Edison 12 retroproject

[soundbrigade]

Maybe I've told you already about Erik Andersson, the guy behind Audio Innovation, Edison and other fine tube stuff.
I thing the Edison series of tube amps were his first and especially the Edison 12 became very popular as it was a cheap, simple amp based on 6922 and EL84 (PP) and other easy-to-find components.
As the design is well proven and a new group project is on the way, I came up with the idea of maybe getting DIYTUBE involved somehow.
Please have a look at this and have an opinion on the matter. I will also try to get in touch with Mr Ahlberg (the project maager) and Mr Andersson about their views. Already asked if they could produce some more PCBs than "needed" by the forum in order to sell to "outsiders".


The Edison 12 (Original):


The Edison 12 thread
Erik Andersson's website

[kheper]

Not a bad looking el-84 amp design. If possible, I'd substitute a choke for the first resistor in the power supply, and use a UL output transformer.

[EWBrown]

I don't read Swedish, so I'll have to "wing it" here, since he doesn't show the operating power supply voltages.

With the 150 cathode resistors, as shown on the EL84s, I would SWAG that the "1" B+ voltage is approximately 250VDC, and idle current approx. 45 mA, for each EL84.

With the "usual" 350 to 380V B+, as used in the ST35, the EL84's rated plate dissipation would be seriously exceeded.

/edB

[soundbrigade]

I guess, as this design is maybe 20+ years old, it will be updated. There's a PCB to come and maybe even for the preamp, which is actually an active RIAA + a passive line amp.

Keep you posted.

[soundbrigade]

The preamp:

[Geek]

Please have a look at this and have an opinion on the matter.

- EL84 cathode resistors should be at least 5W. They are dissipating 1.3W for a typical -14V bias and need to be derated for above 25C. operation by any data sheet I've seen.

- Add grid snubbers to the EL84. 10K to 47K non-inductive (no metal films) will do. Default higher since pentode-mode has very low C-Miller to deal with.

- That 1M5 || 22n made me think for a moment it was a guitar amp. Please explain why it's there? If it's to fix a gimped up top-end frequency response, use a better OPT.

The preamp:

Woah, wait.... a MC position using an ECC83 in grounded cathode? You're kidding, right?

Cheers!

[mr. triatic]

Tjena Magnus!

I have rebuilt three Edison 12, i simply triode strap the EL84's, skips the first ECC88 stage and remove the feedback.
I also lower the B+ just a tad, due to the fact that the amp is constructed when we had 220V from the wall, today we have 230+ Volts and also because i remove the load the first ECC88 was.

Also, consider rebuilding the original Power supply, not even Erik himself was pleased with it, but after all, this was a budget oriented amplifier that had to cut some corners, in any way, it is not a bad amplifier but with some small modifications it shines!

I'm very happy with the result and the guys that owns the two other modified amps are very happy with the sound.

[soundbrigade]

Tjena sjelv grabben!

Good to hear. Got news from HiFi-forum. There will be PCBs offered for enthusiasts outside the forum. This is a non-commersial project, by the way.

[Kranis]

Woah, wait.... a MC position using an ECC83 in grounded cathode? You're kidding, right?

It's was a special solution for the medium-to-high output MC Shinon Red, which was popular in Sweden at that time. I believe it is noted on the schematic. Barely legible though.

/Kranis

[Kranis]

Please have a look at this and have an opinion on the matter. I will also try to get in touch with Mr Ahlberg (the project maager) and Mr Andersson about their views. Already asked if they could produce some more PCBs than "needed" by the forum in order to sell to "outsiders"

Not much really, if you like the original sound.

1. Heavier decoupling of the phase inverter, since it has poor PSRR on its anode. An alternative would be to experiment with Broskie's (TubeCad) PSRR enhancing tricks.

2. Contrary to popular belief, the split-load inverter remains perfectly balanced with excellent bandwidth as long as the loads are balanced. If the grids of the EL84s are driven positive that is no longer the case. A solution would be to add an extra half a tube to each channel to make the classical Williamson front end. It would have the same gain and do a away with cathode decoupling on the driver stage.

3. Add some grid stoppers to the output tubes as an extra precaution

4. Ultra-linear OPT could be an option. However, I don't really see the point here as the EL84s are true pentodes and happy to have their screens close to their plates.

5. Make it an "Edison 24" by using a quad of EL84s on each channel. With an class AB1 output that would be about 35 watts or so.

As for the 1,5 MOhm//22nF in the schematic, I don't think it was used in production. It's marked on the board but there is a wire jumper there in my amp. My guess is that it's a way to control the phase at low frequencies to avoid motorboating. Apparently the OPT does not need it.



Cheers,

Kranis