• 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.

Looking for non-DRD 300B using D3a as driver

Hello,

why not using a high GM pentode in pentode mode :

- no Miller effect
- no bypassing cathode capacitor (local CR)
- only one stage drive, no preamp needed

Philippe
 

Attachments

  • D3a pentode drive.jpg
    D3a pentode drive.jpg
    13.5 KB · Views: 916
  • EL802-300B.jpg
    EL802-300B.jpg
    36.8 KB · Views: 804
The article mentions two follow up articles. Would you happen to have access to those? I will search, but time is limited at work..
Glad you like it.

I will post the second article but to be honest you don't learn anything new. (That is the reason I did not translate the third article) It's more of the same just on a different amp if I remember correctly.

There is another article on psud someone wrote for my diymag. I'll see if I can seperate that and post it here.
 
PhillippeS,
Can you please clarify something in the schematic you presented? There is a junction (I think) below the 27K and above the 47uF where this path to ground crosses the lead out of g2. does this all connect there at that 4-way junction? Or does the wire from the 27K to the 47uF cross-over the lead out from the g2 without actually making a connection there?

Bas,
that was a perfect article! It really answered alot of questions I had about power supplies. I was wondering how one measures the power transformer output resistance/impedence you were mentioning in the article (I think an example had a value of 710R). Is this a simple measurement with a multimeter across the secondaries, basically the DCR of the secondary windings or is it something more than this? If so, how does one calculate it or is it something given in the manufacture data sheet of the transformer (If so, I don't think I have seen it before).

I was wondering what the advantage/disadvantages were of using the D3a in Triode vs. Pentode as well other than the obvious of higher output in pentode. Why would you choose one over the other is what I think I am asking?

To date it has been suggested to me to connect the D3a in triode and that has been my plans so far. I don't think I need the extra gain from a pentode connection either. Also, the pentode requires more parts and is a bit more complex....I want to keep it as simple as possible. I have never worked with zeneers either to date. I think the schematic shows x3 51V zeneers, when a zeneer is purchased are there any other ratings other than voltage I should concern myself with?

Thanks again to all,
Jeff
 
Hi Jeff,

I was wondering how one measures the power transformer output resistance/impedence you were mentioning in the article (I think an example had a value of 710R). Is this a simple measurement with a multimeter across the secondaries, basically the DCR of the secondary windings or is it something more than this? If so, how does one calculate it or is it something given in the manufacture data sheet of the transformer (If so, I don't think I have seen it before).
Here is an answer to your question. Otto Axel wrote an article for my mag to do just that.

http://www.basaudio.net/pubs/DIYMAG_2006_2.pdf page 10-13.
Thanks again to Otto Axel.
 
I promised you to draw a schematic and here it is.
I have drawn the shunt mod, and the output stage as well, but I didnt check values, but used the ones you gave me. The shunt mod has one disadvantage, at full volume it still acts as a voltage divider 2:1. But the d3a has a gain of 70x, so the 35x left is still plenty. I presume you don't have a pre-amp? because if you do, I would leave the potmeter out.

Sorry it is a little late, but I had to put some Ikea furniture together.... Billy the bookshelf x 6 :rolleyes:

Good luck!
 

Attachments

  • Jeffmiller.jpg
    Jeffmiller.jpg
    129.3 KB · Views: 784
Last edited:
Pauldune,
Thanks alot it looks great. I think the only difference is I have 0.33 Russian PIO's I will be using.

I was wondering if I should maybe place some resistance to bring the voltage down some on the CCS....maybe a B+ of around 300-325 rather than the full 435v for the output stage. This should decrease the heat and sinking I will have to use for the CCS. Or am I mistaken? If it is a good idea to bring the B+ down a bit for the CCS is there a particular voltage you would target?

Is there any way to use the CCS for the 300B stage as well? Just curious, I haven't seen this before and I am sure there is a reason. Maybe it is just I haven't seen it too. Not sure. Either way, any use for this for the output stage at all?

Jeff
 
Jeff-
the closest I've seen is the Tubelab SE which uses a different CCS for the input tube (a 5842) and a MOSFET follower for the grid of the 300b.

I don't think the 300b would lend itself to a CCS as a load since the load has to be the output transformer. Maybe a CCS at the cathode?
 
I suggest you keep the fixed bias on the output. With a sufficiently quiet PSU and high quality caps from +B to Common this sounds unusually good (I have done it both ways often enough by now).

I'd even go fixed bias on driver stage, if it was me...

Ciao T

Thanks Thorsten - I will retain active bias. I am not sure what you mean by "high quality caps from +B to Common" though. I also have a few other questions, so will start a new thread in the next few days. I am tired of procrastinating about a build - the time has come for doing.

Cheers
Raymond
 
Hi,

Thanks Thorsten - I will retain active bias.

