ByPassed Tone Stack, Amp Distortion Much Better !

[HaroldBrooks]

Recently I got a bit tired of the highly overdriven (using a boost up front) 'Glassy' and what I term 'Hairy' distortion from my Gretsch 6162 amp (See Schematic). The amp has a rather unusual tone circuit that is represented as a hexagon on the below schematic.

http://www.oldfrets.com/Valco/Schema...retsch6162.pdf

At first, I added grid stopper resistors figuring it was blocking distortion, then filtered out the lowest bass (below 80hz) using smaller coupling caps and smaller cathode bypass caps, and used a Boss PQ-3B equalizer on the input. The problem still persisted, regardless of what I did, including swapping every tube.

I got the idea to bypass the tone circuit from Rob Robinette's wonderful guitar amp website, and now the amp now sounds MUCH better and smoother when driving it with a high gain clean boost. I use a parametric EQ up front to shape the tone, and it's much smoother in the balance without the tone circuit, no question about it, and the amp has become very useful for the garden variety of hard rock I play.

Any thoughts as to what may be the issue with the tone circuit ? It uses all ceramic caps (amp from 1964) so I would imagine they are still in good shape. I should be satisfied with the outcome of all this, and I can install a switch to kill the jumper if I chose to go back to the original tone control in circuit... But it bugs me that I never found out what was causing the nasty sounding distortion in the first place.

The 6973 tubes are not burnt like many are, and definitely not red-plating as I set the bias right at 12 watts each tube. I also used the output transformer shunt method to measure the 6973 tube bias, and they are both very close to 12 watts plate dissipation. The RCA 12AX7s are also all good, and I did spend a great deal of time swapping all of the tubes from end to end in the amp.

Any thoughts as to what may have been happening with the tone circuit are greatly appreciated ! The Tone circuit is an odd one, at least to me.

[eschertron]

I can't say what you encountered, but I can tell you there's a lot of filtering in that preamp. There's some sharp rumble-reduction after V4 with the .01u and 470k stages (6dB/oct each for 18dB/oct with a knee freq of 33Hz), and a little HF rollof just before that (I didn't calculate that, but I assume it's a weak version of the caps seen bypassing anode resistors in some amps). The weird hex circuit, if you unroll it, turns out to be a kind of big muff tone control with 12dB/oct from the ganged filters. Probably a distinctive sound, but I don't think I've ever heard it, so it may not be everyone's cup of tea.
The other thing I'm thinking is that with all that filtering there's some massive phase shift going on. Maybe that's another factor in the sound?

[Dave H]

The hex tone circuit is a mid dip.

[J M Fahey]

Yes, it´s a Big Muff tone control on steroids ... in fact 2 stacked high pass or low pass selectable with the tone pot so as Escherton said and david h simulated leaves a *huge* mid notch if you set it to "5" , extreme highs on one end and rumbling bass on the other.

In fact, an improvement, it must be the widest range single pot tone control and in fact I used something very similar long ago.

Very powerful for Bass, can go from funky slappy sound to deep rumbling with a single twist but can understand that if fed a squarewave sound can be NASTY.
Although maybe Industrial or Doom players may love it.

One very interesting but almost forgotten amplifier, the 40 D cell battery powered Electro Harmonix Freedom, went even further, used the *active* version of this tone control and added an extra knob, the "Bite" one

Coupled to a Celestion Greenback speaker it made this tiny amplifier harder to ignore than the WW2 Hamburg air raid siren .... not kidding.

But admittedly the sound was a little "extreme".

Almost forgot: the extreme high pass is needed because the previous tube has bias wiggling tremolo, and is not self-removed (like on a push pull stage) because it´s single stage Class A and nothing can cancel it, so it must be removed by brute force filtering.

[eschertron]

Almost forgot: the extreme high pass is needed because the previous tube has bias wiggling tremolo, and is not self-removed (like on a push pull stage) because it´s single stage Class A and nothing can cancel it, so it must be removed by brute force filtering.

and... BAM! I learned some more!

[AttachMan]

Always good to have a local copy of any schematic under discussion:

[J M Fahey]

and... BAM! I learned some more!

Oh, you already saw it , check VOX AC30 vib/trem and they also need a strong highpass filter.


Can some kind soul simulate it?

[Justin Thomas]

"Tone Circuit Same As Other Channel" ...
If only all schematics were that polite!
Anyway, something to explore in a build... Always lookimg for effective single-knob tone controls.

Justin

[eschertron]

Oh, you already saw it , check VOX AC30 vib/trem and they also need a strong highpass filter.

Oh, I noted it, I didn't see the why of it. I like to know the "Why" of things.

[Tony Bones]

Oh, you already saw it , check VOX AC30 vib/trem and they also need a strong highpass filter.


Can some kind soul simulate it?

It's a 3rd order high pass filter with a corner frequency of 33Hz. 3rd order means it has a slope of 18dB/octave, so it's 72dB down at 8Hz and 144dB down at 1Hz. Really, more than that as I didn't include the last 4.7n and the volume pot, or the 47n and 1M grid resistor in the calculation.

So, virtually no LFO frequency gets through.

[pdf64]

My understanding is that regular the freq response calculation for the single stage passive RC filter assumes that the source impedance is zero and the load impedance infinite.
Hence it may be better to use impedance bridging for cascading stages that don't have any active buffering between them, eg the R value increases significantly with each stage.
As it is, due to the significant source and load impedances each stage places on the others, the total freq response of the Vox circuit may be rather different to the assumed ideal.
A sim may be helpful in clearing that up

With the Vox https://el34world.com/charts/Schemat...x_ac301960.pdf the summing together at the shared plate load of the balanced stage formed by V8 should act to largely cancel out the modulation signal fed to it from the oscillator's cathodyne.
So the LF filtering from all those 4n7-1M stages may just be 'belt and braces' to remove the residual mod signal, eg due to imbalance.
The V8 plate load of R41//C41 looks like it may give a significant LF boost, which may indicate that those LF filters are eating into the audio bandwidth. Note that the actual value of C41 seems to be 0.047uF, rather than 0.005uF shown on the schematics.

[Tony Bones]

Can some kind soul simulate it?

... the total freq response of the Vox circuit may be rather different to the assumed ideal.
A sim may be helpful in clearing that up

You guys are a pain in my ass.

Juan Manuel, next time I'm in Buenos Aires I'm going to teach you to use LTspice. It took me 5 minutes to setup the sim.



For reference, I assumed 50k source impedance and set the volume control at 50%, though the end result was not very sensitive to either.

Note that the previously calculated attenuation of -72dB and -144dB at 8Hz and 1Hz seem to line up pretty well with the sim. However, the sim includes -6dB due to the volume control at 50%.

[Dave H]

Oh, you already saw it , check VOX AC30 vib/trem and they also need a strong highpass filter.


Can some kind soul simulate it?

Edit: I see Tony has already posted a sim

[HaroldBrooks]

New to this site, but hey, pdf64 is here ! Very Cool !

Good to 'See Ya' once more !

Thanks for your help so far in getting me up to speed with amps and the like. I will continue to learn with your tutelage and experience.

Nice addition regarding higher order series filters ! I still have a lot to learn, so please keep it coming !