Marshall, Matchless etc and other EL84 amp questions

[Toxophilite]

Hi I have a couple of AO-43 Hammond chassis
I want to build a two channel amp
I'm looking for an amp that will give me as much clean headroom as possible within the abilities of a pair of EL84


I've wondered about the marshall 18 , a matchless 18 amp maybe but I have no actual experience with these amps so i have no idea which is better or if there's a better design

The other option is to just build an appropriate pair of preamps in front of the existing power amp and phase splitter (I'm partial to the standard 3 band fender tone stack)


which of these options would be ideal for maximum cleans or is there a better other option ?
All suggestions appreciated


I'll post the schematic. it's the reverberation and power amp part of the schematic.



My other question relates to the Gulbransen Amp I've gotten back to.
I'm including it in this post as I'm trying to accomplish a similar result(lot's of clean headroom)
 I added another 12AX7 socket and tried running into one side of that,out into a fender tone stack and then into the gain stage of the next 12AX7 prior to the one triode phase splitter. needless to say it was a little much with 3 gain stages and the amp made pumping noises as it warmed up and was too gainy


So I bypassed one intial 12AX7 stage and it's still plenty loud but much better behaved and good sounding


My question is: Would I be better off rewiring the phase splitter to be something more inline with an existing build???(marshal 18 or matchless,etc) or is it okay as is?
I just wonder if it's a more unstable design
 It seems to be working and I might use the unused 12ax7 gain stage to run another channel into it and have them share the same preamp.


I'll include that schematic as well


THanks very much

[HotBluePlates]

Hi I have a couple of AO-43 Hammond chassis
...
The other option is to just build an appropriate pair of preamps in front of the existing power amp and phase splitter (I'm partial to the standard 3 band fender tone stack)
...
which of these options would be ideal for maximum cleans or is there a better other option ?

When recycling existing amps, it is almost always best to keep the existing power supply, output tubes, output transformer and phase inverter stock. These 4 things define your output stage, and the existing parts were all designed to work together. There is probably not a different arrangement you could use to gain any more performance than the stock arrangement, unless you start swapping transformers (and if you do that, you eradicate most of the savings of recycling an amp).

You can assume with an old amp (whether PA, hi-fi or guitar amp) that the output stage and power supply was designed such that the output tubes would distort first, then the phase inverter, then whathever preamp stages. In other words, to assure the designed clean-power output, the output tubes would have been the limiting factor and not earlier distortion in the preamp.

You can probably also assume the phase inverter was designed to take ~1v of input from the preamp to drive the output tubes to full clean power. This assumption may need to be checked if feedback is wrapped  around the output section.

For the purposes of the AO-43, take the triode ahead of the split-load phase inverter as being "part of the phase inverter" and keep it as well. This is also the point speaker NFB is returned to, and was designed as part of the total output stage.

If you add the typical Fender blackface 12AX7 preamp (gain stage, tone stack, volume control, gain stage) you essentially have the non-reverb Princeton with EL84 output tubes.

[HotBluePlates]

My other question relates to the Gulbransen Amp I've gotten back to.
...
 My question is: Would I be better off rewiring the phase splitter to be something more inline with an existing build???

The Gulbransen schematic you included is only the power section. Both 12AX7 triode are used as the phase inverter (a split-load inverter, and a pre-gain stage ahead of it, and direct-coupled into it to set the phase inverter's bias).

Again, I'd think you should leave that alone and just add whatever preamp & tone control you want prior to the input jack shown on the schematic. Feedback is wrapped from the speaker to the cathode of the 1st tube stage, and the whole schematic is just an output section.

If you want clean and a single 12AX7 preamp tube with tone control and volume (a la Fender) isn't working for you, then keep the Fender preamp and swap the tube for a 12DW7. This is 1/2 12AX7 and 1/2 12AU7. You want the 12AU7 to be the 2nd preamp gain stage. That way, as the signal is boosted by the 1st triode (fr good signal-to-noise ratio) it is very unlikely to overdrive the input of the 2nd gain stage at any volume setting.

