New build: Princeton Reverb with mods - advice wanted!

[klooon]

Your bias switching looks OK to me. Not how I might DRAW it, but that doesn't matter.

OK, thanks for checking! I will change this in my layout that I am about to finish and see if it still makes sense.

I don;t know what we are doing with the multitap output. I generally stick the NFB on one winding and leave it there. The speaker changes tap to tap to match impedance, but the amp makes the same signal into the transformer regardless. NFB is a voltage. Ask yourself, "why would there need to be more NFB for a 16 ohm speaker than for an 8 ohm?"

I will say this though, if those speaker jacks are two-tip contacts, that is a clever way to switch. If there are two tip contacts, one wired to the transformer tap, the other wired to the NFB, then plugging into the jack shorts the two together through the tip of the plug. That results in selecting that NFB tap.

OK! Instead of the stereo jacks, I could use this as well then? So no problems with current going through the OT winding upwards passing 'a few taps'?

And allow me an observation. Having done a couple cathode/fixed switchable amp conversions, what almost always happens is that the owner then likes one a little better, and the switch never leaves that setting, once he has showed off the feature to his friends. The difference is really more of aq subtle thing in feel, so it doesn;t wind up as some sort of switchable feature like a boost.

OK, but this is the first time that I will experience this, so I will keep it for the moment. Maybe at a later stage, I can remove this! I also learned from one of the posts of tubeswell that the tremolo works better ('deeper') under fixed bias conditions?! Further, I am very curious about the effect of switching the cap in and out in the cathode bias mode.

Same thing with the reverb dwell. Sounds like a new day on paper, but really, all it can do is reduce the amount of signal into the reverb drive. And the majority of owners wind up with the dwell sitting at max. And we don;t need to make a hole on the panel to have that.

I am planning putting it in a Deluxe Reverb Chassis and head. So I have some extra space to play around with. Maybe, at some point I reduce the amount of 'features' (just before the front and back panel with the annotations will be made). But please keep making these comments, as I don't have this experience yet and maybe I can remove some of the features already now (or later).

[Enzo]

The drawing Tubeswell posted is my style as well. I think his example is more readily clear to a passer by.

The OT puts out a voltage, just as the PT puts out voltages. I was suggesting you could pick the one you want for NFB, and wire to it, the one you use for the speaker doesn;t matter.


I am not really trying to get you to omit your features, just more suggesting what to expect. I have done the cathode/fixed conversions, but instead of a big switch on the panel, we left it a small switch on the chassis. You could still change it any time with a flick, but it wasn;t in the way, or taking up room. Same with the reverb, if you already have a hole for it, great. But if you have the time to experiment a little, you could tack a pot into the project and see if it is really useful to you or not, then include it or not.


A fun old trick is to use the "other" amp channel for reverb. Your Princeton only has the one channel, just one pair of inputs. But many Fender amps have two channels, typically the reverb/trem channel and the "normal" channel, The old trick is to take the return cable from the reverb pan, and plug it into the input jack of the Normal channel out front. Now the channel volume and tone controls work on the reverb. A big deal? Nah. But still something to fiddle with a while and might be just the thing for some guys. And absolutely no modification to the amp.

Here is a circuit that has the two channels, coincidentally, it is your Deluxe Reverb.

[tubeswell]

Off topic, slightly...

What editors are you fellas using?

JSchem

[tubeswell]

Thanks! That's the one I used (or tried) for the v001 and v002 above as far as the switching part is concerned.

What about the values? For instance of Ck and Rk?

Just copy the 5E3 values

And further, should I adapt the tremolo circuit (or values) to "fit" both fixed and cathode bias situations?

Maybe take a look at the Fender Excelsior schematic for a method of trem-wiggle insertion for a 6V6 cathode-biased amp.

[klooon]

The drawing Tubeswell posted is my style as well. I think his example is more readily clear to a passer by.

OK, I see. Please have a look at the new schematic below, I hope this is better now. I am trying to learn

The OT puts out a voltage, just as the PT puts out voltages. I was suggesting you could pick the one you want for NFB, and wire to it, the one you use for the speaker doesn;t matter.

OK, apparently it doesn't matter then that there is quite some winding in between then. For instance you connect the speaker to the 8Ohm tap (and ground) and then if you connect the NFB loop wire to the 16 Ohm tap. OK, I will use the 3 mono jacks and wire then to each of the appropriate taps and connect the NFB wire to the 16 Ohm tap from the OT (which is also connected to the tip of the 16Ohm jack, but not connected to the speaker in case of 8 or 4 Ohm ). Right?!

