they went 'all out' for the bass control (seeing as how a change in gain seems to shift the Q of the bass control), yet dump a simple 'cut' circuit on the tail end.

It's more that, the interaction that the bass has (in the circiut) that made me ask to start with. So I knew (assumed) that C1 was affecting response as well, and assume the same for C3. What I didn't know is which one was the 'master' so to speak (as in: overall limiting factor). If C3 was the limiting factor, then that could be opened up and allow a wider frequency band through for C4 to be able to 'fine-tune.' =)

I didn't want to shift the overall Q (resonant freq) point necessarily, just give it more to work with from the onset. Rather than have to consider a redesign of the bass as a boost circuit.

As for C1, that is called a cathode bypass cap. It attaches to the tube cathode and bypasses the cathode resistor. By this we mean it creates a low impedance path around the resistor for the ac signal, which increases the ac gain of the stage.

Ok!! Need to plug "cathode bypass" in (mentally/verbally). Thank you!

The value of the cap will change the frequency response of the gain boost. Some amps will use this as a treble boost or a fat boost feature. If you study early schematics, look at the Tweed Bassman circuit. Each channel has a different cathode cap value on the first input stage.

I do see where the first stage on the 5b6(and rev-a), 5e6(and rev-a), 5f6, all have the 6uf's on the first stage(v1a/b), and then 25uf's on v2a, if I'm following what you were bringing my attention to.

Otherwise, I did see a change when they went to the 6g6 blonde circuit though on v1a, where they used 25uf. But I think I get your general point of notice regardless

I think that there must be a frequency response calculator out there on the interwebs to figure out what value will boost which range of frequencies. Somebody here will probably have a link for it.

If they do that'd be awesome!

f= 1 / 2*PI*R*C should be correct though shouldn't it?
(At least for the 'simple' RV2A/C4 and the R4/C1 concerns....)

Closest I've come link-wise was a pair of calcs:
http://www.ekswai.com/en_highpass.htm
http://www.ekswai.com/en_lowpass.htm

But using R4/C1 as an example:
I came up with 4.82Hz. which is more than ample on the low end cutoff. Anything above that is being passed if I'm on target.

Using RV2A/C4 as example:
I came up with 663.15 hz. Which sets a cutoff that leaves out a lot of bass IMO.

http://www.ultimate-guitar.com/colum...bing_tone.html
I'd guess that I want to get that down to around 70-80hz for passing max range of frequency.

[Audiotexan]

Having done some more calc's on the last bit...

I have to wonder if the 240p wasnt a typo from the layout person, and subsequently built that way at the factory. As when I plug in 2400pf the cutoff comes up nicely to a nice 66.31hz.

2200pf plugs it firmly at 72.34hz
2000pf puts it squarely at 79.58!

Again though, this is all based on extrapolations of my last post. So..IF (big if) I've not lost the plot, then I should be eyeing C4 towards approx. 2000pf to 2200pf.

[nickb]

I was also considering changing C1 in the NFB loop.

With both diodes in C1 shunts all the signal to ground it's not really a NFB loop. With the link removed it shifts the DC operating point of the first stage and so changes the duty cycle of the waveform. So, changing C1 will change the time constant.

Or maybe, why didn't they move the bass control nearer the output like the treble??

It's generally accepted that it sounds better to remove bass before the clipping stage(s) and remove treble after them. Clipping tends to restore the fundamental (lowest frequency) and add lots of high harmonics. The high ones don't sound very musical so it's best to attenuate them.

For instance, why such a large shift in bass response with dependancy on the gain setting

The source impedance into the bass pot varies from zero to several hundred k ohms as the gain is adjusted. That will change the bass control response somewhat.

[J M Fahey]

1 ) 240pF in series with 680K (470+220)=975Hz Fc (cutoff frequency) which is fine for a cut only bass comtrol.
Problem with only 1M parallel pot is that max attenuation is some 7 or 8 dB , poor range adjustment.
But cutting below 1KHz at 6 dN/oct is what most Hi Fi controls do.

2) This pedal is sort of a scam because it´s just a 100 or 200X gain stage followed by diode clipping, same as 99% pedals out there.
An Op Amp would sound exactly the same.
No tube sound will come from it.
But people love the orange glow inside glass, so ....

