Small Push Pull tube amp build, advice sought.

deltafred · Post by **deltafred** » 23 Nov 2020, 23:37

I need a bit of advice and guidance from the valve (tube) gurus.

I’ve been fixing valve gear since before solid state was available (mid 60’s) and have built a load of SS amps and even one class D but never a valve amp.

I recently discovered I had an old push pull output transformer (marked P4133, possibly an early Partridge P4133, probably late 50's or early 60's).

I also have a couple of suitable looking mains transformers (250v @ 60mA + 6.3v @ 3A and 260 – 0 – 260v current unspecified with 6.3v @ 2.5A), a Celestion green back, a spring line reverb unit and a load of suitable plywood and sheet metal so I thought why not?

Q1. Would the P4133 be ok to use with a pair of EL84s or 6V6s?
The Mullard Circuits For Audio Amplifiers book shows the P4133 used with 2x ECL84 (for 8W) and the P4131 with 2x EL84 (for 10W) but these are intended for Hi-fi use so probably not that relevant.

Specs from Partridge datasheet
P4131 8k a-a 43% taps max 10W
P4133 8-9k a-a, 20% taps max 8W

Q2. As as far as I can remember none of the guitar amps I have worked on have used screen grid taps, should I just ignore these and use screen grid resistors?

Q3. Would you recommend a single or double valve Phase Inverter (can’t remember the proper names for them).

Q4. Which tonestack would you recommend?

I’m thinking of building something along the lines of an AC15 / Marshall 18W / Fender blues Junior for use when playing slide guitar (acoustic guitar with piezo under the saddle and basic high impedance buffer).

The proposed spec is -
Single input
Some mild OD without the use of pedals, possibly foot switch controlled
Hopefully EL84 output valves (as I have quite a few)
SS rectifier/s
12” Speaker
Possibly tremolo and/or spring line reverb
Cab and chassis design open to suggestion

What are your thoughts and criticisms please?

Don’t expect to see results immediately as I have a ton of unfinished projects but if I get fired up it may happen sooner rather than later.

ppluis0 · Post by **ppluis0** » 24 Nov 2020, 02:09

Hi Deltafred,

I'm not exactly an tube guru but an authentic tube enthusiast, so there are my opinion:

Answer 1: Transformer P4131 looks OK to be used in a guitar amp project as the frequency range is lower compared to hi-fi and the core size must be enough (IMHO)

Answer 2: Yes. Leave those taps floating and connect the screen grids as pentode with suitable resistors.

Answer 3 and 4: Look at the following diagram:

Replace the rectifier tube with a couple (or diode bridge if the power transformer does not have center tap) of 1N4007 and increase the size of filter caps to your taste.

If you wish to include a reverb you can drive its input (attenuated with a suitable resistor ! ) directly from the secondary of the output transformer.
From the output of the reverb tank build any kind of reverb preamp using tubes op amps or transistors, and send its output to pin 7 of the second 12AX7
Place a potentiometer after the mentioned recovery preamp, lift the 100nF condenser (at pin 7) from ground and connect it to the pot slider.

If you like to have also tremolo, look at this diagram: http://www.prowessamplifiers.com/schema ... icane.html

Use a half of a 12AX7 to act as tremolo oscillator and the other half to amplify the reverb pan output.

In a further post we discuss cabinet options

Cheers,
Jose

Manfred · Post by **Manfred** » 24 Nov 2020, 07:05

The P4133 OT is designed for 8 watts and cannot be used as a push-pull amplifier in Class A, AB and B according to the standard circuits with the El84 and 6V6 tubes.
The transformer would go into saturation early or the winding is not designed for the currents at 18W.
One possibility would be to design the circuit for 8W.
This means to use a lower operating voltage or to operate the output valves in triode mode.
Since the transformer with the 260V-0-260V is available, the triode mode would be the best solution.
In order to achieve the required Raa on the primary side, it is necessary to check to which secondary terminal the loudspeaker is connected with the given impedance.

Partridge info.pdf: (3.27 MiB) Downloaded 155 times

deltafred · Post by **deltafred** » 24 Nov 2020, 10:42

Thanks for your rapid responses guys.

I was afraid that the P4133 wouldn't be up to the job but don't mind limiting the output power if I can still use it. The whole idea is to make something useable out of bits I have lying around.