I am not really trying to get you to omit your features, just more suggesting what to expect. I have done the cathode/fixed conversions, but instead of a big switch on the panel, we left it a small switch on the chassis. You could still change it any time with a flick, but it wasn;t in the way, or taking up room. Same with the reverb, if you already have a hole for it, great. But if you have the time to experiment a little, you could tack a pot into the project and see if it is really useful to you or not, then include it or not.

The reverb dwell should be fine to put at the front. There is some space left. But the cathode/fixed switch: I can see your point, I will put this at the back! What about switching the cathode bypass cap in and out, apparently this gives you a different tone and gain. Switching the cap out will give you apparently a mellower tone and sooner break-up and keeping the cap in the circuit gives you a loud, bright tone more in the direction of the fixed bias. In your conversions, did you made this switchable or not? If not, did you keep the cap or not? I am quite interested in this mellow, sooner break-up tone.
source: Cathode-Bias/Fixed-Bias Switching in Tube Power Amplifiers

A fun old trick is to use the "other" amp channel for reverb. Your Princeton only has the one channel, just one pair of inputs. But many Fender amps have two channels, typically the reverb/trem channel and the "normal" channel, The old trick is to take the return cable from the reverb pan, and plug it into the input jack of the Normal channel out front. Now the channel volume and tone controls work on the reverb. A big deal? Nah. But still something to fiddle with a while and might be just the thing for some guys. And absolutely no modification to the amp.

Here is a circuit that has the two channels, coincidentally, it is your Deluxe Reverb.

Interesting!!

[klooon]

Just copy the 5E3 values

OK, I will put the 25uF and 250R/5W in the schematic. I will order some other resistor values if needed during bias check.

Maybe take a look at the Fender Excelsior schematic for a method of trem-wiggle insertion for a 6V6 cathode-biased amp.

mmm, that's a (for me) totally different circuit. Is it possible to have the same trem circuit for both fixed and cathode bias modes? Would the method I put in the schematic work and/or sound nice???

[Enzo]

It is one thing to think the cathode cap makes the difference you assume, But this is what the breadboard stage is all about. Make the amp without the cap, listen to it. Now tack the cap across the cathode and listen some more. Is the difference enough you'd want a panel switch of it? Or not? If you like it, then install a switch.

[klooon]

It is one thing to think the cathode cap makes the difference you assume, But this is what the breadboard stage is all about. Make the amp without the cap, listen to it. Now tack the cap across the cathode and listen some more. Is the difference enough you'd want a panel switch of it? Or not? If you like it, then install a switch.

I will try! Sometimes it's tempting though to ask advice to save a bit of time. But you're right, try yourself as it can be a matter of personal taste.

I will try to focus on the things that are right and wrong in the circuit.... or really make no sense to try out.

[tubeswell]

mmm, that's a (for me) totally different circuit. Is it possible to have the same trem circuit for both fixed and cathode bias modes? Would the method I put in the schematic work and/or sound nice???

The thing about bias-vary trem in output stages is that the trem-wiggle is fighting the output tube current, and in cathode-bias mode the cathode resistor has the effect of counteracting changes in tube current ('auto-biasing' if you will), which makes it 'harder' for the LFO stage to 'win the battle' with the tube current in order to affect the bias voltage to the desired effect. So you have to have bigger voltage swings out of the LFO stage to get the trem to kick-in (compared to what you need if you are wiggling a fixed bias output stage). The method of trem-wiggle insertion in the excelsior is designed for a cathode bias output stage and the excelsior has a cathodyne splitter (like a BF Princeton) so it seems efficient to not have to re-invent the wheel too much so to speak. (YMMV). But of course you would need to try it out before you can really evaluate it.

[klooon]

The thing about bias-vary trem in output stages is that the trem-wiggle is fighting the output tube current, and in cathode-bias mode the cathode resistor has the effect of counteracting changes in tube current ('auto-biasing' if you will), which makes it 'harder' for the LFO stage to 'win the battle' with the tube current in order to affect the bias voltage to the desired effect. So you have to have bigger voltage swings out of the LFO stage to get the trem to kick-in (compared to what you need if you are wiggling a fixed bias output stage).

Thanks for the excellent explanation!

The method of trem-wiggle insertion in the excelsior is designed for a cathode bias output stage and the excelsior has a cathodyne splitter (like a BF Princeton) so it seems efficient to not have to re-invent the wheel too much so to speak. (YMMV). But of course you would need to try it out before you can really evaluate it.