[sluckey]

I have a couple AO-43s also. In fact, I just built a Plexi6V6 using an AO-43 as a donor. I did a lot of research on the AO-43 and then I did a lot of testing with a 50mA, 100mA, and 120mA load and various rectifier tubes, including SS plug.

The AO-43 has been very popular for converting to a Marshall 18W (model 1974) amp. Many have just stripped the guts and built  the Marshall circuit right in the AO-43 chassis. They even stripped the cone board and loaded it with the Marshall components. The long chassis is very friendly for this type conversion. No reason to not consider a Matchless Lightning or Vox AC-15 type circuit either. Except for one thing...

The most frequently asked question about AO-43 conversions to any EL84 based amp is "Help. My B+ is way too high. How can I drop it?"

There are several reasons for the high voltage.  1) Modern line voltage is higher. 2) Voltage readings on the original AO-43 schematic were taken with a 10,000Ω per volt VOM. Your 10MΩ DVM will read higher voltage because it's such a light load compared to the '50 VOM. 3) The power supply on the AO-43 powered much more than just the AO-43 chassis itself. In fact, the schematic shows a 160mA current draw right off the rectifier tube. Your EL84 based 18Watt amp will draw nowhere near that amount of current. What I'm getting at is when you take that AO-43 out of the L-100 organ, the load on the power supply goes down and the B+ will rise.

Using a 5U4 and fresh filter caps in a typical EL84 amp will put the B+ close to 400V. Using a NOS 5Y3 will drop it to about 370V. The increased B+ will be hard on a cathode biased EL84 amp. Fixed bias amps may not be stressed as much.

Here's an idea you may want to consider...

High B+ becomes a non-issue if you abandon the EL84s and replace them with modern 6V6s. The iron in the AO-43 is well suited for this. Once you change the output to 6V6s you open the door for a lot of clean sounding amps. Several Ampegs and Fenders come to mind. Princeton, Tremolux, Deluxe AB763 can all be built right in this chassis. All of these 6V6 based amps (especially the fixed bias amps) will have more clean headroom than the Marshall 18W or Matchless Lightning.

PS... Do a search for "Hammond L100 schematic" and download the L100 service manual. It's about 57MB and shows the complete schematic.

[Toxophilite]

This is good stuff
Thanks both of you for taking the time!


I was checking out the 'lil' tiger' some people make out of this amp and it looks like they keep the power amp and phase inverter intact just as Hot Blue Plates suggested.


It might be cool to build a another fender style 6V6 amp too..I've got another pair of 6V6s kicking around and 5U4s galore
And I really like that deluxe I built using the electrohome chassis...(I've always secretly wanted a pair of deluxes)


Great suggestions!, gives me a lot of different possibilities to ponder


As for the Gulbransen amp ,that's the entirety of what's in the chassis (not including the actual power supply) . This was the only vacuum tube part of that organ, The preamps were all transistor and elsewhere

I added another 12AX7 socket and and wired it up ala a fender amp (same cathode bypass 1.5k resistor, 25/25 cap) and 100 k resistors on the plates.  Is that an okay template to use in this instance?, It seems to be functioning okay (despite the excessive gain when using the entire tube)

Looking at a princeton(minus reverb) I guess the other route I could go would be to use the other 1/2 of the 12ax7 to make a tremelo circuit as the amp just has 1/2 a 12ax7 gain stage them the tone stack in front of the tube used as a phase inverter/gainstage

I did try taming it with a 12au7 and whilst it was a lot better it was still unruly and made pumping sounds as it warmed up. Coincidentally I just got a 12AD7 the other day which would be something to try as well just for fun, It really is very loud with just the one gain stage in front of the PI 12AX7 and was much better behaved.

