7W DIY clean amp for kid's Christmas present

[antonis]

does removing R6 mess up the input impedance of the buffer?

It does mess up with the input impedance in a promising way (by setting non-inverting input impedance the sole input impedance, as seen by Treble pot wiper) but biasing through Mids//R12+Bass+Treble pots might(*) occur to crackling pots..
(with R6 in place, no DC flows through those pots due to zero voltage difference across them..)

(*)depending on input bias current which can be as large as 800pA, according to datasheet specs

P.S.
As for Tonestack pot values, Tom already answered on behalf of myself..

To be more comprehensive: Suppose all pots set FCW.. Tonestack output is loaded slightly more than 50%..
(VREF is considered AC ground so R6 is effectivelly set in parallel with 500k+500k+100k//100k -> 1.05M)
This might not be a problem in case of pure resistive items 'cause signal attenuation could be easily recovered by IC2.2 stage gain but a Tonestack consists of reactive items (caps) which set particular filters corner frequencies with the resistive items (pots)

edit: Dunno if ROG Tonemender Tonestack values are set taking into account R6 bias resistor loading so you better put your desired values here: http://www.duncanamps.com/tsc/ by setting R6 value (1M) as Load and a couple of Ohms as Zsrc..

[Grubb]

Thanks Antonis, so multiplying cap values x5 and decreasing pot and resistor values to 1/5 of current values? If I have understood you correctly, leaving R6 in place seems like a good idea as it reduces the risk of pot noise.

Also for anyone with an eagle eye, I've fixed up all the opamp numbers and their + and - connections since posting the schematics

[antonis]

so multiplying cap values x5 and decreasing pot and resistor values to 1/5 of current values? If I have understood you correctly, leaving R6 in place seems like a good idea as it reduces the risk of pot noise

Just edited above post recently..

[Grubb]

OK, next version of the preamp schematic:


I changed the values of the tonestack as Antonis suggested and left the bias resistor R17 in place. Simulating the before and after values on the guitarscience online TSC, the new values gave me identically shaped curves but around 2-3 dB lower than the original values. I wasn't sure whether R5/100K needed to be reduced by x5 as well?

I am not sure how much gain to set in my first gain stage. Suggestions? Once I get this section finalised I will just about be ready to do a PCB layout.

[antonis]

I wasn't sure whether R5/100K needed to be reduced by x5 as well?

R5 alters MIDS pot taper (from Linear to, counterfeit, Logarithmic) and is also involved in upper pots resistance so IMHO you should lower its value too..

I am not sure how much gain to set in my first gain stage. Suggestions?

That clearly depends on you personal taste..

As it is, you can have a maximun gain of x6 (1 + 500k/100k -> +15dB) which, IMHO again, walks the line between boost and overdrive..
I'd propose you to make R2 22k and IC1.2 a buffer with gain of about x2 to come close to ROG Tonemender original gain, like below..



PS1
All VREFs, except green framed ones, can equivalently be routed to GND..

P.S.2
VREFs end up to GND (via voltage divider lower resistor) so R2 needs a series capacitor to roll-off DC gain to unity
(or else, VREF voltage appears on op-amp output multiplied by particular stage gain, resulting into saturated output for configurations other than unity gain buffer..)

[Steben]

I am not sure how much gain to set in my first gain stage. Suggestions?

That clearly depends on you personal taste..

As it is, you can have a maximun gain of x6 (1 + 500k/100k -> +15dB) which, IMHO again, walks the line between boost and overdrive..
I'd propose you to make R2 22k and IC1.2 a buffer with gain of about x2 to come close to ROG Tonemender original gain

But it will only add the option of opamp clipping or power amp clipping no?

[antonis]

But it will only add the option of opamp clipping or power amp clipping no?

I was talking for particular stage only..
What comes next will be faced next..

[Grubb]

Hello again folks, thank you for the continued assistance. I'm really pleased with how this is going.

