Any ideas for a PA conversion?

[HotFat]

Hello all!

I'm considering a conversion on a 2 x el34 PA. The PT is dated 1969, but there are no manufacturer markings on the amp other than “Sound 50”. I found this thing left out in the street just as I'd been thinking about building another amp. I plugged it in and it works. I'm going to call it the Manna amp.

I'm fairly inexperienced. I've built the Dave Hunter two stroke amp from his book, serviced my JCM 800 replacing the filter caps and followed instructions to install a neg feedback loop on my Pro Jnr. At the moment I think get the big ideas but I need to work on the maths. For the next few weeks I'm reading all I can to help me pull this off. I've got some ideas but I could do with some help from the experienced members of the forum to sort out what's worth pursuing and help me identify those “unknown unknown's” that I need to read up on.

I'm making notes of the current layout, and attempted to map the schematic, (I need practice at that, it looks like a game of kerplunk!). I'm planning on replacing the circuit board, but I realise that at some point I might want to know what I just ripped out.

 The OT is a decent size, It's marked 116, has a pair of white leads each connected to pin 3 on each el34, a pair of blue leads to pin 4 and a single red lead to the PT via the solid rectifier. There are 3 outputs to 15, 8 & 3 ohm speaker jacks and a ground wire.

The PT outputs appear to use two solid red metal wires with black insulation sleeves connected to heaters, 2 dirty looking white wires to the solid state rectifier and a green and a white wire both running to ground. 240V inputs are brown & orange.

The thoughts I have at the moment are:

1) Can this be built as a dual single ended EL34 amp? I'm looking for around 20 watts and I like the Victoria Regal idea, I love swapping valves in the two stroke. Would I need a new OT and if so what sort of model am I looking at?

2) If dual single ended is unrealistic, I'm thinking about a cathode biased el34 with a long tailed pair PI, cathode follower tone stack, master volume and switchable negative feedback loop, possibly using volume reducing ideas like http://www.el34world.com/Forum/index.php?topic=4401.50. Any recommendations for schematics / layouts you have used that would give me some sparkle and gradual onset of distortion?

3) With this kind of project am I limited to using el34, or could I consider 6l6?

Any thoughts on the original PA, ideas, links or encouragement would be most welcome!

Thanks
Hot Fat
London

[sluckey]

Good find! That's an excellant platform for a Marshall Plexi or Matchless Clubman (push/pull cathode bias) or any of several other 40-50 watt push/pull amps. Your OT is a push/pull ultralinear type and is not suited for any single ended designs. You could use 6L6s but I'd stay with EL34s. You already know the transformers work well with EL34s.

Before you tear it down, measure the AC voltages for the PT. Measure DC voltages on each filter cap. Measure voltages on every pin of both EL34s. And record those readings.

[smackoj]

Post script: I see that one of our leaders, Sluckey, posted a couple seconds before me and I'm glad to see we both hardily agree on keeping the 34s and shooting for the beloved Plexi stars....love them Plexis.   

My orig. post starts here: It looks to me from viewing the pictures, that you have basically a tube power amp....the look is quite a bit similar to a Dynakit Mk III. My suggestions would be like this:

A. rebuild it as-is until it's working correctly and quietly, and use it as a power amp, i.e. don't do anything to build into it a fender, marshall etc. preamp section. Then build or find yourself a nice guitar preamp and hook 'em up for your guitar rig.
B. do the same as A. only resell the amp as a mono tube amp on flebay. Some of them go for 2-400 dollars.
C. if you are determined to make a guitar amp out of it, I would suggest keeping the EL34s and making a Marshall JTM45 or similar Plexi style Marshall build.

cheers    

[HotFat]

Thanks for your advice guys. I'll take the suggested readings asap.

I can see the logic in selling it on, but from the moment I found it I couldn't stop thinking of the possibilities. I doubt I would be as happy selling it and working from scratch, I like the back story and the limitations it places on me.

