Ceriatone Chupacabra mods

[FourT6and2]

I have 145 on v1a

I have changed the PT and I have a B+ of 460 v

How do I raise the v1a?

Been 2+ years haha!

145v on V1a is too low. 460v B+ is ok. What have you done to the amp since you last posted in this thread? Hard to give you suggestions without knowing what else you changed.

But, if everything else is still stock Chupacabra spec, then I recommend:

1. Adding the missing 10K dropping resistor between PI and V2 cathode follower.
2. Lower V1a plate resistor to at least 330K.
3. Change V1a cathode to 2K7 / 0.68uF.
4. Change V1b cathode to 2K7 / 0.68uF.

5. Slope resistor to taste 47K, 39K, 33K, etc.
6. NFB to 47K 8 ohm tap.
7. Depth to 2200pF or 3300pF or leave it at 4700pF if you like it.
8. V1b input grid resistor to 68K.
9. Feel resistor can stay at 82K, but if you downtune a lot or run a boost, change to 68K.

 

[peterc52]

Been 2+ years haha!

145v on V1a is too low. 460v B+ is ok. What have you done to the amp since you last posted in this thread? Hard to give you suggestions without knowing what else you changed.

But, if everything else is still stock Chupacabra spec, then I recommend:

1. Adding the missing 10K dropping resistor between PI and V2 cathode follower.
2. Lower V1a plate resistor to at least 330K.
3. Change V1a cathode to 2K7 / 0.68uF.
4. Change V1b cathode to 2K7 / 0.68uF.

5. Slope resistor to taste 47K, 39K, 33K, etc.
6. NFB to 47K 8 ohm tap.
7. Depth to 2200pF or 3300pF or leave it at 4700pF if you like it.
8. V1b input grid resistor to 68K.
9. Feel resistor can stay at 82K, but if you downtune a lot or run a boost, change to 68K.

I know. Haha ? finally have the room to try some building and playing.

I bought the Yeti transformer 350-350 and installed it last week. Raised the dropping resistor before the PI node from 10k to 20k - perhaps too much?

From the start of the this thread I have changed:

1. Adding the missing 10K dropping resistor between PI and V2 cathode follower.
2. Lower V1a plate resistor to at least 330K.
3. Change V1a cathode to 2K7 / 0.68uF.
4. Change V1b cathode to 2K7 / 0.68uF.

5. Slope resistor to taste 39k
6. NFB to 47K 8 ohm tap.

 

[FourT6and2]

I know. Haha ? finally have the room to try some building and playing.

I bought the Yeti transformer 350-350 and installed it last week. Raised the dropping resistor before the PI node from 10k to 20k - perhaps too much?

From the start of the this thread I have changed:

1. Adding the missing 10K dropping resistor between PI and V2 cathode follower.
2. Lower V1a plate resistor to at least 330K.
3. Change V1a cathode to 2K7 / 0.68uF.
4. Change V1b cathode to 2K7 / 0.68uF.

5. Slope resistor to taste 39k
6. NFB to 47K 8 ohm tap.

Raising that first dropping resistor LOWERS the preamp voltages. Put it back to 10K for sure. Amp will tighten up, get less fizzy/hairy, and sound stronger overall. Plenty of gain on tap already + diodes = you don't need "brown sound" by lowering voltages.

You can also get rid of all the stuff on V1a input grid. All you need is a 1M on the input jack for ground reference and a small grid stopper on the socket, like 27K - 33K. But only do this once you get V1a plate voltage up above 160v.

Another thing to do would be to replace the lower voltage diodes with either MPSA06 or 2N5401. So you have 18v zeners on one side of the switch and a more angry, less compressed sound on the other for modern sounds instead of the super low voltage diodes Ceriatone uses.

Anyway... yeah, go back to 10K dropper and your preamp voltages will come up where they need to be.

The other stuff you already did should be great.

 

[peterc52]

Raising that first dropping resistor LOWERS the preamp voltages. Put it back to 10K for sure. Amp will tighten up, get less fizzy/hairy, and sound stronger overall. Plenty of gain on tap already + diodes = you don't need "brown sound" by lowering voltages.