Very important question first: Would this also work if I switch back to fixed bias mode?

Then:
OK, I have looked at this schematic, it's pretty similar yes.

How to proceed? This is all new to me, please help? The Excelsior has about 270VDC on the PI plate and below the cathode about 80VDC versus the PR has about 200VDC and 50VDC. Adding the 470k and then the same trem circuit to feed the signal?? How do I get these voltages right?

Further, I see in the Excelsior trem circuit that it needs a 320VDC supply. This is what is coming from the 3rd filter stage of the PR, so that's nice?

[klooon]

Apparently, in cathode bias (see Excelsior schem) if the signal is wiggling before the "biasing circuit" this is fine and it doesn't interfere with the basic biasing circuit (there is no signal inserted in the (ground) point between the 220k resistors). Is this observation OK?

Could I use this in principle to switch (practically it's a pain) between trem signal insertion in fixed bias mode inbetween the 220k's and in cathode bias mode in between the (additional) 470k resistors just before the 220k's?

Or could I use this tremolo circuit even for both the fixed bias and the cathode bias mode (so without switching the tremolo signal into one of each; and only switching the bias circuits)?

[tubeswell]

Try it and see

[klooon]

Try it and see

OK, I will try the last option then. I was just curious if it did make sense at all. So, I will use the trem circuit for both cathode bias and fixed bias mode. What about the voltages, should I do some calculations or should I take care about something/ stock different resistor values? I don't want to fry my MM iron... I assume I cannot just copy/paste the values in the Excelsior schem? Maybe you can point me in the right direction (e.g. some links) where to start?

[tubeswell]

I can't see how building the excelsior style trem into the circuit would damage the transformers. If the PT and OT are rated for power output that you will get from a couple of 6V6s in PP (together with the total demand on the heater and rectifier windings), and if you have an appropriate reflected load on the OT, and good tubes, you shouldn't damage the transformers. For power rating you can assume that the PT is at least rated for the total VA you would get on all the secondaries combined (say 100mA x whatever VAC the HT winding is, plus 2.5A (including the lamp) x 6.3V for the heater winding, plus 5V x 3A for the rectifier winding) and then some. Similarly the OT is probably rated for around 20W for a pair of 6V6s in PP, and that will be fine.

The circuit values in the excelsior trem circuit should work. The mosfet they use for a buffer is a high-voltage rated mosfet like an IRF820 http://www.vishay.com/docs/91059/91059.pdf. You will want something like a 12V zener between the source and the gate to protect the insulation in the fet, and you could put a 100R-200R resistor in series with the gate and the previous stage's plate to limit current. If the weirdos at flickr hadn't made it impossible to share the photos there now, I could've linked you to an example of a schematic using an IRF820 buffer, but the sorts of things I mentioned are incorporated in the excelsior schematic anyhow.

[klooon]

I can't see how building the excelsior style trem into the circuit would damage the transformers. If the PT and OT are rated for power output that you will get from a couple of 6V6s in PP (together with the total demand on the heater and rectifier windings), and if you have an appropriate reflected load on the OT, and good tubes, you shouldn't damage the transformers. For power rating you can assume that the PT is at least rated for the total VA you would get on all the secondaries combined (say 100mA x whatever VAC the HT winding is, plus 2.5A (including the lamp) x 6.3V for the heater winding, plus 5V x 3A for the rectifier winding) and then some. Similarly the OT is probably rated for around 20W for a pair of 6V6s in PP, and that will be fine.

OK! This should be fine then as I am using the MM PR transformers, but I was mainly concerned about putting the Exc trem circuit into the PR circuit not knowing anything what it can do to the PR circuit in terms of currents, etc. Just to explain that "just try it"/experimenting doesn't give me a comfortable feeling. That's why I need a bit more guidance...

The circuit values in the excelsior trem circuit should work. The mosfet they use for a buffer is a high-voltage rated mosfet like an IRF820 http://www.vishay.com/docs/91059/91059.pdf. You will want something like a 12V zener between the source and the gate to protect the insulation in the fet, and you could put a 100R-200R resistor in series with the gate and the previous stage's plate to limit current. If the weirdos at flickr hadn't made it impossible to share the photos there now, I could've linked you to an example of a schematic using an IRF820 buffer, but the sorts of things I mentioned are incorporated in the excelsior schematic anyhow.

Great then I will use this! And study the fets a bit more, that will be fine. What about the resistors from the 0.1uF caps to the trem circuit? What is their function exactly? I should use the same value (i.e. 470k)?