Princeton Reverb build: A few questions about differently voiced second channel

[Diverted]

Hi all,

Experimenting again and am wondering if anyone wants to chime in on what I'm considering. Both positive and negative comments welcome!
I am going to start building a Princeton Reverb AA1164 and think it would be fun to try some changes. Here's the schematic:
https://schematicheaven.net/fenderamps/princeton_rev_gz34_aa1164.pdf

Anyway, I am using a Fender-style eyelet board and I have a Princeton chassis. The power transformer I'm using will have perfect voltages with a solid state rectifier, so I will not be using one of the three octal socket holes for a tube rectifier.
I would also like to try adding a second, differently voiced channel akin to what Rob Robinette suggests below. I have used it on a bunch of other two-channel builds, but would like to try it here. Notes on Robinette's lead channel mod:

https://robrobinette.com/Generic_Tube_Amp_Mods.htm#Lead_Channel

Here is my (early) idea:
1. Use the unused octal socket to house an octal 6SF5, which I would wire as the phase inverter. Its operating characteristics are just about identical to the 12AX7 though it's a single triode tube. That would free up a triode from the three 12AX7s in the circuit. I have a lot of these 6SF5s laying around.
2. To keep everything else where it should be on V1-3, I would then use V4 (which originally was PI and tremolo) for tremolo and the second, new channel.
3. As space will be tight on the faceplate/panel I want to use just one tone stack on both channels, and a single input jack.
So I was thinking about using a dpdt switch to switch the input jack from one channel's input grid to the other; the other half of the switch would run the signal from either channel into the tone stack.
4. I'll probably put the new channel triode on the one unused power rail node.

Questions:
1. Will I have issues running a second channel preamp triode so close to the tremolo?
2. In his post, Robinette recommends swapping the modified channel's coupling cap for a smaller value. His example was from a Deluxe, which has a .022 coupling cap.
The coupling cap on the front end of the Princeton is only 250pf. Inadvisable to go lower, and just switch the signal after the coupling cap?

Any thoughts? Is this crazy or ill-advised? I would love to give it a try.
Thanks!

[sluckey]

Just some thoughts...

Using a Fender style board. Where will you mount the extra components for the second channel? Princeton chassis... Where will you mount three/four more knobs and input jacks?

You will need two triodes for another preamp. So, four 9-pin tubes for two preamps and reverb. Plus whatever you decide to do with the PI/TREM.

Guessing you want to use that extra octal socket for PI? Two ideas, Hoffman sells an adapter plate that allows you to mount a 9 pin tube socket in an octal hole (my preference). Or, use a 6SN7 dual triode for the PI and the Trem oscillator.

Look at the schematic. That 250pF cap is part of the tone stack. Let it be.

[Diverted]

Thanks for the input Sluckey.
Maybe a (very crappy) schematic will help explain how I hoped/thought it might work. See below.
Anyway:
1. I'll use a 6SN7 for the trem/PI if I end up doing this. That will free up a hole that was previously used by V1, so I'll put a single triode there for the new channel.

As for the extra knobs, pots etc., as I mentioned I would like to use just the one existing input jack and the existing tone stack. My hope was to use a dpdt switch. Half would direct the signal from the single input jack to the stock V1A input grid or the new channel input grid.
The other half of the switch would feed the tone stack (just before the 250pf cap)from either the stock V1A or the new one I'm adding. The photo should help explain it even though it's pretty rough.
Thanks!

[SILVERGUN]

You are not gonna want to run the plates right to the switch like that.
If you just switch the input signal feed between those 2 triodes then you don't have to switch the plates.
It is ok for a tube to sit there idling without a signal going to it.

It feels like you are trying to re-invent the wheel without considering all of the wheels that have come before yours.
I would try to find a working 2 channel design that comes close to what you are looking for and then mod that.

I do like your schematic snip because we can easily see what you're saying.