Updated schematic as per Antonis' recommendations:

But it will only add the option of opamp clipping or power amp clipping no?

What is the concern here? I am hoping for something reasonably clean but if the opamp clipping is pleasant enough it wouldn't be the worst thing. I am thinking I should probably sim this preamp in LTSpice now to see what the output looks like.

What comes next will be faced next..

What does come next? Happy for any further pointers or recommendations. Thank you!

[antonis]

What does come next? Happy for any further pointers or recommendations. Thank you!

For the time being, try it as it is and here we are for any further recommendation..

[niektb]

in the latest version I dó see a path from VREF to GND through the tonestack, which I think is problematic when sitting in the lower tonestack pot ranges. I would advise a not-too-small series cap between pin 2 of the treble pot and junction with R17 and IC1.2

[antonis]

in the latest version I dó see a path from VREF to GND through the tonestack, which I think is problematic when sitting in the lower tonestack pot ranges. I would advise a not-too-small series cap between pin 2 of the treble pot and junction with R17 and IC1.2

Eagle eye..!!

It was my bad for suggesting Tonestack connection to GND..
(original connection didn't suffer from DC voltage difference between Tonestack ends..)

@Grabb: Either bring back to VREF MIDS pot & R5 or place a 1μF (or bigger) cap between TREBLE pot wiper (lug 2) and R17 (leg 1) joint..
(in case of electro cap placement, positive leg should face towards R17 / IC1.2 pin 5

[Steben]

But it will only add the option of opamp clipping or power amp clipping no?

What is the concern here? I am hoping for something reasonably clean but if the opamp clipping is pleasant enough it wouldn't be the worst thing. I am thinking I should probably sim this preamp in LTSpice now to see what the output looks like.

Or you could build it with the gain anyway and perhaps change the parts if the sound of the stage does not fit into the plan.
It won't sound very classic amp-like.
Something in th eback of my head says If you want the stage to have drive sounds, you are probably better off designing it as such. As a surplus when using well chosen clipping parts, the output will be far less than with the opamp clipping which will render a large part of the master control useful.
I am just spreading a very classic approach. Could be boring but just in case. 

[Grubb]

With Steben's thoughts in mind regarding clipping, I have been experimenting in LTSpice with a combination of my favourite mild overdrive and the FMV tonestack from the Tonemender schematic. I am aware that this is very much a "lego blocks" approach and I may have missed something important in the details. For example, I'm not sure how much the filtering on the first opamp stage messes with the tonestack's ability to shape the sound.

Preamp:


Output in LTSpice, stepping through the Gain control:


The overdrive circuit is the Lightspeed, it's a circuit I have experimented with a lot and whose behaviour I understand well.
Is that level of peak-to-peak output voltage still too hot for the power amp?
I have also reduced the Gain pot to 250k as this seemed to give more intermediate steps between the original signal and full output, are there any unintended consequences of this change?
Thoughts? Suggestions? Corrections?

Thank you all

Edit: the schematic above is missing an RC filter that is between the opamp stages in the spice model, consisting of a 5K6 resistor, then a 10n film cap to ground.

[Steben]

I definitely like the idea the voltage swing out of the EQ will be smaller than out of the earlier version last preamp stage.

[Grubb]

I put together a PCB design for this. I still haven't fully committed to the preamp section, so I think I will sleep on it and think some more.

[antonis]

I still haven't fully committed to the preamp section, so I think I will sleep on it and think some more.

Judging from current build progress, I think your kids will get their Christmas present near Easter ..

[Grubb]

This is fast for me! 

[Grubb]

So I did indeed get this built in time for Christmas. It sounds better than I expected in terms of tone, but not as loud as I was anticipating. At any rate, it is plenty loud enough for bedroom practice. I just wanted to thank the folks who helped me out with this project again, I'm excited to hear what my kids do with this.

[Steben]

Dashing!

[Phend]

Nice present !

but not as loud as I was anticipating

Next build, a booster or equivalent, DIY pedal.!!