RH ECL86 Amp: Dave?

[pjanda1]

I've decided to hang onto the console amp I had posted for sale, and instead begin my first tube amp project. (Telefunken, se ECL86's and a while bunch of other stuff.) I don't know anything (I've read a bit, but that doesn't count for much). I'm looking for something simple and cheap that sounds as good as possible. I don't need much power. It will mainly be a learning exercise, my Hemp BIB's are plenty efficient, and I've got 50W of chip amp when I need it. I'd like to build it with as many salvaged parts as possible (mostly from the Telefunken console). I've been looking over ECL86 circuits. I know the RH84 has been popular, and it looks simple. Planet 10/Dave posted this schematic awhile ago (a couple of years), which evidentally is a stab at an RH ECL86. Is this useable? Has anyone built it?

http://www.diyaudio.com/forums/attac...amp=1113328953

pj

[planet10]

The idea is certainly sound. You'd need to get the bias right (i can't rember where i pulled the values for that map, but i likely took an existing ECL86 SE and just changed the feedback to RH style.

It should likely have a bit lower B+ than a regular RH84, and a bit less power. I'd also work to get a bit more robust supply for the screens.

Have a look for Gingertube's push-pull variant -- his 1st one used ECL86.

dave

[mach1]

The feedback ratio in the ECL86 version has been radically increased when compared with Alex's RH84.

Feedback is determined by the ratio of the feedback resistor (100k) to the paralled combination of V1 plate resistor + (internal plate resistance of V1+ {u +1}Rk) + grid resistor of V2.

Using rough mental calcs, Alex's design ballparks at 10%, with the ECL86 version approx 35%. Raising the value of the feedback resistor to around 470k would restore the balance.

[mach1]

ECL86 as per schematic with 100k feedback resistor (assuming triode Ri of 62.5k and mu of 70 with Rp as per schematic)

feedback = 1 - (100k/ [100k + 1/{(1/100k + 1/(62.5k + 71k) + 1/470k)}]

= 1 - (100/150.82) = 34%

ECL86 with 470k feedback resistor

1- (470k/[470k + 1/{(1/100k + 1/(62.5k + 71k) + 1/470k)}]

= 1 - (470/520.8) = 9.8%

[planet10]

I changed the feedback, added a screen supply bypass, and hinted at regulating the screen supply.

More suggestions? B+? 250-275V?

dave

[pjanda1]

Thanks for you input thus far. Circuit design considerations are way over my head. I, of course, like the look of the simple screen supply. Are there big gains to be had? Dave, you mentioned bias early on. I've looked around at the various published circuits and found a bit of variance in the size of the cathode resistor, but you're 180ohms seems to be in the middle. There are a few at 170 and one or two at 200. I don't think I need much power, and I have a little stash of used and I believe NOS ECL86's, if either of those plays into the consideration of where to bias it.

I assume that my power tranny will produce an adaquate voltage. I should pull it out and see. Given the complexity of this receiver, it has plenty of little taps too. In this application, is the choke worth the $$ over big electrolytics? If you believe so, I'm certainly up for it.

Finally, I have some parts quesitons as I plan this thing. I assume that if I test enough of them I'll find some good carbon resistors. Is there any chance I could reuse a pair of the axial caps for the coupling caps?

pj

[planet10]

Originally posted by pjanda1
Is there any chance I could reuse a pair of the axial caps for the coupling caps?

I wouldn't even consider reusing the caps. Modern caps tend to sound way better IMHO.

dave

[pjanda1]

I stumbled upon a free Ward's Airline SE EL84 console amp this weekend. The transformers all appear to be Hammonds, and the outputs are slightly larger than the Telefunkens. They are also probably have 8 ohm taps, whereas the Telefunkens had 5.5 ohm speakers. Fate is forcing me to build this amp, or something along these lines.

pj

[becki200]

Hi try this.. very good amp

[fyzygy]

has anybody built one of these RH ECL86 amps?