DIY Audio Projects Forum

Trying to provide some answers:

Glo: Thank you for the compliments!

Beardy: I don't think anyone knows what you are saying. At least I sure don't. Try again?

Mullet: Without a bleeding resistor, you WILL have high voltage for a considerable amount of time after power-off as the large filter capacitors hold the charge. Drain them before you work on the amp by putting a decent (~1K, 2K) 10-watt resistor between chassis ground and B+ for a bit to drain off the voltage. I use a 2.7K, 10-watt resistor myself. Also, the buzz you are hearing is probably the power transformer in operation. Mine "buzzes" as well. Nothing to worry about. I'm told that Edcor power transformers are a bit noisy and I can confirm the same on mine.

Hope this helps.

OK I checked my value : I use a 150 K resistor, it drops the B+ voltages to less than 10V in 2 minutes (voltage continues to drop with the usual asymptotic RC curve).

Thanks for inviting me back...
I am having issues with noise.
I have two boards; a "PSU board" which is pretty much as drawn on the original schematic, and a "head board" with tube sockets. And the transformers which are mounted on the chassis. The only deviation from the two schematics on Bruce's original design is that the last two caps across the filament supply are up on the headboard as opposed to down on the PSU board.

My guess is that my errors are likely in the grounding or interconnection between these sub assemblies. So I am looking for guidance on where I should run the returns and where I should isolate.

There are three wires coming up off the PSU board: B+ left, B+ right and the Filament heater 6V supply. The first two go to the transformers and the supply to the heater filament goes up to the "head board". There are also the two caps which are located physically close to the transformers and one end of the cap goes to ground.

I have been led to believe that whenever there is a supply or signal wire, then the return or 'signal reference' should go back along the same path, and best as part of a twisted pair. So, presumably there needs to be a return line from the cap that hangs off each transformer that goes back to the ground connection on the PSU board (where the B+ comes from) and this should be part of a twisted pair with the B+ supply.

There are several `signal reference' connections on the head board:
1. One end of 100k ohm resistor that runs from input signal
2. Pin A of voltage regulator/pin 9 (internal shield) of tube
3. One end of 47k ohm resistor that hangs of pin 2 of the tube
All of these come together on the headboard and I am guessing they go back to PSU board as part of a twisted pair and as the 'return' for the filament supply?

The signal common on the input jack needs to go up to the head board along with the signal input. It is the origin of the signal reference from the source.

I am reasonably confident about the above, but please confirm or point out the errors of my ways...

However, what about the common at the output?
The two output channels come from the two transformers. Where should the signal reference come from for the output jack?

Thanks

Beardy,

I'm not an expert on hum, but here is my feedback:

I'm not sure you need to twist filament heating wires if your filament heating is done with DC. They will not emit 50 (or 60) Hz. Still you may want to twist wires carrying audio signal, e.g. from input to pot and from pot to amp board so they do not catch whatever parasitic waves flying around.

Myself I have used a "star" config for ground. See on the pic below, the red stick point to a small ground connector where all the ground wires (brown) come. One from PSU board, one from tube board, one from the jack output (bottom right) and two to each pin 8 of my OPTs. Plus, the two small blue capacitors from B+ with a wire extension isolated in black.

You may note the respective orientation of the PSU transformer and the OPTs, so that they do not induce fields from one to the other. The other important point is how you connect input earth to audio ground. I used the exact setup from the original scheme (120R + 0.1 uF-X2).

Another potential source of hum is a ground loop. Try to use a floating source for your tests, e.g. a battery-powered smartphone with only an audio cable to the amp and not connected to a power source. Or short-circuit the input, the output should be silent.

The other things you see are not relevant: a led and its resistor at top-right.


The way it is now, it's dead silent without music; and extremely good sounding with music.
I'm not very proud of the box and the cabling, I used the wrong wood (beech) and it's twisting a lot. I will rebox the thing in a metal-case over the week-end.

So after several years of having this project sitting in my closet in various forms, I've finally got everything built up and I'm a very happy camper. No hum. Great tunes. Have a nice variety of NOS ECC88/E88CC tubes to try - Siemens, Telefunkens, Valvo, Mullard, Tesla. So far I like the Tele ECC88s the best. If I'm running these tubes at 158-160v B+ am I hurting the tubes? I did buy some 1.5k R2 resistors to swap in and bring down the B+. I should get them later this week or next week. I have an EHHA Rev A (ECC86 based amp) designed by Alex Cavalli and I have to say this rivals that amp at a fraction of the cost.

A special thanks to Vixr (for prototyping boards), Muskyhuntr (for implementing Vixr's prototype), and of course Bruce Heran for this super elegant design and really making this amp happen.

Here's a few pics...

Wow! Looks good. This is definitely one of my favorite amps.

Me

Mullet: That does look very nice! Good job mate!

My thought on your plate voltage is that 158-160V might be a bit high but if it sounds good, good enough for me. You are good to ask though as keeping things safe for your expensive tubes and getting the most runtime out of them is the question.

Hi, Nice looking build. This is one of my favorite projects....that almost didn't happen. All but my main system at the time had headphone outputs. One day I wanted to do something with my main system and thus was the cause to design and build the amp. The voltage across the tubes may be a little lower as you need to subtract the voltage on the cathodes from that applied to the anodes. The case on yours reminds me of the ones I just used for a set of modified Mini Blocks for my main system.

Good listening
Bruce

_________________
Some of my DIY Tube Amplifier Projects:

• PoddWatt: 5751 / EL84 Push-Pull DIY Tube Amplifier Project
• OddBlocks: KT88 Push-Pull DIY Tube Amplifiers (Mono Blocks)
• ForeWatt: ECC802S (12AU7 / ECC82) DIY Tube Preamplifier Project
• GrooveWatt: Tube (Valve) RIAA Phono Preamplifier Project

Thanks! Yeah, unfortunately I don't have the means to fabricate and mill the chassis myself so I pay someone to do it for me. For those wondering, Horace Atkinson of IAG Audio does the chassis work for me. As you can see, he does very good work.

Hi, I buy mine much of the time as well. The one in the photo came from a guy in Texas. I'll try to find the link to him. Quite reasonable and really nice. very heavy gauge aluminum (about 1/8th inch) with fitted bottom (with vent slots)that had threaded bolt holes to fasten it as well as nice wood sides and mounting feet.

Good listening
Bruce

_________________
Some of my DIY Tube Amplifier Projects:

• PoddWatt: 5751 / EL84 Push-Pull DIY Tube Amplifier Project
• OddBlocks: KT88 Push-Pull DIY Tube Amplifiers (Mono Blocks)
• ForeWatt: ECC802S (12AU7 / ECC82) DIY Tube Preamplifier Project
• GrooveWatt: Tube (Valve) RIAA Phono Preamplifier Project