[Diverted]

Thanks, I was wondering about that DC to the switch issue.
Not trying to reinvent the wheel, but I wanted a PR 1164 and thought it would be a shame to built it without putting SOMETHING in that vacant octal hole 

Having never used a 6SN7, anything I need to keep in mind when using it for the PI/Tremolo oscillator?
How about the plate load resistor size?

Thanks,

Ted

[shooter]

considering all of the wheels that have come before yours.

to SG's point;
I like this wheel (stole from Marshall?)
use the switch hole for the 2nd ch gain pot, your 1 in.  dial one or the other OR both, I know that's a lot of choices while on-stage being a rock star 

[Diverted]

Well I'll never be onstage being a rockstar but I'll take your word for it

Anyway, that schematic looks pretty interesting.

One question, getting back to a point Shooter made:

If you just switch the input signal feed between those 2 triodes then you don't have to switch the plates.

Think I'm following you: So, I would only need a SPDT then for the input jack, with both plates hard running directly into the tone stack? Would I want to use any isolating resistors coming out of the plates? Or should I wire it so the grid of whatever triode isn't being used is grounded out, so I don't induce any hum from the unused channel?


Thanks.

[sluckey]

Or, use a 6SN7 dual triode for the PI and the Trem oscillator.

I meant to say 6SL7.

[SILVERGUN]

Take a look at Sluckey's Supro input switching - attached
You could use this if you wanted both input triodes to be available at the same time or simplify it to select one or the other only. His switching scheme takes care of grounding the grid of the unused triode.

[Diverted]

This is fantastic. I like the option of having both channels together. I was thinking if I wanted to ground that unused grid I would use using and on-on DPDT in that arrangement, but hadn't considered an ON-ON-ON. That's pretty awesome and is probably going to be how I go I think.

Thank you!

[Diverted]

Hi all,

Starting to work on this amp and I have a question.
Per a few suggestions, i am only switching between the input grids of the two triodes. Each triode is receiving power at all times. I am using the DPDT switch configuration someone posted per Sluckey's schematic. It's not shown in the image here. The red dots are each triode's plate; wondering how to incorporate them both into the 250pf cap.
My question is this: How would I feed the plate signal to the 250pf cap? Stock, the preamp plate goes directly into it. I'm wondering about the effect of running two different plates directly into this spot. Am I going to run into issues? Or should I put a small cap between the new channel's plate and that 250 pf to keep DC away from the stock channel's plate?
Hope this makes sense.
Also attached is my work on the chassis so far.
Thanks!

[sluckey]

I would just tie the two plates together. Probably be a bit neater to use V1 for both preamps and feed the volume control into V2.

[Diverted]

Thanks Sluckey.

Question about that: I was hoping to use a 220K plate load resistor for the new channel but if I end up tying them together that's not going to be possible. So then, just live with it? Or is there a way to accomplish that?

Thanks!

[sluckey]

I was hoping to use a 220K plate load resistor for the new channel but if I end up tying them together that's not going to be possible.

Sure it's possible. 220K in parallel with 100K is same as a single 68K plate load. Nothing wrong with that.

[PRR]

IMHO: tying the plates together makes the "two channels" so similar it isn't worth the trouble.

To do what you think you want, use another switch between the plates. (Ah, it may POP.)

Side-issue: the tone stack is a low and variable impedance. Driving it from 270k may not be best.

[Diverted]

That’s what I was worried about (making them too similar). I wanted there to be a clear difference in tone.
That’s one reason why I thought switching the plates would be important..
Someone above had said it wouldn’t be a great idea to switch that high DC voltage but if I used a heavy duty switch would I be able to get away with it?
I’m currently using a DPDT switch to toggle between the input grids, grounding out the triode that isn’t in use. I don’t want to put in a second switch for the plates... would a 3PDT switch like this be OK? It’s rated 250VAC@15a but will it have issues switching DC? The node supplying these triodes would be about 240vdc.