3) the only strong frequency response limiter is C6 in series with a 333K load (3 x 1M pots or resistors in parallel) so fixed low cut frequency is 220Hz, not bad at all for a distortion pedal (as an example, MXR Dist+ cuts below 720 Hz.)
Considering next amp input impedance of 1M would rise that somewhat, still way below what other pedals cut.

Experiment with raising C6 value, but too much of it will increase bass ... and mud.
Let your ears be the judge.

[nickb]

Here is something you can try that will increase the bass control somewhat and reduce the volume control affecting the response. The schematics show the new and original schemes and there is a plot of the two frequency responses. You can play around with component values as you please. The difference isn't huge but I think it will be a move in the right direction.

Schematics (ignore component references):



Frequency Responses:

[Audiotexan]

Hi Nick! Thanks for your input as well!
I do have a quick question though, as observing what Bill's noted below, and being in accordance with the impression I was under to start with (hence my mention of NFB loop), your first observation kind of "throws a wrench in my works." lol

Not that I mind in the least, but it of course raises some questions for me!

As for C1, that is called a cathode bypass cap. It attaches to the tube cathode and bypasses the cathode resistor. By this we mean it creates a low impedance path around the resistor for the ac signal, which increases the ac gain of the stage. The value of the cap will change the frequency response of the gain boost. Some amps will use this as a treble boost or a fat boost feature.

To me, that implies (whether I'm interpreting it correctly is a whole 'nother issue lol) that it would be a point of re-entry for the signal path, and that (even if to limited extent) signal is being passed back into v1a from v1b with diode clipping added for flavor. I realize that usually those clipping diodes go straight to ground. But with the way it's drawn here, I assumed it was an alternate way to inject signal onto the plate, rather than the grid.

With both diodes in C1 shunts all the signal to ground it's not really a NFB loop. With the link removed it shifts the DC operating point of the first stage and so changes the duty cycle of the waveform. So, changing C1 will change the time constant.

With how you explain it here, I take it that the anode of D1 and the cathode of D2 might as well be tied to ground (signal wise). But if that's the case, why didn't they just tie it to the neg. side of C1 or any other ground point?
In short, I don't follow (pardon my ignorance) :x

"With the link removed it shifts the DC operating point of the first stage"
With the link connected, and assuming DC is blocked from reaching that point in the path by C5, and only the AC signal is hitting the diodes...how would the DC operating point be shifted?

It's generally accepted that it sounds better to remove bass before the clipping stage(s) and remove treble after them. Clipping tends to restore the fundamental (lowest frequency) and add lots of high harmonics. The high ones don't sound very musical so it's best to attenuate them.

Easy enough to follow!

The source impedance into the bass pot varies from zero to several hundred k ohms as the gain is adjusted. That will change the bass control response somewhat.

Again, thanks for the information! Didn't even think about the change in impedance.

[Audiotexan]

1 ) 240pF in series with 680K (470+220)=975Hz Fc (cutoff frequency) which is fine for a cut only bass comtrol.
Problem with only 1M parallel pot is that max attenuation is some 7 or 8 dB , poor range adjustment.
But cutting below 1KHz at 6 dN/oct is what most Hi Fi controls do.

Noted.

2) This pedal is sort of a scam because it´s just a 100 or 200X gain stage followed by diode clipping, same as 99% pedals out there.
An Op Amp would sound exactly the same.
No tube sound will come from it.
But people love the orange glow inside glass, so ....

I respect your viewpoint. But I figured that it would be getting more natural tube sound than any comparable 'tube pedal' (as you infer) simply because most of those don't even get run anywhere near 'tube voltages'. I think/thought it'd be more tube influenced due to running at 250v, rather than the 'cheapos' at 9-18v. Even though they 'cheat' and use clippers. ;P
I guess I was thinking the clippers were only a minor 'flavor' influence. You're saying they're doing the bulk of the work?

What would it take then to clip this sucker naturally? (Or can that be done)

I know it was tossed around in another thread on another forum that I saw where they talked about cutting the 'diode loop' out altogether to see, but from what I remember, it completely shifted the sound. And not for the better (at least in that person's opinion). I should probably see if I can look that thread up again, just for a reminder/reference/comparison.