[Toxophilite]

S. Lucky
Out of curiousity
You mentioned a tremolux which I thought was a 6L6 amp(I could be wrong)
That got me thinking
From your research on the AO-43 do you think a 2 6L6 amp like the vibroverb could be built on this platform?
or would that be pushing it too far
It does have a pretty beefy power transformer., and I do have 2 6L6 pushpull OT from a Baldwin(It's PT is a Boat anchor the size of a babies head!)
Just curious

[jjasilli]

Another solution (which I have used) to drop B+ to EL84's is a 12V bucking transformer + a dropping resistor.  If 120VAC produces 400VDC B+, that's a boost factor of 3.3.  A 12V bucking tranny will drop wall voltage by 12VAC X 3.3 = 40VDC.  You will now have B+ @ 360VDC.  If you want you can drop up to another 50VDC B+ with a suitable dropping resistor.


When using a bucking tranny or (variac) make sure filament voltage for 6V tubes has at least 5V, loaded.  Assuming your loaded filament voltage is 6.3VAC; then 120X x = 6.3; x = 6.3/120 = 0.0525;  0.0525 X 12 = .63; 6.3 - .63 = 5.67.  This suggests that there's an option use a larger bucking tranny, say 18V or 24V, but do the math. 


Moreover, your filament voltage may be high anyway, if the PT's filament secondary was expected to be loaded with more tubes elsewhere in the organ. 

[sluckey]

I was thinking 5G9 Tremolux (because of another current thread). The PT may be just on the small size for 6L6s and the OT is definitely on the small size for 6L6s unless you really choke back the current and run the OT with a 4Ω speaker. But then why not just use 6V6s?

No more than I play, 6L6s would probably hold up in my amp but I'd be nervous to use it for a 4 hour gig even in a small venue.

[Ed_Chambley]

Could you add a couple more EL84?  I am not familiar with the chassis and how much room.  Of course, change the OT's secondary load.  Sort of like making an AC30 or the DC30 Matchless.

I just guessing since it is old organ iron the OT could handle more clean watts.

[sluckey]

Could you add a couple more EL84?  I am not familiar with the chassis and how much room.  Of course, change the OT's secondary load.  Sort of like making an AC30 or the DC30 Matchless.

I just guessing since it is old organ iron the OT could handle more clean watts.

No room. But, the AO-43 is a 15 to 18 watt amp. The OT is about the size of a DR OT. I suspect it will handle 15-20 watts just fine. But I would not try to make it handle 36 watts. Would you add a couple more EL84s to your ac15 and use the same OT?

[Toxophilite]

Thanks for clearing that up (regarding the 6L6's )
Curiousity was getting the better of me


I'm probably going to try and build one like a deluxe and maybe leave my options open for the other
As S.Lucky said the chassis is of a nice length to accomodate it and I like the sound of a deluxe and if it's going to have a big organ PT it might as well have a bit more headroom


I've been playing the Gulbransen amp and it sure is sounding fine, especially through the JBL speakers I'm trying it through (D130 and D123) I find the JBLs are almost like a secret weapon(if you like cleans). Lovely jangle, nice warm jazz tones, Lot's of articulation, breaks up a little at about 8-10 on the volume but it's almost a pretty sound.


I'm curious about my currently unused 12AX7 triode section.


Right now I'm running into 1/2 of a 12AX7 through a fender tone and volume stack and then straight into the existing power amp as shown on the schematic (12ax7 PI with gain stage into EL84s)


I'm wondering if I can do the exact same thing with the unused triode? (hypothetically, there might not be room )
Make another channel that goes into it and then out into a fender tone stack then through a mixing resistor into the existing power amp (and have a mixing resistor on the initial channel too )


Would that work or would the EQs and volumes become too interactive with each other??
Seems almost too easy!

[HotBluePlates]

I'm probably going to try and build one like a deluxe ...

If you use 6V6's, you may get an inadvertent leg-up in your pursuit of clean-tone.

6V6's have lower Gm than EL84's, therefore require more bias and drive voltage for the same output power and circuit conditions. So the same output power will require bigger drive from your preamp, meaning you're less likely to distort the output stage.

If you're not careful, you could transfer the distortion problems to the phase inverter or preamp. I'd have to have a sense of what bias voltage 6V6's land on in real-world use in this circuit, as well as B+ voltages throughout.

