ECL82 power supply question

[chrish]

G'Day,

Next project is a point-to-point ECL82 push-pull. Previous projects were a Tubelab SimpleSE and a point-to-point 6P15P SE. This project is intended to be a learning process, I will be using it to power some Fostex FE127E speakers in 4.5 litre cabinets in my office with the source being my Mac. I want to learn techniques for building an amp and gain some experience before embarking on a more complex EL84 PP "Baby Huey" designed by GingerTube for my main system. First ECL82 effort was unsuccessful, a little too ambitious as a first point-to-point build. Subsequent 6P15P I paid particular attention to grounding and managed to get a hum-free result straight up.

Now my question. I have chosen this project http://www.bonavolta.ch/hobby/en/audio/ecl82_3.htm There are a couple of issues. First, I managed to pick up a cheap power transformer while in Singapore a few weeks ago. Trouble is, it only has 2A + 2A 6.3V. That is sufficient to run the 4 ECL82s, but not enough current to run the EZ81 rectifier. It has a 3A 5.0V tap, so I am going to run a GZ34 or 5U4G rectifier. From what I can gather, the specified 100uF first filter cap will be too large. I have a 50+50uF JJ multi section cap that I can use. Can I use the 50+50 in place of the two 100uF caps, or will this be insufficient filtering? Next point is that as this is a "learner project" and funds and real estate in the chassis are limited, I have decided to omit the 8-10H choke. Can I simply substitute a resistor of about 150R in its place? Next question, in the power supply schematic, there appears to be three 470K 3W bleeder resistors. Are all of these necessary? I have the book that this schematic comes from, and there is no mention of design goals with the power supply.

Picture of the cheap power transformer attached...

Any help or advice is greatly appreciated!

Thanks,

Chris

[Eli Duttman]

Chris,

Look at the 6CA4/EZ81 data sheet. 50 muF. is the limit for the I/P cap. So, you know the designer of that circuit doesn't respect published limits.

Using a 5AR4/GZ34 is a good plan. A single section of your 2X 50 muF. cap. in the 1st position is within published limits. Parallel the 2nd 50 muF. section with a 100 muF. part. The total capacitance is the same, but it's distributed in manner that respects published limits.

IMO, the 1 MegOhm value shown for R1 (the I/P grid leak part) is a selection of dubious merit. High freq. roll off could easily occur. 100 KOhms is, in my thinking, a better choice.

FWIW, I have reservations about the phase splitter circuitry too. For very good reasons, LTPs and "concertina" splitters are "in vogue". That circuit uses neither of the preferred topologies. It appears to be paraphase.

What is going to drive your 6BM8/ECL82 project?

[chrish]

Thanks for taking the time to respond Eli,

This project is just a "learner". Using output transformers (cheap 10 watt Edcor PP open frame) salvaged from my first ECL82 project http://www.bonavolta.ch/hobby/en/audio/ecl82_2.htm This project did not work, but I do not consider it a failure. It hummed and buzzed like crazy, but did amplify the signal... I tried to get too much into too small a chassis for my first point to point project. Learned a lot about grounding on the way, my next effort was a lot better due to the lessons learned.

That original ECL82 project (Mullard) looked a little complex, one extra tube than the Rainer zer Linde design I listed in the first post. I have no attachment to the design and am happy to abandon it in favour of a better option. My aim is to keep things simple, trying to keep it to just the four ECL82 tubes if possible. A question about the Mullard design ( http://www.bonavolta.ch/hobby/en/audio/ecl82_2.htm ), the first tube introduces a lot of gain, was this originally designed as a phono pre-amp stage? If so, maybe I could modify this design slightly by removing the first stage.

To answer your question Eli, this amp will be sourced from the headphone jack of my macbook pro, playing iTunes. Absolute hi-fi not required, using this project to gain building skill before attempting a more complex EL84 project with constant current source etc.

Have seen your "el cheapo" design and was thinking maybe of adapting it, but was hoping for something a little simpler to concentrate on issues such as layout and alowing simpler fault finding if required.

If it is some help recommending options, here are the parts I have available:

Power transformer: 280-230-0-230-280v 150mA, 5V 3A, 6.3V 2A, 6.3V 2A, 12.6V 1.5A

Output transformers: Edcor XPP 10-8-8K (10 Watt, 8K:8 ohm Push Pull)

Rectifiers: GZ34, 5U3C (Svetlana 5U4G), EZ81

Caps: JJ 50uF+50uF, JJ 100uF+100uF, 80uF PIO Motor Run, several ASC 20uF+4uF motor run caps

plenty of Russian 0.22uF PIO 500 V coupling caps
some solen 0.47Uf coupling caps

Thanks for the help,

Chris

[chrish]

Have done a little more searching. Does this look like a better schematic? http://www.geocities.com/asl_wave8/schem/wave8stock.pdf

Thanks again,

Chris

[planet10]

My 1st thot would be to just use the 4 6BM8s to do a simplified El Cheapo... a primary simplification could be the replacement of the CCS for just a big resistor to a negative supply . Something like what Poinz did on the front of this...

http://www.audiotropic.net/Projects/machine1.html



Eli might have some ideas too.

dave

[chrish]

Thanks for the response Dave.

Still very much a beginner with all of this, hence the idea of a simple build. For the negative supply, can I simply tap off one of the unused high voltage secondary windings? For example, with my 280-230-0-230-280 transformer, if I use the two 280 volt taps for the B+, can I use one of the 230 volt taps through a diode to provide the negative supply?

And for info, the amp will be powering these speakers. Look familiar?

Cheers,

Chris

[planet10]

We'll have to lean on Eli for your 1st question -- as soon as you said 4.5 l FE127 i knew exactly...

dave

[Shoog]

Be careful of overloading the input triode. Most designs will only bias it up at less than -1V. Also The gain on the triode is about 60x with a perfect load - what are you going to do with all that gain ? Think of a voltage divider on the input such that the grid will never see more than 0.5V. Even better a step down transformer will sound a lot better.
The ECl82 is a tricky candidate to deal with.

Shoog

[Eli Duttman]

Originally posted by chrish
Have done a little more searching. Does this look like a better schematic? http://www.geocities.com/asl_wave8/schem/wave8stock.pdf

Thanks again,

Chris

Is the Pope Catholic? Do bears sh*t in the woods? You bet that design is better. The "Wave" small signal circuitry is common cathode gain DC coupled to a "concertina" phase splitter. Excellent!

Copy the 56 nF. cap. at the I/P from "El Cheapo". You protect the O/P trafo cores against saturation.

[chrish]

Thanks again Guys for your time and your patience.

I suspect that perhaps I have fallen for the trap of asking "I have these great wheels, what car should I build" type situation. Yes I have the 6BM8/ECL82 tubes, but I also have some extra:

ECC81/12AT7
6N2P (similar to 12AX7/ECC83)
6N1P (similar to ECC88)

6P15P (similar to EL83)
6P3C (Similar to 6L6GT)

Have I been limiting myself by only considering an ECL82 build? Limitations are space in the chassis, build will be in the same chassis as for this 6P15P SE build New SE 6P15P project small output transformers (10 watt 8K) and the power transformer as listed above.

Eli, I assume that you mean to put a 56 nF capacitor after the RCA input, before the 100k input grid leak resistor (this will be a straight power amp, no volume control)? I am also guessing that as this forms a HP filter at about 28 Hz, it is the low frequencies that you are trying to keep out that will saturate the core of the small output transformers?

Thanks for your patience,

Chris