PSE EL84 Amplifier

[Paul Barker]

Between 3 and 4 Watts estimate

Speaker output or LCL driver for GM70 amp for which loudspeakers not connected.

Grid capacitor (2uF pio) before rca out interconnect rca in grid choke on ground of rca @ GM70 chassis.

[Paul Barker]

The sheet of steel for dressing the top plate from a fresh start.

[Paul Barker]

[Paul Barker]

I Trod this path before

open gas stations in my area

[Paul Barker]

find diesel near me

[Paul Barker]

[Paul Barker]

Correcting errors in diagram, such as no grid stopper drawn, but present.



Sadly there is a lack of info on the grid lines right of the typical quiescent operating point 250v -9v.

So my plan is to buy an up to date UTracer and trace the unknown area so that information is clear and complete.

In the days of these valves nodody imagined some idiot would build this design.

[Paul Barker]

Now I have to figure out how to guillotine 1.7 mm stainless steel without access to said equipment. I know if I cut it with a thin blade in my angle grinder it will skate about and ruin the job. Might take a walk up Barry’s Lane and see if the metal worker is still in his unit.

Years ago I was in a model engineering club equiped with all such machinery. The members died years ago. Its a new world. No club survived. Soon noone will give a toss about chokes used in this way. Just throw them in my grave then.

[Paul Barker]

It has to be said some Mullard valves not unrelated to EL84 are available at Langrex for £14.

In this amplifier they require less bias but should be worth trying same power out, but a Mullard. My EL84’s if British are not nos but well worn. I do have Russian equivalents which Ive always liked. But this opportunits to get nos Mullard valves.



Be aware g3 and kathode are ic and g2 is where g3 of the EL84 is found.

I note they have 1,229 in stock.

I found a set of four Mullard on ebay for £24 so I’ll alter the bias and evaluate them. They look new but are reported as used test good. No signs of wear in the picture.

[Mike H]

Just IMO, I don't like the idea of 2 x EL84 in parallel, seems like a step too complicated. Also getting them biased the same with fixed bias couild be a headache. So then what, separate bias circuits with trimmers?

I assume the issue is must drive a 2.5k primary?

Why can't the secondary load be altered to make it 5k primary? Just wondering aloud.

As said just random musings

[Paul Barker]

Its more about getting 4 watts instead of 2 watts. The fixed bias is to avoid requirement for autobias bypass cap. Thanks for the warning.I could check the bias with the clamp meter though so it will show if bias balance is an issue.

The 4 watts would be fine at home, but 2 watts not.

I wound the opts with a great deal of inductance even though they are 2k. The 2k was designed for the 6080. Using the 6080 was to get good bass. The transformer design made sense where my head was at the time last century.

[simon]

NOS Mullards for £14 has its attractions. Be interested what you think of them Paul.

[Mike H]

+1.

OK next silly question, why 4 Watts

PS: 2 matched EL84 may bias together, but unlikely to stay matched with use (my personal experience).

[Paul Barker]

Why 4 watts? because on the speakers 2 watts would be a waste of project.Wad did a pse 2a3, I did a pse 801a. I didn’t find it a problem, but I think you are saying fixed bias makes valve balancing is then likely to drift.

I could just build the driver amp only and go back to IT coupling focus on the GM70 project before I go too far with a dead duck. Just revert to this as a driver.

Id be still in the state of wandering what these 2k transformers sound like. But I could always use them as a choke. and parafeed a matching transformer. The only element that makes them 2k primary is the speaker impedance. The inductance says otherwise.

Funily enough Im awake and suddenly had a brainwave about phase splitting. I loved the IT phase split amp I built. Had the se elements then the lower distortion of the pp output though in class A was no more power than pse, less distortion.

I cant found those IT’s for love n’or money. Im not really in a condition to get my transformer winding together again.

You can parafeed phase split with a choke. My brain wave was. Only connect the primary of a pp opt as a centre tapped choke. My best pp opts are the Sowters from Dr John. So they must stay as the opt’s.

I have the lesser opts from Steves balanced pp amp. Those opts can be the centre tapped choke.

So for now I’ll scrap this shift in project to make it a amplifier which can also drive the gm70, back to the gm70 driver. I can still use the power supply and with half the signal current less difficult and more effective the shunt regulator. I had exceeded the boundaries of 190 mA amount I thought I could get away with, but it will be better not beyond design parameters. One less compromise.

What do you think of the opt as a phase splitter?

What smplifier that project becomes may be el34 or a dht by the way. Its not forced down the el84 path. Which was only chosen by Nick as I didnt then need to change the volt amp.

[Paul Barker]

PS I still listening to a real nice el34 amp and its 6 watts is all required to be honest. Quad 57’s need that as a bare minimum Im saying. Though the greater power of the 6550 shone out, the emotion wasnt thrre, its a cold sound. I never liked any versions of the wad kt88 I heard and the last iteration was worse still. My fav was the el84.

So anyway thats corrected by the el34 enough. Its not a heavy metal enforcer which the 6550 was, but Eva Adel Norah Jones blues and crooners (except sny Phil listens to) probably shouldnt have said crooners. More Willy Nelson Nat King Cole. get a playing. More my scene snyway so I want for nothing as I am. Shame I dont have mullard el34’s but im actually auditioning chinese and they sound as good as russian. Maybe Im warming to chinese option.