[Toxophilite]

Any insights into this question?
Thanks
I've been playing the Gulbransen amp and it sure is sounding fine, especially through the JBL speakers I'm trying it through (D130 and D123) I find the JBLs are almost like a secret weapon(if you like cleans). Lovely jangle, nice warm jazz tones, Lot's of articulation, breaks up a little at about 8-10 on the volume but it's almost a pretty sound.
I'm curious about my currently unused 12AX7 triode section.


Right now I'm running into 1/2 of a 12AX7 through a fender tone and volume stack and then straight into the existing power amp as shown on the schematic (12ax7 PI with gain stage into EL84s)
I'm wondering if I can do the exact same thing with the unused triode? (hypothetically, there might not be room )


Make another channel that goes into it and then out into a fender tone stack then through a mixing resistor into the existing power amp (and have a mixing resistor on the initial channel too )


Would that work or would the EQs and volumes become too interactive with each other??Seems almost too easy!

[HotBluePlates]

Right now I'm running into 1/2 of a 12AX7 through a fender tone and volume stack and then straight into the existing power amp as shown on the schematic (12ax7 PI with gain stage into EL84s)
I'm wondering if I can do the exact same thing with the unused triode? (hypothetically, there might not be room )

If this arrangement works for you, then it's fine.

Typically, you'd have the Fender arrangement of gain stage, tone/volume, gain stage, output section. The reason is tone stack imposes significant signal loss, so you boost to ensure the output section is fully-driven. Maybe more-gain-than-needed is available, to accommodate a range of pickup strengths/outputs. The designer might figure out needed gain from the 2nd stage by assuming the volume control is at 1/2 or 7/10ths, allowing both turn-up and turn-down, as a shortcut to accounting for pickup variations.

If you don't want distortion under any circumstance, the answer might be as you've done: provide no make-up gain after the tone circuit/volume and cripple drive to the output section. Unless something is designed way-wrong, the output stage would distort first under either plan, and you'd only be limiting drive to the output stage to prevent it ever being driven hard enough to distort. If you're not careful with the approach, you could reduce the maximum output power of the amp, because the output section is never driven hard enough to reach its full clean output. That might be alright for you, because there is likely to be a range of increasing "furriness" before obvious distortion sets in.

All the above is why I suggested a 12DW7; if the 2nd gain stage is the 12AU7 section, you get a little re-boost but very much less than if there was another 12AX7 triode there. After looking at a Standel, I think folks are missing the boat when they swap the input stage for a low-gain tube but leave 12AX7's for later stages. The Standel Preamp schematic floating around the internet shows a 12AX7 triode, then a 12AT7 triode, then a 12AU7 triode, then phase inverter. Each successive stage can handle larger input signals cleanly (as well as amplify less), so a maximally-clean clean gets relayed along.

Would that work or would the EQs and volumes become too interactive with each other??Seems almost too easy!

The Typical Fender solution of tone circuit, volume, then mixer resistor will likely work ok. You'd be combining a blackface-style tone circuit then volume control with the 5F6-A style volume control then mixer resistor.

Will there be some interaction? Probably. I don't know if it will be "too much" ... better to just try it and see what you think of its performance.

[Ritchie200]

This may or may not have any bearing on this particular discussion...  I have always wondered how a el84 based amp would sound UL.  Obviously in this case we would need to replace the original with another op tranny, but if the desire is loads of headroom, this would be a way to get there.  I am not aware of any UL el84 guitar amp designs?    For that matter, I have never looked to see if there was any appropriate op iron available with taps.


Jim

[sluckey]

Magnatone M10A uses 7189 (close to EL84) in a UL circuit. Sounds very clean.

