diyAudio (
-   Tubes / Valves (
-   -   Design using EL42 output pentodes (

phil_2 25th August 2008 02:21 PM

Design using EL42 output pentodes
Hi, this is my first post here but I've been reading this forum for some time and I'm learning by the day.

I have a pair of mono valve amplifier chassis from old pye black box record players (bought cheaply off ebay), they each use a single ECC83 for the input stage, two EL42's for the output stage and one EZ41 rectifier.

Unfortunately they aren't a matching pair, one is the early mk 1 circuit the second being the mk 2 which differs in that it uses 'an ultra linear push-pull output stage'. so the output transformers are different, one has three wires for an input the other has five.

I have tested both amplifiers and they both work, they have an output of about 2 watts each I would estimate.

now I honestly don't know much about valve amplifiers and I would like to know if these amps would be a good basis to build a stereo amplifier or if these tubes are too lo-fi for any practical use? does anyone have a schematic for an amplifier using these valves? I have the schematic for the black box but I assume the original design, being designed for a crystal stylus and not a CD player would need serious modification which i'm afraid I don't know how to do.

any advice/thoughts or schematics using these valves would be greatly appreciated :)


richwalters 25th August 2008 06:46 PM

What makes you wanna to go for EL42's ? Awkward 8 pin sockets. 2W is a crumb low. A push pull pair should get 7W with 250V B+ on anodes other than this, in my 1950's audiobook (with cover missing) mentions "quality audio amp".
I strongly suspect one model has the o/p stage with a common cathode resistor (unbypassed) with reasonably matched tubes. Or using o/p transformers with differing inductances.
I suspect the anode to anode impedance is around 15 K....bit high implying a ratio of 70:1 for a 3-4 Ohm LS

As Franks data sheet mentions....this tube may have seen it's debut in early day car radios with vibrators.

I will dig the circuit from my archives.


Eli Duttman 25th August 2008 07:36 PM


I have the schematic for the black box but I assume the original design, being designed for a crystal stylus and not a CD player would need serious modification which i'm afraid I don't know how to do.
Phil, please post a scan of the schematic. Little change is required to switch from crystal phono cart. as the source to CDP as the source.


Unfortunately they aren't a matching pair, one is the early mk 1 circuit the second being the mk 2 which differs in that it uses 'an ultra linear push-pull output stage'. so the output transformers are different, one has three wires for an input the other has five.
Ultralinear (UL) topology involves connecting the screen grids of pentodes to taps on the O/P trafo primary winding, instead of directly to a high positive voltage. The British description of partial triode is very apt. UL mode is, in fact, a form of local NFB.

Obviously, assuming the 2 O/P trafos are otherwise equivalent, you will have to use pentode or triode mode in both channels and tie the UL tap wires off.


As Rich has indicated, things start with the EL42 data sheet. The rim lock socket is a PITA. Fortunately, as shown here, the type is electrically equivalent to the Noval based EL85. The EL85 data sheet contains a good deal of useful info. that confirms Mr. Walter's excellent intuition.

mach1 26th August 2008 07:04 AM


The classic telefunken V73 uses EL42 tubes. This push pull amp has a cult following in Europe and commands very high prices. Even some US tubeophiles appear to think highly of this amp.

In their day, Pye tended to dominate the opposite end of the budget spectrum, so I doubt whether the EL42s would be the limiting factor in their amp's performance.

mach1 26th August 2008 07:11 AM

V73 schematic

phil_2 26th August 2008 11:38 AM

Thanks for the help guys, much appreciated.

The reason I'm trying to use these valves and transformers is simply because I have them and I'm hoping to build an amplifier on a low budget. 7w per channel is also more than I hoped with these little tubes :-)

as for the tube sockets I'm hoping to reuse the sockets I already have if I can.

I will try and post the schematic I have on here tonight, unfortunately I have to head off to work now..

Thanks again for your responses guys.


phil_2 26th August 2008 09:10 PM

Hi Again, I've opened a photobucket account so that I can post clear images of the tatty service sheet I have, hopefully these images will work...

The schematic is for the Mk2 version, on page two there is a discription describing the differences between this and the earlier model.

Eli Duttman thanks for letting me know that the modifications are possible. It would be fantastic if you could modify the schematic for me so that it will be possible to connect a CD player directly to it and also, if possible remove the tone control as I would prefer not to have this in the circuit. I would be very greatful, please have a look and see what you think.

I am thinking that it will be easier to downgrade the later model than upgrade the earlier model as I think I will be able to find a 'three terminal input' output transormer more easily than another 5 terminal one, is this a sensible idea? or is loosing the ultra-linear feature a bad idea? obviously I need a matching pair for a stereo amp.

also thanks mach1 for the link to the V73, I had no idea there was so much potential in the EL42's!

thanks again, Phil.

Eli Duttman 27th August 2008 02:00 AM


You are definitely on the right track by wanting to eliminate the tone controls, as they are very nasty.