I'll keep you updated

Hot Fat

[1rebmem]

Having built 4 guitar amps from old PA's, I have some advice.
First, build the EL34 amp you desire.
Find a schematic you like.  Strip the wiring and components from the chassis, you can reuse components if you verify they are good, i.e. resistor drift and cap values/leakage. I personally never re-use electrolytic cap's. Replace the tube sockets, I've found they can be problematic and yours are very weathered looking. Replace the circuit board with a turret board or install an aluminum cover in place of it etc. I've never used turret strips, only solder strips to build mine since the chassis already had them.
Then add a Hall Amplification VVR ($32) to scale down the amp output. They work good.

The expense of building an amp is the tubes, iron and chassis. Resistors, capacitors, sockets and jacks are fairly cheap. 

[HotFat]

Thanks for the advice 1rebmem. This is really useful! Without this forum I'd be trying to build a swiss army amp!

[Scsoul]

   That's sweet, with that circuit board it makes it easy to switch pre-amp you could set up
a breadboard type pre-amp just get the output section working right, get some board building
supplys mount the sockets on the board and build anything you want  looks sweet     scsoul

[PRR]

> basically a tube power amp....

Lot of knobs and more little-tubes than you suspect.

I'm thinking 3 mike-amps, a mix-stage, and a driver.

Document the power and bias supplies. Leave the power bias EL34 and OT wiring alone, that stuff never varies (much).

It is VERY possible that, after leaky caps are fixed, you can jack guitar into a "MIC" input and play. But cheap mike inputs tend to overload, and you don't need three inputs.

I'd get real tempted to trash the PCB, mount sockets on a turret board, and build (or clone) anew.

[drew]

The chassis is Hiwatt-ish, with the transformers at the opposite ends, the power tubes and pre-amp tubes in different rows, and the power tube heat-shield.  I'd poke around Mark Huss' site for inspiration.  E.g., http://hiwatt.org/pix.php?p=DR504_81bChassis

[HotFat]

"I'm thinking 3 mike-amps, a mix-stage, and a driver."

Absolutely right.  There are 6 inputs across 3 channels, two high channels and one low. Each input has its own volume. There is a 12au7 that handles the two high channels then another two 12ax7's. There's a treble and a bass control which push up the volume, and the channels can be jumpered.

The amp works, not really loud but very clean and no filter cap issues! I'm guessing it wasn't used for years, but the caps must be 40 years old so I'd expect the caps to be shot.

There's what looks like a bridge rectifier on there and I'm leaning towards leaving that and the EL34's alone as suggested. The number of holes that will be left when I remove some of the jack inputs will allow some easy switches.

Just a question of what to put in as a preamp. I was looking at the Matchless Clubman as per Sluckey's post, but I'm wondering if that would work with the stock power section. I want something that gives a bit of chime and sparkle and gradually distorts. The HiWatt thing interests me but I always think of them as being really LOUD before breakup, although that Townsend sound is amazing. Any other ideas for a preamp?

I'll get some readings done tonight.

[sluckey]

There's what looks like a bridge rectifier

I don't think it's a bridge. You have two diodes in series connected to one PT wire and two more diodes in series connected to another PT wire. And the cathodes of those two series strings connect together and go somewhere on a red wire. Using series diodes increases the voltage rating. Those old top hats rarely had a rating over 400V, maybe 600V on high dollar diodes.

[HotFat]

Thanks for all your help. I've checked the voltages.

Power Transformer:

Green & white wired to ground
Red 3v AC
Dirty White 360v AC

Red wire from cathodes of diodes in series to OT 485v DC
White wire from diodes to cap in can 482v DC
Voltage between 5th top hat diode and 3 black resistors 179v DC

Caps
Can cap to ground 370v
two caps above can cap 36v (both in series, measured from top to bottom of each cap)

Left orange huge filter cap (to ground) 474v
Left orange huge filter cap (across) 228v
Right orange huge filter cap (to ground) 0v
Right orange huge filter cap (across) 242v

16uf cap on board (next to 12AU7) 161v
16uf cap on board (middle) 204v
16uf cap on board (left) 255v