[Toxophilite]

Hot Blue plates I did try a 12au7 in there prior to bypassing a  a gain stage and it was really gainy (super loud at 2 or 3) and I'm pretty familiar with what loud is). Also the amp made strange and wonderful tremelo/pumping sounds when warming up. I might try it again just to give it a thorough testing. While I'm at it I'll try the 7247 too and get back to you


With the the way it's set up now it's like the princeton (minus reverb) that you were mentioning and has the definite beginnings of distortion between 8-10. It's pretty damn loud there but the JBLs I'm playing through make most amps sound loud.


I have a few 7189s but alas no ultralinear OT. That would be neat but i think I will stick with the iron the amp has as it's a budget build.
My super has an ultra linear OT and I like my super a lot despite it being often considered the least desirable silverface.

[Toxophilite]

Hello
Hooked the estra gain stage back in
input -1/2 12AX7 - tonestack and volume - 1/2 12AX7 - Phase inverter


-Crazy pumping noise on start up
- amp started distorting alot around 4 on the dial.
-Lot's of noise(buzzy hum)
-seemed out of control


Even with a 12AU7 in V1


Switched it back to


input - 1/2 12ax7 - tone stack and volume - phase inverter


I'm pretty sure only one half of the 2nd 12ax7 is functioning as a Pi and the other half is another gain stage which makes sense from the results I was getting


Sounds great set up this way and is Very loud with minimal distortion

[Toxophilite]

And to provide further context
I have a 34K input grid resistor to emulate the louder input on fender amps and I'm using a guitar with 7ohm single coil pickups (single soil harmony d'Armonds), that's my loud guitar! all the others have 3 ohm pickups (old gretsch's)

[HotBluePlates]

-Crazy pumping noise on start up
...
-seemed out of control

This is not "normal operation." A 60's Fender Princeton is 2x 12AX7 gain stages, then a pre-gain stage and split-load inverter; it neither has a pumping noise nor is out of control.

Is there a way for you to upload a schematic of how you wired your amp and/or upload photos?

- amp started distorting alot around 4 on the dial.

This might be normal, depending on how you incorporated the 12AX7 gain stage. Compared to a Standel amp, a Fender Princeton gets slightly fuzzy around 4-5 on the volume control. There are ways to dial back gain when using a 12AX7. A better option is the 12DW7/7247 like I mentioned earlier.

-Lot's of noise(buzzy hum)

Not normal. Suggests a wiring error somewhere.

I'm pretty sure only one half of the 2nd 12ax7 is functioning as a Pi and the other half is another gain stage

Yes, but the split-load inverter provides no gain of its own. It needs the pre-gain stage to allow a reasonable ~1v RMS signal from the preamp to drive the output tubes.

Your output tube bias is not provided on the Gulbransen schematic, but you have EL84's so maybe 10-11v is a reasonable guess. For the split-load to deliver 10v peak out to each EL84, it needs 10v peak applied to its grid. That may be more than a the average preamp can deliver cleanly. So the 12AX7 stage ahead of the split-load boosts the signal to enable the preamp to drive the output section.

OPTION: Replace the 100kΩ resistors on the split load plate & cathode with 22kΩ. Yank the 12AX7 and install a 12AU7. The pre-gain and split load now accept larger input signals without overloading, and the pair doesn't amplify so much (obviously the EL84's are getting more than enough drive in your application). The 680Ω cathode bias resistor for the 1st 12AU7 may need adjusting upwards, but I'd want to see voltage readings at all pins to understand how the circuit reacts to the changes mentioned so far.

Also, the stage ahead of the split-load inverter has feedback returned to its cathode. What did you do with that when you removed that stage? Or did you remove that stage? Omitting the feedback makes the output stage easier to distort, which you probably don't want. Making the 68kΩ series feedback resistor too small could boost feedback so much that the output stage oscillates; however there should be some room to experiment with its value and amount of feedback.

[Toxophilite]

Hi
I'm just looking at a AA964 Princeton schematic right now
it has 2 12AX7  2 6V6s and a GZ34


there's input - 1/2 12AX7 - tonestack and volume- 1/2 12ax7 - 1/2 12ax7 PI then into the 6V6s
The other half of the 12AX7 is used for a vibrato circuit


So there's 3 12ax7 stages prior to the power tubes one of which is the PI


THe other is the vibrato circuit which works on the bias of the power tubes and isn't part of the preamp as far as I can see


Are we thinking about different Princeton reverbs??


