• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

Valve PA amplifier - what to do with it?

Drawn the circuit (as I see it) out by hand. First time I've drawn a circuit so go easy on me 😁.
A lot of the resistors are quite difficult to find using the old colour codes. The values I've written are as they tested, I have no idea how much hey haven't have not drifted! Clearly there's a lot of variation, some identical pairs test differently. There's also a few capacitor values in not sure of.

I don't know if the ECC81 and ECC82 are in the correct positions, they might need to swap around.

Not sure if I've drawn the large power supply capacitor correctly, it has a common connection to ground on the bottom.

I'm slightly confused about the phase of the gramaphone signal when it enters the grid of the first half of the ECC81. It enters here with the microphone signal which has already been through a stage and will have reversed phase compared to the not yet amplified gram signal.
Unless the microphone transformers (not drawn on diagram but discussed earlier in the thread) have already reverse the phase of the signal? In which case the first stage would have changed it back making it in phase with the gram signal?

Please feel free to correct me if I've misinterpreted or missed anything out on my drawing so I can amend it, I'm a beginner! If any resistor/capacitor values look way off I'd appreciate a helpful point in the right direction.

Some of the caps say 0.15, is that uf?
Also, In terms of what type of cap to replace these old ones with?

IMG_20231010_233210572_MFNR.jpg


IMG_20231010_233325725_MFNR.jpg
 
  • Like
Reactions: ChrisABC
The caps, like 0.15, would be in uF. Experience and exposure to a range of circuits and their design can quickly indicate if pF, nF or uF, based on the application.

The 0.22uF coupling caps from the first stage anodes to the pots would connect to the anode terminals, not the power rail.

Phasing of input signals is not a concern, as the signals are independent of each other.

Does the HT rail feeding the input stages (ie. after the 47k feed from HT+) have a power supply decoupling cap to ground ?

That is a negative feedback connection - so is correctly shown in your sketch, and that total output section is quite typical with its feedback stage followed by concertina phase splitter to push-pull output stage.

The EL34 anode resistors are likely failed if they measure 99k and 65k, as they were likely circa 100 ohm and used as an 'anode stopper' to avoid parasitic oscillation. The 44k screen dropper for the EL34's seems a little high for practical use, so that resistor may have degraded.

The power transformer HT secondary is not correctly shown. Each pin 1 anode is fed from a balancing 150R to a HT winding arm. Likewise the pin 7's would go through balancing resistors to another HT winding arm, with the HT winding having a CT connection going to 0V. The 150uF dual filter cap would have one side going to before the choke, and the other side going after the choke, and each side's negative terminal would be the common terminal going to 0V.
 
  • Like
Reactions: peacocksuit
When I had a similar nut to crack I made myself get to grips with LTSpice. With a little practice it becomes a great tool for creating and editing a schematic. It is very easy to shuffle whole stages around, for instance. Then if you are prepared to go to the next level, with a moderate amount of effort you can simulate the circuit, and verify it works correctly (verifying the schematic). At that point you have a baseline for testing planned changes.
 
  • Like
Reactions: peacocksuit
Something less powerful that has all the symbols needed to draw an amp schematic would be good, not bothered about simulations. LTspice seems unfathomably complicated. I wasted a lot of time trying to find the right directory to add valves into. From memory it was in a wierd place and hidden. I think it's designed with older versions of Windows in mind maybe. Very tedious.
 
When I had a similar nut to crack I made myself get to grips with LTSpice. With a little practice it becomes a great tool for creating and editing a schematic. It is very easy to shuffle whole stages around, for instance.
I doubt i could get a version of LTSpice into this iPad. And it’s the only computer I use for DIYAudio.com or anything related. (How to keep malware off your financial computer? Don’t use it to PLAY - just business.) So most stuff posted here get hand drawn. I do ALL my tube design by hand anyway, and a good bit of the solid state. Old habits, dating back to 8088 (or even TRS-80) times. Eventually things make it into a real computer, if they go anywhere.

Hand drawn is fine by me - usually it indicates (to me anyway) the person is a thinker who doesn’t blindly accept any answer the computer throws your way. Some of our best designers at work were buried in paper in their cubes.
 
  • Like
Reactions: peacocksuit
The caps, like 0.15, would be in uF. Experience and exposure to a range of circuits and their design can quickly indicate if pF, nF or uF, based on the application.

Got it.
The 0.22uF coupling caps from the first stage anodes to the pots would connect to the anode terminals, not the power rail.

IMG_20231011_115703343_MFNR.jpg


Sorted?
Phasing of input signals is not a concern, as the signals are independent of each other.
That makes sense.
Does the HT rail feeding the input stages (ie. after the 47k feed from HT+) have a power supply decoupling cap to ground ?
IMG_20231011_115052559_MFNR.jpg

There is yes, I've added this to the diagram, hopefully this looks right now.

That is a negative feedback connection - so is correctly shown in your sketch, and that total output section is quite typical with its feedback stage followed by concertina phase splitter to push-pull output stage.
Got it.
The EL34 anode resistors are likely failed if they measure 99k and 65k, as they were likely circa 100 ohm and used as an 'anode stopper' to avoid parasitic oscillation. The 44k screen dropper for the EL34's seems a little high for practical use, so that resistor may have degraded.
I'll re check these.
The power transformer HT secondary is not correctly shown. Each pin 1 anode is fed from a balancing 150R to a HT winding arm. Likewise the pin 7's would go through balancing resistors to another HT winding arm, with the HT winding having a CT connection going to 0V.
IMG_20231011_114332071_MFNR.jpg


I didn't draw this initially but I've added it in, should be right now?
The 150uF dual filter cap would have one side going to before the choke, and the other side going after the choke, and each side's negative terminal would be the common terminal going to 0V.
This part is alluding me somewhat. I'm just not sure how to represent that on the diagram.
This is how it looks actually:

IMG_20231011_114023150_MFNR.jpg


And how I've drawn it (which I know is wrong...)