EL34  nearest PT
pin 1 & 8 - 10v DC
pin 2 - 2.6v AC
pin 3 - 481v DC
pin 4 - 482v DC
pin 5 & 6 - 21v DC
pin 7 - 3v AC

EL34 nearest OT
pin 1 & 8 -10v DC
pin 2 - 3v AC
pin 3 - 476v DC
pin 4 - 479v DC
pin 5 & 6 - 21v DC
pin 7 2.6v AC

Thanks
Simon

[HotFat]

I've mapped out the connections and I'm still thinking about what board to put in. There is a component that I can't place though. It sits between pins 1&8 of the el34's and ground, but I can't identify it so I've no idea how to test or replace it?

[eleventeen]

That would appear to be a small-value resistor, probably 1 ohm or 10 ohms, that we like to have holding the cathodes of the output tubes off ground to make it easy to measure current through the output tubes. Exactly as you described. We adjust bias to achieve a nominal amount of idle current through the output tubes.

Across a 1 ohm resistor, 10 ma (10/1000 = 1/100 amp) of current will show up on a meter as .01 volt. We do not like to measure current directly...it is certainly possible, and newer DVMs handle it OK, but in the old days of analog meters, it was a sooner-or-later thing of blowing up your meter if you were in the habit of trying to measure current and mismatched the range of the meter to the measurement task. Run 10 amps through your Simpson 260 and you were likely to flame it before you realized what happened...and maybe bend the meter pointer, to boot. To measure current, you also have to interrupt a given circuit. Given that the resistor you picture is hanging out there like that and connected to the EL34 cathodes, that's my guess as to what it is. Nice feature, frankly. I wouldn't necessarily mount it in that fashion but it's a nice feature. Most folks mount two tip-jacks under the chassis so that the meter probes can be stuck in the tip jacks, they will hold themselves there, and then you go to the bias control and tweak away. You can also use three tip jacks to measure the individual currents through either one of the EL34's.

Personally, I usually use a ten-ohm resistor as the cathode hold-off resistor because when you are talking about .01 volt, the resistance of the test leads themselves starts to matter. But same idea.

[jjasilli]

Those trannies appear to be quality; full audio frequency spectrum; and heavy, especially for a guitar amp.  These factors tend to favor PRR's suggestion.  Could also be used for mono audio amp, such as the center-channel tube power amp in a home theater system.

[HotFat]

Once again thank you all for your advice.

I'm planning on going for a JTM45. I'm wondering whether to keep the diode rectifier circuit and the existing chassis mounted filter caps? Would you rip them out and go with a GZ34 and new caps or just put the preamp in and hook it up to the filtered power supply?

[sluckey]

I don't think your PT will support a GZ34 rectidier.

[HotFat]

Thanks for the advice sluckey. I think I'm mostly there now!

If I keep the original power stage, rectifier and filter caps it looks to me like there are three connection points on the replacement board, (plus the heaters, bulb, neg feedback loop and ground).

I've mapped out the power stage and modified a layout of the JTM 50 board to show where the I believe the 3 (X, Y & Z) connections are. I apologise for my skills mapping the power stage but I hope it makes sense. I also put up the original JTM 50 layout for anyone wanting to see what I've wiped out.

Assuming that it's basically correct, the next issue seems to be that the B+ entering at Point Z is reading at 368v, but the 5F6-A schematic I've used as reference is showing 385v. Is this something I can overcome by changing the values of the 10k 2W resistor before the second gain stage and the 100k & 82k resistors leading to the PI?

Thank you all for your advice. I do appreciate it!

[HotFat]

Thanks to all the advice on this project. In the end I went with a JTM50 / Plexi cross. Switches on the front panel allow me to split the cathode between JTM50 and Plexi specs, switch between 0.1uf and .022uf PI coupling caps and between 47k, 27k and no neg feedback/presence control. Round the back I have the LarMar PPIMV and Pentode / Triode switch. I also rewired the PT to non ultra linear.

It sounds amazing!

Cheers