 

and in case it's been misunderstood, none of the problems/symptoms mentioned manifest themselves when it's wired like the Princeton below(minus the reverb)

[Toxophilite]

My preamp wiring looks just fine
THe 12ax7 is wired like a standard fender front end 1.5K and 25/25caps on the cathodes, 100K resistors on the plates. I believe I checked the voltages and they seemed good. I'll check again (i did that a month or so ago)
Everything else is as the schematic included in my 1st post


One thing that could be making this amp VERY loud and a little out of control when given lots of front end gain is the fact that it has a 68k resistor on the negative feedback which seems to me to be huge. The Princeton has a 2.7 K which is obviously letting a lot more through resulting a tamer, much more civilized amp.
I just checked this right now as I hadn't thought about it before
The sound of this amp seems to bear this out as it has all the pretty high end that you get with less negative feedback

[Willabe]

For that matter, I have never looked to see if there was any appropriate op iron available with taps.

I bet they've been used in hi-fi.


                  Brad   

[shooter]

I'm looking at a AA964 Princeton schematic right now it has 2 12AX7  2 6V6s and a GZ34
there's input - 1/2 12AX7 - tonestack and volume- 1/2 12ax7 - 1/2 12ax7 PI then into the 6V6s
So there's 3 12ax7 stages prior to the power tubes one of which is the PI
and in case it's been misunderstood, none of the problems/symptoms mentioned manifest themselves when it's wired like the Princeton below(minus the reverb)

Previous quote:

Switched it back to input - 1/2 12ax7 - tone stack and volume - phase inverter
I'm pretty sure only one half of the 2nd 12ax7 is functioning as a Pi and the other half is another gain stage which makes sense from the results I was getting
Sounds great set up this way and is Very loud with minimal distortion

I'm following, trying to gain some knowledge but I got off at the wrong stop, I think.

So you get good sound from just 1/2 12ax,TS,Vol, PI, and you don't have problems when wired like the schematic?  what configuration causes the thumping?

[Toxophilite]

Hi
there has been much reiteration so it is getting confusing


I think the problem is the amp is designed with very little negative feedback (see schematic post 1(Fig 2-9 'tube type amplifier'). Likely it gets fed a very quiet transistor signal, So if I put 2 gain stages (one 12AX7) in front of the existing amp  It gets hugely loud and unstable
which to me
makes sense now
I might try swapping out the NFB resistor


Though with just the one gain stage it sounds pretty damn cool.(and that's clean)

[Toxophilite]

I think also the Princeton reverb topology is a moot point as it actually pretty different in some respects
Just confusing the issue



THis gulbransen amp has another curious feature that perhaps someone can explain
It's curious to me because I've not seen it before


just so you know I've not done anything to the existing part of the amp shown in the schematic.

on the 1st stage of the PI 12AX7 the cathode (which the NFB loop attaches to) is tied to the 12AX7 heaters with a 680 ohm resistor.

Why is that?

[sluckey]

on the 1st stage of the PI 12AX7 the cathode (which the NFB loop attaches to) is tied to the 12AX7 heaters with a 680 ohm resistor.

The schematic doesn't show the filament circuit. But that 680Ω cathode resistor must be connected to ground. I suspect that one side of your heater circuit is also tied to ground. That was common practice back in the '50s. If that's the case then the grounded filament pin is just a handy tie point for the 680Ω cathode resistor.

Here's a schematic for a Champ that shows the grounded filaments...

     http://el34world.com/charts/Schematics/files/fender/Fender_Champ_5E1.pdf

[HotBluePlates]

Hi
I'm just looking at a AA964 Princeton schematic right now
...
So there's 3 12ax7 stages prior to the power tubes one of which is the PI
...
Are we thinking about different Princeton reverbs??