When I clear some jobs off my bench I will breadboard it. I can power the heaters from a separate transformer and hook the HT transformer to a variac and put a scope on it to see what voltage works best.

mozz · Post by **mozz** » 24 Nov 2020, 13:23

Nothing wrong with the output transformer. Post a picture of it showing size.

soulsonic · Post by **soulsonic** » 25 Nov 2020, 01:39

I'm very pleased with the preamp and phase inverter section of my re-built 6V6 mystery amp. Check out the second schematic I posted that shows how I rebuilt it. I used Merlin's book as a guide and I recommend it.
You can then design the power supply and power amp to work with the voltages you have available with the power transformer, as per the suggestions others have already given.

deltafred · Post by **deltafred** » 25 Nov 2020, 11:52

soulsonic wrote: ↑25 Nov 2020, 01:39 I'm very pleased with the preamp and phase inverter section of my re-built 6V6 mystery amp. Check out the second schematic I posted that shows how I rebuilt it. I used Merlin's book as a guide and I recommend it.
You can then design the power supply and power amp to work with the voltages you have available with the power transformer, as per the suggestions others have already given.

Reading your thread was a big part of the inspiration to do this. Another was that in the last 10 years I have repaired quite a lot of low wattage guitar amps and discovering that I had an OT, which I had always thought was a mains transformer, clinched it.

My kids are always stuck for things to buy me for Christmas so I will drop a few strong hints about Merlin's book. I have a good basic knowledge of valve theory, when I started at college they were still teaching valves, and since the mid 60's have repaired countless valve amps (and radios and TVs when they still used valves) but have never delved into modifying them. I see a new rabbit hole before me!

deltafred · Post by **deltafred** » 25 Nov 2020, 12:47

mozz wrote: ↑24 Nov 2020, 13:23 Nothing wrong with the output transformer. Post a picture of it showing size.

The outer dimensions of the laminations are 76mm x 64mm x 25mm thick (3" x 2.5" x 1")
The core is 28mm approx* x 25mm x 37mm long. (1,1"* x 1" x 1.5" long)
*This is hidden by the former so could be less.

The former is ink stamped P4133 which is also hand written on the outer casing.

phatt · Post by **phatt** » 29 Nov 2020, 13:42

deltafred wrote: ↑23 Nov 2020, 23:37 I need a bit of advice and guidance from the valve (tube) gurus.

Specs from Partridge datasheet
P4131 8k a-a 43% taps max 10W
P4133 8-9k a-a, 20% taps max 8W

Q2. As as far as I can remember none of the guitar amps I have worked on have used screen grid taps, should I just ignore these and use screen grid resistors?

Q3. Would you recommend a single or double valve Phase Inverter (can’t remember the proper names for them).

Q4. Which tonestack would you recommend?

Re this;
"Specs from Partridge datasheet
P4131 8k a-a 43% taps max 10W
P4133 8-9k a-a, 20% taps max 8W"
Umm,, you will find they are talking HIFI Whatzezz so that is Clean watts before clipping.
So highly likely they can handle more.
My guess is 8W clean and likely easy handle 12~14 Dirty guitar watts.

If they are are hifi then likely interleaved for better fidelity so they might be a bit bright. Most of the older small wattage Guitar amps used basic and cheaper Tx's with limited BWidth so these might make the treble brittle,, suk and see I guess.

Re Q2 agree not used for guitar but I believe you can use the taps to lower the primary,, others might know more on that?
Q3, I believe the term is LTP (Long Tail Pair) yeah I'd go with that.

Re Q4, Fender Pro junior works ok 2 knobs Volume & tone,, or if you want more than the basic try the Hiwatt tone.
It's in my new circuit that I'm messing right now that might suit,, so when I finish the R&D bench test I'll post the schematic for you. Sorry no Rev or Trem though.

Phil.

deltafred · Post by **deltafred** » 03 Dec 2020, 16:27

I've had yet another Laney 15W amp across my bench and this time took a note of the transformers.

The OT was about half the size of the one I have (P4133) and the mains transformer was not quite as big as it.

I don't think that I need to worry about it not having enough iron!

jarrodthebobo · Post by **jarrodthebobo** » 03 Dec 2020, 20:44

Just happened to notice this thread so I thought I'd hop in!

If you have a single ended tranny that you're hoping to use for a fun little project and feel restricted since most guitar amps are PP, I'd say go for it anyways.

This thing here is a project that started its life really small; using a single el84 or 6v6 in the power section. The preamp was originally just the Bright channel of a superlead, but has since been played with quite a bit.

With 3 12ax7s in the preamp, you have quite a lot of possibilities in terms of tonal options, and without the need for a phase inverter that's 3 12ax7s, or 6 triodes that are free for pretty much any topology you can think of.