As is so often the case with "ancient" units, considerable simplification is a must. Earlier, you mentioned that you were not looking for the peak power possible from PP EL42s. Triode wiring the "finals" in both amps allows you to use the O/P trafos already in your possesion. Just tie the UL taps off on the part that has 'em. ;) Figure on an approx. 50% reduction in peak power O/P, but all sorts of pitfalls disappear, when that price is paid. To triode wire, connect the screen grids to the plates via 100 Ohm resistors.

Are you willing to go back and forth between the 2 units when setting listening level and channel to channel balance? If so, a simple log. taper pot. at the I/P of each amp is all that's necessary.

For more than 1 reason, the small signal circuitry needs to be revised. Once the tone control stuff gets tossed, you have too much gain. The bandwidth a CDP provides is considerably broader than the BW a crystal phono cart. delivers. The posted schematic shows a common cathode gain block cap. coupled to a split load phase splitter. The splitter is cap. coupled to the PP "finals". That many coupling caps. in an amp with global NFB and low quality O/P trafos is a prescription for instability. :( Aside from excessive gain, an ECC83/12AX7 section has a high RP and low gm that make it a poor splitter, especially for driving the substantial Miller capacitance of triode wired O/P tubes. Fisher, Dyna, et al, DC coupled the splitter to the voltage amplifier, which makes a world of difference in the stability dept. Perhaps some other twin triode could be used with the existing B+ PSU in a DC coupled arrangement.

Do you happen to know how much current the pilot lamp draws? If it's 150 mA. or more, that current could be used to increase the possibilities for the driver/splitter position.

phil_2 27th August 2008 12:43 PM

Thanks Eli, I think I follow you on the tone control removal, would it be ok to replace pot R18 for a resistor of the same value, removing the wire that connected the 'wiper' of R18 to C2 and removing C2, is this all thats required?

regarding the volume control am I right in thinking I can connect a log pot (say 50k?) directly to the grid of V1a and earth and I can do away with R1, R2, R3, C1 and R4? Sorry if this is a dumb question, also, do i need a decoupling cap? if so where should it go?

now you said "To triode wire, connect the screen grids to the plates via 100 Ohm resistors" I'm afraid I'm not used to this terminology, by plate do you mean the Annode?

now I must admit you lost me a little with the small signal circitry revision, I would prefer to use the tubes I have, they are all original Mullard and also I would like to keep the cost of this project to a minimum, but if thats unwise then I'm happy to change them for something better.

I haven't seen the lamps used in the black boxes, but I would assume its just a small flashlight style bulb, about 150mA seems a fair estimate.

thanks for your time, I hope you don't mind all these questions.


Eli Duttman 28th August 2008 01:55 AM


Yes, plate = anode. Please use the 100 Ohm triode mode "straps".

Mullard made ECC83s from the 1950s are very good tubes. Unfortunately, they are electrically wrong for the task at hand. Store the ECC83s very carefully. You will use 'em in a future phono preamp project. :D

"Stealing" the current used by the on/off indicator lamp allows the use of a 12AV7 family type. Other members of the family are the 5965, the 7062/E180CC, and the CV10175. Gain is still substantial, but low RP and high gm make for a much better driver/splitter.

I'm hoping to use DC coupling between small signal stages. Eliminating the coupling cap. there is important for stability's sake.

Please use "Google" to research Miller capacitance. That phenomenon is yet another issue with the 12AX7/ECC83.

High gm triodes are vulnerable to parasitic oscillation. You deal with the issue by mounting 1 KOhm Carbon composition resistors right at the grid socket lugs. The "free" end of a stopper resistor is where the grid connection is made.

A 50 KOhm log. taper volume control is OK. Couple the wiper of the volume control via a 12 nF. cap. to the grid stopper. Connect the grid stopper to ground via a 510 KOhm resistor. A high pass filter that "corners" at approx. 26 Hz. is formed. That filter provides important protection against core saturation in the O/P trafo. All the OEM parts in the I/P grid circuit get discarded.

R18 and C2 get discarded. R17 becomes a 1% tolerance 1 KOhm metal film part and connects to the junction of R5 and R6, not the cathode of V1a. R6 becomes a 1% tolerance 100 Ohm metal film part. The new value of R5 is to be determined (TBD). That cathode bias part has to be bypassed by a large electrolytic capacitor.

C3 is replaced by a KOhm Carbon comp. grid stopper. R8 and R10 are discarded. The junction of R9 and C6 is at the cathode of V1b. R9 and R11 are of an identical value, which is TBD. I'm hoping that SY will speak up in this thread, as he can work DC coupled "concertina" phase splitters out, "on autopilot".

The thought of 50 Hz. heater power flowing in the chassis makes me VIOLENTLY ILL. Tightly twisted pair wiring must be used for AC heater power. Lift that 6.3 VAC connection to ground!

All times are GMT. The time now is 02:55 AM.

Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 18.75%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Copyright ©1999-2018 diyAudio