It definitely has a different flavor (and sounds more natural) than my BK Butler (3-knob) which is just a basic 'glow bucket'. But it was as cheap as a SS dirt pedal the 20 years ago when I bought it, so I take that for what it is (lol). But I don't consider that one a tube pedal though. I think I gave 30-40 for it. So it serves it's purpose. It's got a nice and DEEP mid-scoop to it, that complements my Analogman-modded TS-9's heavily mid-humped sound. The TO-2 is kind of the middle ground. That's why I was wanting see about modding it to get the bass back where it should be. There's no massive mid-hump, and there's no deep mid-scoop. It would be the perfect in-between.

I'm very tempted (if folks think it would be worth it, and willing to lend a guiding hand), to start with this as a 'base' and rework the whole thing IF I could yank the diode clipping out, and use both stages to get somewhere near the same level of gain/tone naturally -strictly reliant on clipping the tube.
I wonder if that little PT would support a second tube...hrmmm. >.> lol

3) the only strong frequency response limiter is C6 in series with a 333K load (3 x 1M pots or resistors in parallel) so fixed low cut frequency is 220Hz, not bad at all for a distortion pedal (as an example, MXR Dist+ cuts below 720 Hz.)
Considering next amp input impedance of 1M would rise that somewhat, still way below what other pedals cut.

On the C6 note, as you reference it, that one's the treble cut -unless I'm misunderstanding your math reference of 3x1M!

I'm using this as the first/second pedal in a chain, so if I need to tailor output impedance as well, I can. Say the word!

Experiment with raising C6 value, but too much of it will increase bass ... and mud.
Let your ears be the judge.

In reference what's being fed back into the 'loop' (allowing leeway for my lack of certainty with that topic as yet)? Or in reference to what's being limited 'on the way out' of v1b?

Thanks again for trying to help!

[Audiotexan]

Here is something you can try that will increase the bass control somewhat and reduce the volume control affecting the response. The schematics show the new and original schemes and there is a plot of the two frequency responses. You can play around with component values as you please. The difference isn't huge but I think it will be a move in the right direction.

Schematics (ignore component references):

[ATTACH=CONFIG]29970[/ATTACH]

Frequency Responses:

[ATTACH=CONFIG]29971[/ATTACH]

Very much appreciated! Also, curious to what you used for plotting? =D

Wouldn't the blue line drop/cut the bass even more?

I'm trying to add bass that's missing back to the signal.
The control definitely can remove bass as it is. lol
It's just that too much has already been removed by their factory cap selection.

[J M Fahey]

On the C6 note, as you reference it, that one's the treble cut -unless I'm misunderstanding your math reference of 3x1M!

C7 is the treble cut , comes after R10 and its right end goes to the cursor of the Treble pot, which either shorts it (no treble cut) or grounds it (max treble cut).

C6 is in series with R10 and then meets 3 1 M resistors to ground (combined value 333K) which are RV2b // RV1a // R11 .
To that we must parallel the unknown next stage impedance, usually 1M .

Note: despite the way it´s drawn, the net C5 , D1//D2 , C1//R4 is NOT an NFB loop, neither DC (because C5 stops DC) nor AC because C1 impedance at any audio frequency we might care about is a short, for all practical means.
In fact those diodes should go straight from C6-R10 to ground.

Note 2= I agree that pulling the diodes would kill a huge part of its distortion ... and by the way I´m not a tube purist, quite the contrary, simply that to clip the second triode (forget about the first) you need way higher signal.

To put it in numbers: those diodes as shown clip with roughly 1400 mVpp, the second triode by itself with roughly 200/220Vpp (basically +V supply minus 12AX7 saturation voltage, some 20/30V) so to regain same level of tube distortion you need some 200/1.4=150X extra gain
More than what a single triode can supply so probably 2 with some attenuation between them.
What, say, the Soldano Supercharger does.
So using those diodes must have simplified the Guyatone guys life for sure

[nickb]

Very much appreciated! Also, curious to what you used for plotting? =D

Wouldn't the blue line drop/cut the bass even more?

I'm trying to add bass that's missing back to the signal.

The control definitely can remove bass as it is. lol
It's just that too much has already been removed by their factory cap selection.