My mistake!!!  The AA964 Princeton is indeed the circuit I was envisioning, but I remembered wrong, because the Reverb model has an extra gain stage. But that's not what you want, you want the non-reverb model.

Looks like you had the correct setup all along with only 1 gain stage added to the Gulbransen. I'll be sure to actually look at the schematic I refer to in the future and not rely on (faulty) memory!!

The Princeton Reverb has a bit more gain than the non-reverb model, but the reason it doesn't have excessive gain like you experienced is there is a big loss network between stages in that model, which facilitates the mixing in of the reverb signal.

One thing that could be making this amp VERY loud and a little out of control when given lots of front end gain is the fact that it has a 68k resistor on the negative feedback which seems to me to be huge. The Princeton has a 2.7 K which is obviously letting a lot more through ...

1st problem was I boogered.

But Re: the feedback, you cannot just compare the series feedback resistor (68kΩ vs 2.7kΩ). You also have to look at the shunt feedback resistor, which is the resistance between the series NFB resistor and ground (680Ω vs 47Ω). The 2 resistors define a voltage divider which set the amount of feedback.

What happens is for a given output power, there is a set a.c. voltage across the speaker. The series & shunt feedback resistors divide this voltage to a lower amount, then apply it to an earlier point in the amp.

In the Princeton, a good bias resistor was chosen for the stage where feedback is injected, and gain of the stage was increased by adding a cathode bypass cap. To add NFB without disturbing that arrangement, a discrete resistor (47Ω) is used as the shunt feedback resistor, and this is inserted between the existing cathode resistor & ground. In the Gulbransen, the extra gain of a cathode bypass cap was not needed/wanted, so the existing cathode resistor was used as both a cathode resistor & a shunt feedback resistor.

We can estimate the gain from the stage where feedback is injected through to the speaker by using R_series/R_shunt. For the Princeton that's 2.7kΩ/47Ω = ~57, and for the Gulbransen it's 68kΩ/680Ω = 100. So the Princeton appears to have a bit more feedback (a factor of 2), but it also uses output tubes which amplify less (mu ~10 for 6V6, Mu of 19.5 for EL84, or a factor of 2). So really the amount of feedback is likely pretty similar; to know for sure we'd need to calculate open-loop gain for both amps and then the gain after feedback, and the ratio of the two is the applied feedback.

I'd not be inclined to tinker with the Gulbransen's feedback, but if I did I wouldn't drop the 68kΩ below 20-30kΩ without expecting oscillation problems.

Sorry for leading you astry on the extra gain stage thing! It looks like you'd already landed at the sweet spot for clean tone.

[Toxophilite]

Thanks S.Luckey
Strangely this amp is likely from the late 60s. the organ was a transistional one which had a tube power amp but everything else was solid state
However I just noticed that the heaters are tied to ground at that point(it was a little hidden) So there you go


And Hot Blue PLates , THanks for the explanation regarding the NFB
My ignorance was revealing itself in all it's glory!


And regarding the Princeton, I find email and communication via the internet is often fraught with misinterpretation and confusion, No big deal.


One thing that I had messed up on good was, I believe I forgot to check the plate voltages before(or I did and forgot a month ago) So I was getting only 65 vdc on the plates of V1 and about 285 on my El84s
I've been monkeying with the power supply dropping resistors and getting some results(particularly for V1) that are more in line with what I would expect.


THanks for everyone's patience and help as i stumble through this.

[Toxophilite]

Hey S. Luckey


When you made your 6V6 plexi in the AO-43 chassis did you keep the power tubes in the same spot but just widen the holes?
That's what I was thinking, and move the choke over to power supply area
(for a deluxe build)

[sluckey]

I didn't use the AO-43 chassis. My 6V6s are on 1 3/4" centers.

[Toxophilite]

Right I understand. You used the parts, not the chassis
THanks

[brown4308]

I have gutted and built two matchless lightnings using ao43s, they run with higher b+ but both of mine with nos 5y3 are at 340v. These were my first two builds actually. Just be prepared for a tight layout! I have no regrets.