As of now, and as of that video posted above, the first 2 tubes are used in a JCM800 style topology; 3 gain stages and a cathode follower. The second gain stage (cold biased 10k stage) can have its cathode bias resistor changed to a typical marshall normal channel value (820ohms) and is rerouted to be in parallel with the initial bright stage (2.7k) for a superlead style topology. With additional switches for different cathode bias caps/resistors you can get almost any Marshall/modded marshall combo you could really want out of the preamp, since nearly all the famous marshall tube amps follow these 2 topologies.

With the first 2 tubes out of the way, the third tube is used as both a effect loops send/return, and a "fake phase inverter". The send is just a cathode follower in order to better drive effects and to avoid tone-sucking. The fake phase inverter is just a typical gain stage to both make-up gain lost due to the effects, and to add some additional distortion/breakup that tends to be missing in SE amps using a typical Marshall topology.

Finally for the power amp itself you're really only limited as to how much your transformers can handle in terms of current draw. Since this thing has been through so many design changes, I ended up throwing in some absolutely stupid-big transformers; including a 35w output transformer (still single ended). This means I have huge current headroom to do whatever I want tube-wise in the power amp.

Whether you want to use a single tube in the power amp or 2 in parallel is also up to you; but I always recommend using 2. Why? It sounds better due to the mojo of an added tube of course

. Also, if you use a cathode bias power amp, you can intentionally mismatch tubes (one 6v6 and one el34 for example) for some cool sounds. I was using 2 6550 tubes for the longest time in this build and pushing 30-ish watts; but holy hell was it loud! It's now rocking two (mismatched) el34s that breakup nice and well and are plenty loud.

Single ended amps are really fun for us tweakers since they're so simple in cathode bias arrangements. Go crazy! Try one el84, or one 6v6, or a mix of each... or some el34s... or 6l6s!

deltafred · Post by **deltafred** » 03 Dec 2020, 23:34

As it happens I do have a small SE output transformer but my original intention was PP so that is what I am going to try first. I might have a play with SE next though.

I've made a start on the breadboard, I found a chunk of plywood about 1 foot by 2 (300 x 600mm) and have built up the HT supply and made a start on the output stage but other jobs keep getting in the way of progress.

jarrodthebobo · Post by **jarrodthebobo** » 04 Dec 2020, 02:00

deltafred wrote: ↑03 Dec 2020, 23:34 As it happens I do have a small SE output transformer but my original intention was PP so that is what I am going to try first. I might have a play with SE next though.

I've made a start on the breadboard, I found a chunk of plywood about 1 foot by 2 (300 x 600mm) and have built up the HT supply and made a start on the output stage but other jobs keep getting in the way of progress.

Shoot my bad! I must have mis-read one of the previous posts in this thread; I had thought that you had found out the tranny you were using ended up being SE for some reason.

Ignore my whole little rant up there then

My only experience building a PP amp is the little "preamp" thing I built. Basically its a 2-3ish watt PP amp using a 6sn7 at the output and a small PP output tranny. Works pretty nice and sounds really good too.

Another self plug

Again same rules apply with the preamp as stated before however! You can match whatever you want up to any power amp that you want as well as evident by this little crap-box here

. You just gotta ensure you leave at least 1 extra tube for the PI (or one extra triode if using a cathodyne phase splitter). I'd recommend going with a typical "marshall" style PI however (long-tailed pair), unless the amp you're going for uses a different approach, like paraphase or something else odd. You can use a transformer to do the phase splitting too, but that's typically designated to the hifi world and not usually a common answer in the world of instrument amplifiers, but it IS possible.

Tube amps I've found are a lot easier to screw with than the small solid state device equivalents; mainly because it's very, very hard to blow up tube components unless you're REALLY doing something wrong. The only downside is the huge voltages involved that can definitely kill you if you're not paying attention!

Get yourself a small turret board that you can populate with stand-offs and the like; it'll make designing and building sections of the amp much easier.

I'd suggest designing a power supply first and making a board for it; including all the separate B+ nodes and the like. You can likely keep the first reservoir cap separate from this board if you'd like to use a big can cap or the like. Once you have the power supply all good to go, and have verified all your voltages are correct and power isn't being drawn more than it should, the rest of the build should be rather easy.

Build a power amp board, leaving all the tubes utilized (ie 12ax7 PI and el84s) off the board. The sockets can either have components wired directly to them, or wired off to the little board you made. A lot of people tend to keep the resistors as close to the tube pins as possible though so, ymmv. Hook up a jack to the input of the circuit, or inject a signal into it, and give her a test through a small crappy speaker to ensure that signal is passing (after testing for shorts and the like first of course!) . If you'd made it this far with no issues, building the preamp should be quite easy. Again, keep the tube sockets off the board, create a new board for the preamp components (or keep everything soldered directly to the pins 'floating' if you wish; not advisable since its messy but, it's what I do 90% of the time since I'm a lazy asshole

) and connect it up to the PA.