You can use traditional fixed bias as well, it is just annoying to regulary have to re-adjust things.

I am not sure what you mean by "high quality caps from +B to Common" though.

Common or Ground is in the fixed bias amplifier the "reference" to which the 300B Cathode is directly connected. The signal current in the output loop circulates from cathode via the anode and output transformer back to the cathode.

As a result the capacitor in the power supply that offers the lowest impedance will carry this current. If the supply has a high impedance element (choke, electronic choke etc.) this capacitor will be the final PSU capacitor from the supply to the Transformer and 300B anode to ground. The differences in quality of this capacitor are quite audible.

Ciao T
 
As a result the capacitor in the power supply that offers the lowest impedance will carry this current. If the supply has a high impedance element (choke, electronic choke etc.) this capacitor will be the final PSU capacitor from the supply to the Transformer and 300B anode to ground. The differences in quality of this capacitor are quite audible.

Ciao T

Hi, I am wondering what properties are the most important when you say it is audible. Losses, ESR, etc....what in you opinion makes the capacitor "high" quality in this regard. I guess I should also ask what your preference is in this final cap....Poly, poly in oil, Black Gate or other special composition electolytic types (SimilicII, etc..)? I have used either a Film in oil ASC type or Solen PP as my final cap before the output stage with excellent results. I have yet to try a BG mostly due to cost of the power tank types. Are there other electrolytic parameters that I should look for if I choose to use an electrolytic....as stated before, there are different specs I am aware of like the ESR but I am unsure which of these actually make the cap "good sounding"

I have recently considered the Obbligato Film power filter caps but I have a premonition they will behave like a Solen or any other descent PP cap.

Jeff
 
Common or Ground is in the fixed bias amplifier the "reference" to which the 300B Cathode is directly connected. The signal current in the output loop circulates from cathode via the anode and output transformer back to the cathode.

As a result the capacitor in the power supply that offers the lowest impedance will carry this current. If the supply has a high impedance element (choke, electronic choke etc.) this capacitor will be the final PSU capacitor from the supply to the Transformer and 300B anode to ground. The differences in quality of this capacitor are quite audible.

Ciao T

Thanks again Thorsten.

I had not seen (or more likely forgotten!) the term "common" used in this context; signal "ground" I kind of understand. Your description of the current loop in fixed bias also helps. I have not really looked at fixed bias designs seriously; seems it is time for me to do some reading.

Jeff,

IIRC Thorsten does not care for Solen metalised, but has some regard for the Obbligatos... at least the film in oils. That said, Thorsten appears to have changed some of his design preferences during his absence for the forums, so this may have changed as well. As ever, impermanence (change) is a constant.

Cheers
Raymond
 
Pauldune,
Thanks alot it looks great. I think the only difference is I have 0.33 Russian PIO's I will be using.

I was wondering if I should maybe place some resistance to bring the voltage down some on the CCS....maybe a B+ of around 300-325 rather than the full 435v for the output stage. This should decrease the heat and sinking I will have to use for the CCS. Or am I mistaken? If it is a good idea to bring the B+ down a bit for the CCS is there a particular voltage you would target?

Is there any way to use the CCS for the 300B stage as well? Just curious, I haven't seen this before and I am sure there is a reason. Maybe it is just I haven't seen it too. Not sure. Either way, any use for this for the output stage at all?

Jeff

The 0.33 pio's wil be more than fine.

The lowering of the B+ seems like a good thing to do, but when I tried it, the CCS I used (the ixys 10m45s) started oscillating. Directly connected to B+ it didnt do that, so for me the choice was simple. Anyway, its not that much power, so with a nice heatsink it shouldnt be any problem.
With 300v B+ you still have about 200v pp swing, and thats more than you need to drive the 300b. just put a 6k8 10W resister between B+ and the CCS.
But I wouldnt do it, you shift 2.7 watt from the CCS to the resistor. Plus you need an extra resistor and an expensive cap.

You can't use the same CCS for the driver and the end tube at the same time. You can however use another one. You have to use the output stage parafeed with a decoupling cap. But it wont do the same thing as with a driver. With a driver, because of the very high impedance of the next stage, the current through the drivertube stays practically constant. The loadline becomes a horizontal line.
With an end-tube, its a different story. the current through the ccs remains constant, and for dc it flows all through the tube, but for ac it splits. part goes through the tube and part goes through the output transformer. so the load line isnt horizontal, because the anode current isnt constant. What the loadline would look like, I wouldnt know, but i suspect something in between.

One of the advantages of using a ccs for the output stage is the decoupling from the powersupply, its not in the signal path anymore. also, the powersupply sees a constant load; it doesnt matter if you play hard or soft, transients etc.

I tried to use those qualities in my design:here