https://m.grainger.com/mobile/product/2X594?cm_mmc=PPC:+Google+PLA&ef_id=Cj0KCQjwv8nqBRDGARIsAHfR9wDsyPn-eTVEdXRa7Z23R1rXYBgRM0Ltxx6t95AY_y2cb5xPZnMVkjUaArd-EALw_wcB:G:s&s_kwcid=AL!2966!3!216788409968!!!g!294737965147!&gclid=Cj0KCQjwv8nqBRDGARIsAHfR9wDsyPn-eTVEdXRa7Z23R1rXYBgRM0Ltxx6t95AY_y2cb5xPZnMVkjUaArd-EALw_wcB

[Diverted]

You are not gonna want to run the plates right to the switch like that.

I don't doubt it, would just like to understand the reason? The plates will be seeing about 240VDC. As far as switching voltage, many, many amps put a standby switch on the B+ line right at the first filter cap, so they're seeing in some cases up to 500+VDC.

Why is it a bad idea in this application? Thanks! Not doubting anyone, just don't have the knowledge and would like to know. Thanks.

[shooter]

Why is it a bad idea

I see it more as a "best practices" thing, you're trying to switch ~100mVAC, why break out a sledge hammer when a brad hammer works well, used millions of times?
also from a safety/engineering "look", when a switch shorts to the finger or chassis 100mV's is mo-better than 300vdc

[Diverted]

Good points. So, what would be the brad hammer?
Perhaps its best not to do it at all then. If I cant use different plate load resistors on each triode and then switch them on/off, the benefits of doing the whole thing kind of drop off. Without those different plate loads I'm wondering how much different the two channels will actually sound.
Thanks!

[sluckey]

Just use your orig schematic but add a .02µF to .1µf coupling cap between the plate and switch. One cap for each tube. Then use any size switch you want. May have pop issues. Deal with that later if so.


EDIT... You can't use that CK 7211 switch to switch grids and plates.

[Diverted]

Thanks Sluckey. I'll give it a try. I'm just trying to learn more and I've found that at least for me the best way to learn is to experiment and ask a ton of questions when I uncover issues I didn't know to think of. Thanks!

Ted

[Diverted]

Anyone care to critique this? I went ahead and switched the output from the plates with the suggestion to use .02 coupling caps between.
To feed the new channel voltage, instead of using a 220K resistor off the node I put a 120 in series after the 100K resistor that feeds the stock channel. It would have gotten circuitous if I hadn't done it this way. Is it a problem?

Thanks!

[sluckey]

The 120K is connected to the wrong side of that 100K.

[sluckey]

There is another problem... with the grid switching... Whichever grid is not selected will be left floating. This screws up the bias on that tube. The fix is to put a 1M from each grid to ground. Then you can eliminate the 1M on the input jack.

[Diverted]

Thanks Sluckey.

As for the 120 .... if I put it on the other side of that 100K as you said that new channel would only be getting a 120K plate load then, right? I was shooting for 220K to that new channel.
I'll go ahead and pull that 120 entirely then and just run a 220K off the main node.

Thanks for the 1M suggestion. I never knew what those 1M resistors on the jack were for.

[sluckey]

I'll go ahead and pull that 120 entirely then and just run a 220K off the main node.

Yes

[Diverted]

Thank you. How's this look?

[Diverted]

One unrelated question:

The way I have this wired the extra triode is running off the D node. This node already has five triodes on it. Will putting an extra tube on it cause issues?
The C node is unused on this amp ... wondering if it is an option to go off this one and adjust the load resistor accordingly to get the right voltage at the plate?

https://schematicheaven.net/fenderamps/princeton_rev_gz34_aa1164.pdf

[shooter]

cause issues

my un-scientific rule of thumb, no more than 2 "active" gain stages per node
(to translate, if you use 4 gain stages on 1 node but only 2 are "on" at a time you don't violate my rule:)

[Diverted]

Thanks. I’ll let it ride then👍

[Diverted]

Got the amp more or less done. Sounds great, nice and quiet and there's a clear difference between the channels which is good. Thanks for all the help. I adjusted the coupling caps from plates to switch; on the stock channel it's .05; on the new channel it's .01, sounds great.
Plates are at 410 and it's biased at about 60 percent. Really happy with the transformers I used on this ... Musical Power Supplies, US-made trannies that are being sold on eBay. I've bought maybe 10 from him and I like them. If anyone is ever thinking about trying them, go through his website instead. Owner (Tim) gives a discount off the eBay price. He will also wind transformers to your spec and his turnaround time is pretty fast.