You can also get rid of all the stuff on V1a input grid. All you need is a 1M on the input jack for ground reference and a small grid stopper on the socket, like 27K - 33K. But only do this once you get V1a plate voltage up above 160v.

Another thing to do would be to replace the lower voltage diodes with either MPSA06 or 2N5401. So you have 18v zeners on one side of the switch and a more angry, less compressed sound on the other for modern sounds instead of the super low voltage diodes Ceriatone uses.

Anyway... yeah, go back to 10K dropper and your preamp voltages will come up where they need to be.

The other stuff you already did should be great.

Thanks 46&2!

 

[peterc52]

Raising that first dropping resistor LOWERS the preamp voltages. Put it back to 10K for sure. Amp will tighten up, get less fizzy/hairy, and sound stronger overall. Plenty of gain on tap already + diodes = you don't need "brown sound" by lowering voltages.

You can also get rid of all the stuff on V1a input grid. All you need is a 1M on the input jack for ground reference and a small grid stopper on the socket, like 27K - 33K. But only do this once you get V1a plate voltage up above 160v.

Another thing to do would be to replace the lower voltage diodes with either MPSA06 or 2N5401. So you have 18v zeners on one side of the switch and a more angry, less compressed sound on the other for modern sounds instead of the super low voltage diodes Ceriatone uses.

Anyway... yeah, go back to 10K dropper and your preamp voltages will come up where they need to be.

The other stuff you already did should be great.

Now V1a is at 173–174

 

[peterc52]

Raising that first dropping resistor LOWERS the preamp voltages. Put it back to 10K for sure. Amp will tighten up, get less fizzy/hairy, and sound stronger overall. Plenty of gain on tap already + diodes = you don't need "brown sound" by lowering voltages.

You can also get rid of all the stuff on V1a input grid. All you need is a 1M on the input jack for ground reference and a small grid stopper on the socket, like 27K - 33K. But only do this once you get V1a plate voltage up above 160v.

Another thing to do would be to replace the lower voltage diodes with either MPSA06 or 2N5401. So you have 18v zeners on one side of the switch and a more angry, less compressed sound on the other for modern sounds instead of the super low voltage diodes Ceriatone uses.

Anyway... yeah, go back to 10K dropper and your preamp voltages will come up where they need to be.

The other stuff you already did should be great.

Is it the 0.047 cap at the input you would remove?

https://ceriatone.com/wp-content/uploads/2021/01/Chupa50-28-Jan-2021.jpg

 

[FourT6and2]

Is it the 0.047 cap at the input you would remove?

https://ceriatone.com/wp-content/uploads/2021/01/Chupa50-28-Jan-2021.jpg

The early ones (and all the ones I've been in), had the 1M resistor right on the socket, connected to ground via a tag strip, with the input grid resistor, and a cap along with it, but in the wrong order. The 1M should be on the input jack. The grid stopper should be right on the grid pin of the tube socket. You can remove the cap. Or... don't touch any of it if you don't want to bother. It's not doing any real harm, but there's a tiny difference in the feel of the amp with it wired in the traditional way.

 

[FourT6and2]

Oh and looking at the Chupca schematic again and your photo in your first post, I suggest changing the two 0.1uF PI output/coupling caps to 0.022uF and the PI grid leak resistors to at least 150K, if not 220K. But it's personal preference. The 0.1uF and 110K combo makes the amp more bass heavy and looser, with lower output. 0.022uF and 220K is typical Marshall flavor and will have more punch and that typical Marshall authority. Marshall used 150K in amps equipped with 6550s.

 

[glpg80]

Oh and looking at the Chupca schematic again and your photo in your first post, I suggest changing the 0.1uF PI caps to 0.022uF and the PI grid leak resistors to at least 150K, if not 220K. But it's personal preference. The 0.1uF and 110K combo makes the amp more bass heavy and looser, with lower output. 0.022uF and 220K is typical Marshall flavor and will have more punch and that typical Marshall authority. Marshall used 150K in amps equipped with 6550s.