IMG_20231011_115825351_MFNR.jpg



Thanks everyone for your input. I'm having fun with this! Learning a lot.
 
I'm starting to think about replacing some of the parts. I think we've seen that there are plenty of of spec resistors. I have loads of resistors so I'm not too concerned about them.

Capacitor wise, the DC blocking caps look like they've been replaced as the solder points look fresher. These are Wima (polyester?) type. Doing some reading up it seems that these are quite reliable and age ok. I'm wondering if these could be tested before powering up and night be usable? I have a Variac as well so I can bring the amp up to voltage slowly the first time I energise it.

IMG_20231011_200244974_1_MFNR.jpg


There are caps on the output transformer between the primary and secondary side, I'm not entirely sure what these do if I'm honest, I'm guessing these will need replacing? They're not branded (as far as I can see), I'm not sure what type they are. The same type are also used the bass/treble circuit. Guessing these will need replacing?

IMG_20231011_200412918_MFNR.jpg


And then there's the green hunts capacitors (electrolytic?), I've read a lot about these being notoriously unreliable so I guess I'll be replacing these. The big reservoir cap is also hunts so this as well. I assume this will become 2 separate capacitors as it's currently a combined type.
 
Big electrolytics that old are probably in bad shape. Might still “work”, but low in capacity and high in ESR. What you usually end up doing is leaving the original cap on the chassis, disconnected and wiring two or three new ones under the chassis.

I was wondering about those Wima caps - probably not originals. Polyester or polypropylene of that quality just never go bad.
 
  • Like
Reactions: peacocksuit
There are the 'usual suspects' that have a deservedly bad reputation, like brown Hunts caps and Rifa polyester ones. Then there are old electrolytics which split the vintage radio world, with many collectors trying to preserve their equipent and therefore giving them a fair trial before signing a death warrant.

That one looks like a polyester cap and ought to be fairly bullet proof.
 
  • Like
Reactions: peacocksuit
There are the 'usual suspects' that have a deservedly bad reputation, like brown Hunts caps and Rifa polyester ones. Then there are old electrolytics which split the vintage radio world, with many collectors trying to preserve their equipent and therefore giving them a fair trial before signing a death warrant.

That one looks like a polyester cap and ought to be fairly bullet proof.
Maybe that was replaced at.some point when the Wima's went in. Interesting. Might have been my grandad for I know! Maybe he had this working at some point. I have literally no history on this thing at all. No idea how or why he had it. He was pretty into electrical stuff. He passed away over 20 years ago unfortunately so I can only guess.

These look newer also:

IMG_20231011_212326920_MFNR.jpg
 
Your schematic revisions look ok. Imho as long as a sketch is accurate then that is the main issue - no imperative to make a sketch into a formal schematic, but it can be worthwhile copying the sketch to use on the bench to add voltage readings and highlight if operation for a particular circuit section is ok or not quite right or changes much if you swap tubes.

You may need to do some comparisons with other amp circuitry with similar stages and B+ voltage, that also have known bogey operating dc voltages, as a way to check what you measure in your amp given there may be doubt on some actual resistor values and tube types. Confirming the choice of a particular tube type can get detailed and involve preparing a loadline and looking at valve datasheet characteristics to indicate if a particular model of valve is being biased ok and can amplify and achieve symmetric clipping.

Given you have a variac, then there is a procedure for raising the B+ level and checking that the coupling caps are not leaking (by measuring the dc voltage on the grounded side of the cap when the following valve is removed). That is a practical check that should imho be made whether you keep or replace particular coupling caps. In addition that procedure can also be used to reform e-caps and confirm they have an acceptably low leakage current. But that assumes you are ready to start energising the amp and have upgraded all the relevant mains safety aspects, and have a list of what dc voltage levels you can bring sections up to. The doc in post #8 has a procedure description in the testing section that can be adjusted to suit. Or alternately you can just turn it on and hope for the best 🙁
 
Definitely won't be just turning it on with my fingers crossed... I have read your pages in part before in the past bit have started having a look through again to go over safety upgrades etc. I will use the info on testing procedures as a guide when I do energize it. Some great info in there but there's quite a lot to go through!

I was initially thinking of an IEC socket on the chassis for the mains (chassis earthed ofc). But I'd prefer to have it on the rear, as well as a TRS socket(s) for the speaker output(s). The issue is, the chassis lifts up and away from the wooden enclosure. So, if the socket was located on the rear of the enclosure, I'd have to have enough cable on the inside between the mains socket terminals and the place where it terminates on the chassis to allow the chassis to be lifted away when necessary i.e. for maintenance. I was imagining this could just be neatly dressed into the space at the bottom on re assembling each time. I thought about the Live neutral and earth wires using spade type insulated crimps to terminate onto the IEC socket so the chassis could be completely separated from the enclosure when necessary after lifting out/away.

But, why not just hard wire it with a suitable 3 core flex, with a pass through grommet in the rear or side of the enclosure? Whenever the chassis was lifted out of the enclosure for maintenance, the cable would be able to move freely though the grommet to allow this. Then I'd only have to take the plug off if I wanted them completely separate. If working on the amp if would be nice to remove the enclosure completely although there would be enough length on the cable to just have them side by side on the bench.

Does the IEC socket offer any other advantage over a hardwired cable?