Yes it would cut it even more. I guess as I was mislead as you said you wanted more range.

..the Bass control is almost the opposite. Very limited in the overall frequency range, and very little difference in applied effect, though noting it's range/effect changes with regard to the gain setting.

With the bass control set to zero ohms it doesn't remove any bass at all so it would be impossible to add more (with a bass cut control).
I used ltSpice to simulate the network and draw the plots as I wanted to illustrate the difference. You could lower C6 a bit but I think JMF has covered this.

Oh, and JMF is quite right about the DC level shift thing. It doesn't happen since once C5 charges up it's game over. The funny thing is though why on earth didn't they just take the diodes to ground? Also, along the line of funny things, if you were to add a discharge path through a resistor from C5, then you could get a level shift with one diode but I suspect that some tweaking of values would be required to make it effective. It's food for thought.

[Audiotexan]

On the C6 note, as you reference it, that one's the treble cut -unless I'm misunderstanding your math reference of 3x1M!

C7 is the treble cut , comes after R10 and its right end goes to the cursor of the Treble pot, which either shorts it (no treble cut) or grounds it (max treble cut).

I didn't mean to phrase it in such a manner that implied that C6 was the tone cut cap. I knew that was C7. Sorry again for adding confusion. :/

C6 is in series with R10 and then meets 3 1 M resistors to ground (combined value 333K) which are RV2b // RV1a // R11 .
To that we must parallel the unknown next stage impedance, usually 1M .

This is what I thought you were saying, but I should have said everything following C6. But for some reason I second guessed and thought you might have been factoring the two 470k's (approx. 1m together with the 220k) and the two 1m's (bass/gain) in the bass-cut section. I should have known better. You and Enzo rarely make mistakes as such. Mistake was mine as usual.
Received loud and clear though!

Note: despite the way it´s drawn, the net C5 , D1//D2 , C1//R4 is NOT an NFB loop, neither DC (because C5 stops DC) nor AC because C1 impedance at any audio frequency we might care about is a short, for all practical means.
In fact those diodes should go straight from C6-R10 to ground.

Well, I get why they drew it that way, as it is a reverse engineered schem, and those are the physical connections made. But it's nice to be on the same page finally.

(I'm not trying to 'call Bill out' but I've got to ask) Bill, did you have the same impression I did (probably just an easy oversight for you) or did I misinterpret you and subsequently mis-apply your quote when I was thinking that was a 're-injection point'? If I was wrong in my understanding of your post, then I owe you my apology sir! Not trying to throw anyone under the bus. Especially someone trying to help! But I was thinking you shared my POV.
(Nick/Bill: If you haven't seen a couple of other threads of mine, sometimes it takes a good couple 'whacks' to get through to me. So thanks for bearing with me! Phrasing is everything for me, for whatever reason.)

Note 2= I agree that pulling the diodes would kill a huge part of its distortion ... and by the way I´m not a tube purist, quite the contrary, simply that to clip the second triode (forget about the first) you need way higher signal.

Ok. Confirmed. Diodes are currently the work horses. *grrrr*

To put it in numbers: those diodes as shown clip with roughly 1400 mVpp, the second triode by itself with roughly 200/220Vpp (basically +V supply minus 12AX7 saturation voltage, some 20/30V) so to regain same level of tube distortion you need some 200/1.4=150X extra gain
More than what a single triode can supply so probably 2 with some attenuation between them.
What, say, the Soldano Supercharger does.
So using those diodes must have simplified the Guyatone guys life for sure

Pfft. If by simplified you mean 'cheaped out' then yeah. LOL
But yeah, diodes are way cheaper than another tube/etc...I get it. =P
Thanks for the Supercharger reference as well, that spurred me to go get the schem for the GTO, and I've already glanced over that. My only concern is, it's twice/triple the size of a normal pedal obviously due to the transformer they've chosen. I also note that the schem I found shows basically the same HV lines (barring one exception of v2b plate which is 310v).

Given that Mr Strand noted that the heater's in the TO-2 would carry 5VA to 7VA, would that be enough to handle two tubes (with the idea of a possible rework* in mind)?

*the rework being more of a learning experience-type project. After I've tried playing with C4/C6 values to see if I get some bass back into this puppy!