The 0.1uF is sized to be roughly 10x larger than the 0 degree phase reference for textbook low frequency stability of the phase inverter. The real problem for the loose attitude is that the in phase coupling capacitor is too large. Remember that the gains of the phases are compensated due to triode differences of each envelope. The same exists for frequency stability hence the large variance of component sizes. At the end of the day you can change whatever you want - but I suggest tightening up an amplifier in other methods and not messing with the phase inverter.

 

[jasonP]

Oh and looking at the Chupca schematic again and your photo in your first post, I suggest changing the 0.1uF PI caps to 0.022uF and the PI grid leak resistors to at least 150K, if not 220K. But it's personal preference. The 0.1uF and 110K combo makes the amp more bass heavy and looser, with lower output. 0.022uF and 220K is typical Marshall flavor and will have more punch and that typical Marshall authority. Marshall used 150K in amps equipped with 6550s.

I really like the feel and depth of the .1uf/100-110k combo here with this amp (that is what fortin/growler use - you know this already) but then you have to cut the bass on the first and possibly the second stage and run the stock 47k slope, cuts a little low mids in the tone stack (I run 2 .001uf's paralleled off the first stage coupler rather then the .0022uf). I found if you leave the feedback on the 8 ohm tap and go to like 85k'sh on the feedback resistor is just rips man. Everything else like you said with the 2.7k/.68uf on the first 2 stages, 330k on v1 plate. I run on the 18v zener all the time. This is with the 100 watter.

 

[FourT6and2]

The 0.1uF is sized to be roughly 10x larger than the 0 degree phase reference for textbook low frequency stability of the phase inverter. The real problem for the loose attitude is that the in phase coupling capacitor is too large. Remember that the gains of the phases are compensated due to triode differences of each envelope. The same exists for frequency stability hence the large variance of component sizes. At the end of the day you can change whatever you want - but I suggest tightening up an amplifier in other methods and not messing with the phase inverter.

It's not "messing with the PI" to return the 0.1uF and 110K grid leaks back to the most commonly used value/combo in the history of guitar amps. 0.022uF + 220K is tried and tested. But again, it's personal preference.

Why neuter the amp by running 110K in the PI output? Run the sims on AmpBooks.com. 0.1uF/110K vs 0.022uF/220K.

The Chupa circuit is ridiculous. Very loose, fizzy, so much gain, voltages are super low. I mean really... PT running 325v secondaries (B+ is 400v) + 470K V1a plate + 820R/0.68uF in every stage + TWO 4700pF bright caps + zener clipping + 0.1uF PI caps. It's just a messy amp. The stock voltage on V1a is 85v lol. That's why there's all that junk on the input. It's because there is such low voltage there it leaks DC to your guitar's volume control.

 

[FourT6and2]

I really like the feel and depth of the .1uf/100-110k combo here with this amp (that is what fortin/growler use - you know this already) but then you have to cut the bass on the first and possibly the second stage and run the stock 47k slope, cuts a little low mids in the tone stack (I run 2 .001uf's paralleled off the first stage coupler rather then the .0022uf). I found if you leave the feedback on the 8 ohm tap and go to like 85k'sh on the feedback resistor is just rips man. Everything else like you said with the 2.7k/.68uf on the first 2 stages, 330k on v1 plate. I run on the 18v zener all the time. This is with the 100 watter.

Yeah, I tried 0.1uF/110K and it was ok at first. But after A/B testing like every value combo I could think of, I think Marshall just got it right with 0.022uF/220K with EL34s. Fortin used 0.1/110K on a small number of one-off mods if I'm correct. The Meshuggah, for example, is good ol' Marshall through and through (Lead spec, not Bass spec).

I agree with the NFB. I've been running 47K/8ohm for a while, but it does rip pretty hard with 85K-100K ish. I just worked on an amp today and ran a pot in series with 47K NFB and the client settled on 94K with a blind play test.

 

[glpg80]

It's not "messing with the PI" to return the 0.1uF and 110K grid leaks back to the most commonly used value/combo in the history of guitar amps. 0.022uF + 220K is tried and tested. But again, it's personal preference.

