• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

#26 pre amp

Member
Joined 2009
Paid Member
PSU for tube 26 preamp

Hey guys,

Finally started my first 26 PSU only for B+, hybrid rectifier AZ1, Bartolucci one power transformer Model CF125 125VAC 100W rendering 90% regulation 0,5% 4 Kg & Bartolucci one choke MR8L 10H 100mA VDC 200 DC resist 10 ohms 3 Kg, inside box only 2 schottky diodes 4A 600V Cree Z-Rec Sic + one MKP, next box 26 PSU heaters & yes I will go for 4 boxes (two PSU B+ + one PSU Heaters + one 26 tube preamp), a pic just for fun.
 

Attachments

  • DSC00051.jpg
    DSC00051.jpg
    439.7 KB · Views: 653
I'd use 3 boxes. I would put the filament supply in a metal box for screening if you have chokes in it - you need a 1 metre umbilical at least because if you have chokes they can hum. You may build without chokes first of all, but you may want to change the filament supply to a larger box if you do decide to add chokes.

A further option I'm exploring is putting two stages in the preamp box, since an AZ1 will comfortably feed four tubes. You'd have the option of direct coupling stage 1 and 2 and finishing with an output transformer. You'd then only need the output stage in the amp box, and you could experiment with a few different options including PP 2a3s or even a hybrid ss stage which could produce 20+ watts. If you put the PSU in the amp box you'd eliminate the preamp PSU and just have 3 boxes in total, with a larger choke input PSU that would power the whole amp.

andy
 
Last edited:
Hi Felipe,

the need for an external supplky has nothing to do with filament bias or cathode bias. For a preamp it's always better to have the PSU separated.

I see no need for three boxes however. I always put B+ and filament supplies in a single PSU chassis. As long as you have a final decoupling of all voltages either passive or with regulators in the preamp chassis, you are fine with a single PSU box.

Best regards

Thomas
 
Hi Felipe,

I see no need for three boxes however.

Three boxes is the deal killer that has kept me from building this amp (with the coleman regs.) I know I need these regs due to past experiements.

What about putting the B+ and filament transformers in a box with their rectifiers and a little filtering then up to the amp box via umbilical that holds the regulators and sensitive OPT's ? If we are regulating B+ with a SSHV this option would be OK?
 
If you can get the kind of high-quality, low flux leakage transformers that Thomas uses, and you are using SSHV, then 2 boxes will be fine.

Please take care that the wiring between trafo and rectifiers/capacitors is short, for both supplies, and place the transformers 6" [150mm] apart in a Alu case.

With my regulators, 1000uF of Panasonic FC, right on the PCB will suffice. No capacitors to chassis should be necessary, but you could try some Arcotronics or Panasonic stacked-MKPs (220n 100V) and see if the sound is better.
 
Hi!

I really don't see a need for three separate boxes for a 26 preamp. Two are sufficient if done right. This should be possible even with lesser quality transformers than mine (stupid to say that since I of course like to sell them).

Important for this is that each voltage has local decoupling in the preamp, close to the 26s.

I use LCLCLC filters for the B+ and LCL for the filaments. The first two LC sections of the B+ reside in the PSU chassis. A single B+ supply for both channels. In the preamp chassis I split into two separate LC decoupling sections per channel.

For filament, The initial LC is in the PSU. These are separate for each channel.
The final L (no C after it) reside in the preamp.

This should also work with regulators. B+ and filament regulators should be placed close to the 26 tubes.

Best regards

Thomas
 
Well, folks are looking for a general recommendation, but actually, much depends on the detailed design of your supplies.

If anyone wants to see which is best for their own chassis and power supply, I suggest some measurements.

You'll need an oscilloscope, or even better an electrical interface (protection, ac-coupling and maybe preamp) to your PC sound-card, and a nice open-source Spectrum Analyser.

1. Connect up the power supplies with the exact capacitors, rectifiers, and chokes that you plan to use.

2. Don't connect them to an amp though - attach a resistor dummy load, for the expected current.

3. Disconnect the mains to the B+ supply, disconnect the B+ rectifier, too.

4. Power-ON the filament supply, probe the B+ trafo secondary. Adjust the gain of the scope or preamp, and look for noise - noise only present when the filament supplied is powered. 50/60 or 100/120Hz stuff is not too important, but any sign of high-frequency pulsing is bad.

5. Repeat with Filament trafo disconnected, B+ powered.

6. Move one or both trafos to minimise any noise coupling you find.

In general, noise coupling will vary with transformer design, and degree of saturation. The length of the rectifier wiring, and the value of capacitance at the rectifier are also big factors, so it is hazardous to generalise too much. But if you measure, you will know!
 
Yes, naturally to avoid hum but what is the source of the hum?
If you're going to fix something it's much easier if you know what the problem is.



Rod's recommendation is a very good one. With a similiar amp's output I measured a 60 hz bump at the output about -90 down, which I could watch drop significantly the further the power transformer was to the OPT.

This sort of thing may sound immaterial if you have 95 db effecient speakers and don't mind a little hum <2ft away as they always say in the reviews. But for a headphon setup its critical.

Rod or anyone else,

Where can I find info on a interface/preamp that would handle B+ measurements with a good soundcard?