• 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.

Regulation options for 807 g2

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hi Everyone!

I'm currently planning a SE 807 amplifier based at this design:

807 Single Ended Amplifier at Triode Dick

The actual amplifier design will remain the same but not he PSU. First of all, I plan to skip these zeners used for stabilizing the g2 voltage. And here is the area for questions - I have found three options for generating a stable g2 voltage @ 250 V:

1. The zeners - what I have read and heard is that zener diodes are noisy. That's the reason I will not use them first hand.

2. Gas tubes (OC3, OD3) - this was an idea I saw at this forum. Sounds interesting and certanly is a show-off thing in the final amplifier. But how are the regulation properties for these tubes (accuracy & noise). A 10k resistor would be in series with the tubes and the entire thing will be fed from a B+ @ 450 V.

3. MOSFET regulator circuit. This is the option I know best as I have used it in previous amplifiers for regulation of B+. I'm not afraid of using semiconductors where they can give benefints. But I have never used them in an application like this and does not know how it will affect the sound. The MOSFET regulator will be (in case that is the best option) based on this circuit: 150V regulator


I hope You have some idea about this.
 
Noise is not usually a problem for G2 supply but regulation is, VR tubes have much lower dynamic resistance than high voltage Zeners and therefore make better regulators stand alone than Zeners. In noise sensitive applications it should not be forgotten that Zeners give an almost white noise but with increasing 1/F noise below a few 100Hz, therefore it is not that easy to make them quiet in applications where this 1/F noise can be a problem for instance in RIAA stages, VR tubes dont have this 1/F noise problem, (1/F noise in Zener based regulators was a big issue when I was designing low phase noise oscillators for microwave)

I have only designed one G2 regulator, that was for the AB1 amplifier with 807's and later 6146's I designed originally almost 35 years ago, I then used a VR tubes based regulator with a 6AU6 and a 6AQ5 if I remember correctly, I believe I got the circuit from the ARRL handbook, this circuit still works and now my nephew is taking over this amp and convert it into a guitar amp, still using the same circuit for G2 regulation with original tubes, quite a strong case for tube based regulators I would say...

Regards Hans
 
Good answer. There are always new things to learn.

Learning from this, I could go for the tube regulator for the g2 voltage and save the MOSFET regulators for the input stages B+.

That will be a first to me - just have to find a source for these OD3's and OC3's...

What is the estimated lifetime for these kind of tubes? Do I have to put some of these in store for the future?


I have one more question regarding the output stage: how critical is the B+ @ 450 V for the 807? There could be a problem finding a mains transformer giving these 380 VAC, so I want to know how low I can go with the B+ without modifying the circuit too much (the other voltages for input stages and g2 will not change). The only suitable option from Hammond is a 374BX that gives 375 VAC @ 175 mA. Could that one fit or is the current value too low (schematics states 300 mA for the 380 VAC)?

My primary option is the
807SE transformer set from Automatic Electric Europe, but they haven't answered my mail yet so I need to find a backup plan for the transformers to this amplifier. The output transformers are not a big issue (Hammond 1628SE) but the power transformer is.
 
diyAudio Senior Member
Joined 2002
Hi,

That will be a first to me - just have to find a source for these OD3's and OC3's...

No need to go for the higher current octal VRs here, a stack of OA2 and OB2 seven pin VRs will give you about 255V and will take up a lot less space.
All you need is a dropping resistor so they get the proper firing voltage and a small bypass cap.

What is the estimated lifetime for these kind of tubes? Do I have to put some of these in store for the future?

About 10.000 hrs, I think but they can still fail any day of the week so I'd definetely recommend keeping a few spares around.

Cheers,;)
 
Well with lower B+ you could have the option of going triode mode for less power and more easy design. 807s do pretty well in triode and you can probably put 350V on g2 with no worries.

Have you considered buying an ultralinear SE output tranny? It would eliminate regulators and is a nice comprimise between triode and pentode.

I am currently working on a PP 807 with fixed bias, will post some pics when done. Luckily my tansformer had a 200V widing, so with a diode bridge and a cap that will give 300V exactly, should be perfect!
 
Have you considered buying an ultralinear SE output tranny? It would eliminate regulators and is a nice comprimise between triode and pentode.
Got any suggestions of transformers? It is rather hard to find these things at a reasonable price. Hammond & Lundahl are the two options I have when staying local.

I have looked deeper into the VR tube thing and found (by reading some data) that the regulation accuracy is rather low (about 6 V for a 150 V tube). Is that critical for the g2 voltage? I assume a zener-controlled MOSFET regulator will have better precision in the voltage output, but may be overkill?
 
I have looked deeper into the VR tube thing and found (by reading some data) that the regulation accuracy is rather low (about 6 V for a 150 V tube). Is that critical for the g2 voltage? I assume a zener-controlled MOSFET regulator will have better precision in the voltage output, but may be overkill?

Yes, it sounds reasonably correct with 6V for something like 20-30mA change in load, if you compare withj Zeners you will see that that is worse. Either a VR tube + tube regulator or a Zener + mosfet or bipolar SS gives better performance as the current trough the reference can be made almost constant. However I believe that 6V is completely OK and is not a problem so maybe direct regulation using VR tubes is the way to go.

Regards Hans
 
Thank's everyone for the help.

The design have finalized and the practical part have started. The transformer issue was solved by "Ask Jan First", who could provide a custom made power transformer for a decent price. From the same place, I was able to source the output transformers too. Tubes are soon arriving from UK (bought via eBay). The remaining hardware could be easily found here in Sweden.

The final design got these properties:

* Actual amp design not changed.
* G2 stabilization by 0C3 + 0D3 (prefer octals)
* MOSFET regulators for 300/350V to input/driver stages
* Stabilized heaters for 6SN7's
* Filter capacitors will be larger in value.

I hope this will be a decent design. Lot of stuff for just 10W, but I hope this will be nice sounding 10W...
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.