I stated you can change whatever you want wherever you want - tonal preference and theoretical fact are two different ballgames but the former does not invalidate the latter. I’m positive its in the Radiotron handbook and also a writeup by Aiken.

 

[FourT6and2]

I stated you can change whatever you want wherever you want - tonal preference and theoretical fact are two different ballgames but the former does not invalidate the latter. I’m positive its in the Radiotron handbook and also a writeup by Aiken.

You lost me. What's in the Radiotraon handbook and also in a writeup by Aiken?

 

[glpg80]

You lost me. What's in the Radiotraon handbook and also in a writeup by Aiken?

Go back and actually read post #49. I was just sharing facts and you went off.

 

[FourT6and2]

Go back and actually read post #49. I was just sharing facts and you went off.

I'm speaking pretty calmly here. I didn't go off.

You said there are better ways than "messing with" the PI. And my point is that returning the PI setup in this amp back to what Marshall and many other amps throughout history used is not "messing" with anything. Your word choice implies you think doing this is bad?

There's nothing wrong with adjusting the PI values, especially in conjunction with the other tweaks already mentioned.

So what's the alternative?

 

[glpg80]

You’re stuck on 0.1uF/110k. I’m only talking about textbook reasons why the in phase differential input decoupling capacitor should be 10x smaller than the out of phase input. Fuck the resistor. I don’t rightfully care what Marshall are ceriatone or anyone for that matter prefers for tonal reasons.

And you keep quoting “messing with” as if I haven’t explained the theoretical reason why already, if you can’t follow and that’s all you get from my comment then you’re looking for a posting pissing contest. If you want more of an explanation then I’ve also quoted free sources - I don’t need or care about going to ampbook.com

 

[SpiderWars]

I’m only talking about textbook reasons why the in phase differential input decoupling capacitor should be 10x smaller than the out of phase input. Fuck the resistor.

I'm not even sure which caps you're referring to here. Can you identify these two caps in the typical LTPI schematic? Do any guitar amps actually have this '10x smaller cap' between the in-phase and out-of-phase caps?

 

[jasonP]

Yeah, I tried 0.1uF/110K and it was ok at first. But after A/B testing like every value combo I could think of, I think Marshall just got it right with 0.022uF/220K with EL34s. Fortin used 0.1/110K on a small number of one-off mods if I'm correct. The Meshuggah, for example, is good ol' Marshall through and through (Lead spec, not Bass spec).

I agree with the NFB. I've been running 47K/8ohm for a while, but it does rip pretty hard with 85K-100K ish. I just worked on an amp today and ran a pot in series with 47K NFB and the client settled on 94K with a blind play test.

Totally get it man, not for everyone at all. Your descriptions on what the amp was stock is right on as well. In my 50 watter I really like the .022uf/150k combo as I think it's a little tighter feeling and more mid focused and punch with awesome pick attack like a Marshall should be. The best advice you have given me is, if there is a resistor there in question that you want to focus on and tweak that value to get right just place a pot in its place and adjust until it's right. I've used that trick a lot!

 

[fusedbrain]

I'm not even sure which caps you're referring to here. Can you identify these two caps in the typical LTPI schematic? Do any guitar amps actually have this '10x smaller cap' between the in-phase and out-of-phase caps?

I was wondering the same thing, but I think he's referring to the .022uf from the tone stack/mv into the pi and the.1uf on the other pi input.
The Friedman Dirty Shirley has a .0022uf on the pi input fwiw....

Seems the majority of this discussion is about the pi OUTPUT couplers and the bias feed resistors. The size of these components affect "bias excursion" as there is a time constant involved. As mentioned, Aiken has written about this, and there is a calculator an ampbooks for these values.
The gist being you have to watch the size of the caps and adjust the resistors accordingly, or the recovery time will be too slow.
I tend to agree with 46&2 that .022uf and 220k is the way to go, although you could go with 150k and leave open the ability to run 6550's if desired.
I don't hear much difference between 150k and 220k, and I like swapping in 6550s in some Marshall style circuits, especially when diode clipping is being used.