There are tons of videos on youtube of people building amp chasis from ground-up from kits/scratch and it'll give you a real good handle on the most efficient ways to get everything done.

deltafred · Post by **deltafred** » 04 Dec 2020, 09:30

jarrodthebobo wrote: ↑04 Dec 2020, 02:00 Shoot my bad! I must have mis-read one of the previous posts in this thread; I had thought that you had found out the tranny you were using ended up being SE for some reason.

Ignore my whole little rant up there then

It's all good and glad you stopped by. I regularly misread stuff so know how easy it is to get on the wrong track.

I have thought about building a SE amp for years because I knew I had the transformer but when I discovered I had a PP transformer a couple of weeks ago that kicked me into action and nothing like starting a thread about it to keep me from getting sidetracked.

phatt · Post by **phatt** » 04 Dec 2020, 13:57

Hey Fred,, I'm still bench testing a circuit using my ideas mixed with a few others and so far it's working well.
Preamp is slightly modified Clean/Crunch channel of the SLO, Into my Modified HiWatt tone stack then into a PP 6V6.
I Don't need or want the Lead as it's too much for my styles so only 2 Ax7's for preamp. yay,, simple

This page will give clues as to what you can do with this circuit.
https://robrobinette.com/How_the_Saldan ... _Works.htm

Even a circuit for an AU7 1 watt power section,, With some tweaks it could work with EL84.

Here is what is on my bench today it might change but will give you an idea of what can be done. This is a much sweeter/smooth sound than a lot of other valve circuit ideas I've tested in the last few years,, or 30

Phil.

deltafred · Post by **deltafred** » 04 Dec 2020, 17:30

Thanks Phil. I think I've collected most of the common EL84 and other low power OP tube schematics so I can now sit down and compare their similarities and differences then get busy with the soldering iron.

Manfred · Post by **Manfred** » 06 Dec 2020, 18:10

Here is a proposal for a Class A push pull amplifier with EL84 or 6V6 tubes. To get the low power of 9W for the P4133 OT one tube must see about 10kOhm as Ra,
because in class A operation the tubes work in parallel at the OT, I have chosen 12kOhm.
This value can only be reached with a 16 Ohm loudspeaker as load at the 3 Ohm secondary winding by the given winding ratios, with all other combinations this is not possible.
With this current load, about 360VDC can be achieved with the 260V-0-260V mains transformer with full-wave rectification.
Unfortunately the data sheet of the P4133 does not specify the maximum voltage for which it is rated.

: 6v6_PP_ClassA.bmp (272.9 KiB) Viewed 3648 times

deltafred · Post by **deltafred** » 08 Dec 2020, 11:24

Thanks Manfred. I will try this out.

I have a choice of 16 ohm 12 inch speakers I can use, a Celestion G12 or a Hiwatt out of an old PA cab.

I've been reading up on transformer turns ratio and impedance, something that I hadn't really thought about before, it's all starting to make sense now.

deltafred · Post by **deltafred** » 11 Dec 2020, 10:54

Number 5 is alive!

PI and output stage, as suggested by Manfred, on the breadboard and after a minor delay because I'd picked a faulty ECC81 out of my spares box (only half of it working) did a few measurements.

Quiescent current 45mA @ transformer CT (no input signal).

The following measurements taken with a 15 ohm dummy load on the 4 ohm tap with the output stage at the onset of clipping.

Output transform CT - 345V (I used the 0 - 250V MT)

PI supply - 323V

25V peak to peak into dummy load at onset of clipping which - about 5W (hence the no. 5 quote).

More or less flat FR from about 50Hz to 20kHz (subject to the accuracy of the scale on my sig gen and estimating 3db down on a scope).

The MT was just warm to the touch after about 30 minutes at full power (5W), the OT still felt cold.
Next job is a build a preamp and get a speaker out of the loft to see what it sounds like.

Manfred · Post by **Manfred** » 11 Dec 2020, 18:28

I wrote:

...because in class A operation the tubes work in parallel at the OT...

This is wrong the output tubes work serially.
Since these currents are inverted to each other and the windings are opposite with respect to the tap, the two generated magnetic fluxes are in the same direction and add up.
Also I followed with the Ra of the winding an example, which I copied years ago from an Internet page.
In the meantime I have read in my record and books, and found out that the contribution of this author is incorrect.
He had A and B operation thrown together.
Ok, the circuit works, but the output power should be too small in my opinion.
I will redo the dimensioning with the new facts.
What type of tube did you use.