Anyway Sluckey and others predicted I might be getting pops toggling back and forth between the two channels. That's what happening; it's not too bad but audible. I would love to hear some suggestions; haven't had any experience trying to eliminate pops in this kind of arrangement. Thanks!

Ted

[Diverted]

Dusting off this thread to ask a few more follow-ups.
The amp's done and in my friend's hands. He's played it for a while and loves it, but has one small issue: He said "there is RF getting through" which is very low, but very slightly noticeable when he's recording in an isolation booth with it. I haven't gotten back the amp yet so I don't know what he means: Reverb hum? General background noise? The amp seemed extremely quiet when I had it .. I'll have to go over it.

I was trying to think of the things I'd look into when I opened it up, and even if it has nothing to do with this minor issue I keep coming back to the D node of the filter section and would like advice: Stock this last 20mfd cap has five plates on it. However I added that extra triode for the second channel to that last stage, and now that last cap is serving six plates. I'm wondering if upping the last filter cap to 40mfd will have any appreciable difference on noise (ripple) etc?
Or coiuld I do something like run a second final filtering node (18K>20MFD) to the C node, and then just run that to the new triode's plate via dropping resistor?

https://schematicheaven.net/fenderamps/princeton_rev_gz34_aa1164.pdf

[shooter]

If you can't sit in his booth to hear, have him record it.  that way you have a direction other than random

[Diverted]

I'll be swinging by today to take a listen in his space, and then will bring it home.

[sluckey]

He said "there is RF getting through"

RF is radio frequency. He should be hearing a radio station or CB radio, etc. Ask him if this is the case. If he is not hearing anything intelligible then ask how he knows it is RF interference.

For RF interference the cure may be as simple as adding a 47pF cap across the input jack tip and sleeve. Another thing I noticed is that you only have one 33K grid stopper resistor and it appears to be connected between the input jack and the selector switch. Change that. Put the grid stopper resistor directly on the grid pin of the tube socket. You will need two resistors, one for each tube grid.

[Diverted]

Thanks Sluckey.
I'm familiar a little bit with RF due to having recapped old 1930s radios for years, which pick up all kinds of stuff. Maddening. But when I pressed him on RF he just said "noise," so not sure what it is and I will have to take a look. Thanks!

[Diverted]

Well I went to grab amp and it’s stray, very faint voices/radio? he’s hearing, when listening to it mic’d into a board, with headphones. He only heard it twice over the past four or five days and it was faint, but noticeable he said. Otherwise the amp is very quiet.
I’ve never had this issue before and won’t be able to replicate at my house the conditions in his studio when he heard it.
Preamp input leads are shielded. Thinking about putting a metal shield above the chassis (glued to underside of top of cabinet) as a starting point. Other than that, any suggestions?


Thanks!

[shooter]

to it mic’d into a board

I'd suspect that 1st, mic are a finicky thing.
do some digging on a line out for the amp, they're pretty simple to kinda complex, but that might be a fix

[Diverted]

Thanks. I would wonder about his setup being the cause but he has a few other amps he uses regularly in same setup, and said he hadn’t noticed interference on them. So to be safe I’m going on the assumption that it’s sneaking into my amp specifically somewhere.

[Tony Bones]

Follow Sluckey's advice about grid stoppers on the input stages.

I had a SS amp on the bench once that was picking up radio through the mains wiring.  That one stumped me long time.