• 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.

DHT preamp, 2A3 preamp

I built my first DHT preamp 30+ years ago without success because of hum. Recently, I built a DL92 preamp with very good result; low noise, great sound that outperforms all my preamps - ML38S, ML380S, Herron Audio VTSP-3, DIY WE437A. With this success, I am going one step further trying to build a 2A3 DHT preamp using cap output.

Yet, after searching this forum, I can't find any discussion in this topic. Is there anywhere that I can look for more information about this?
 
Just a guess, most DHT line-stages you will find in a search are based around 5V or > filaments, vs. the 2.5V 2A3, because of the difficulty in keeping hum low to acceptable levels. More difficult in a linestage then a SET power amp. You might have better luck if you search for 2A3 Line-stage rather then Preamp ...
 
Member
Joined 2004
Paid Member
Most 2A3 prone to microphony.

Some years ago I designed (and redesigned, and redesigned, and redesigned....) an OPT (Noguchi) loaded preamp to my friend. He tried several IDHT and DHT tubes, the 2A3 tubes was the most microphony tubes.
Now he use 801 graphite plate tubes.
 
I wouldn't use the 2a3 for one simple reason - it can't be used in filament bias. Filament bias for me is a must - I don't want to be using a cathode bypass cap. I hate any caps in the signal path, and try to reduce them to the best of a necessary evil - polypropylene caps throughout the PSU and one teflon coupling cap with a plate choke or gyrator anode load.

Choices for filament bias - 26, 01A, 10Y as higher Ra or 2P29L, 4P1L with lower Ra. The last two can be used with a plate choke of around 100H. The other three want more than that, even double. So what I'm constructing and hoping to finish today is a 10Y preamp stage with 2 of Hammond 126C in series as anode chokes, which will give me just over 200H at 15mA. In theory this is just what I want. I'll know soon if in practice it works.

So before fixing on the 2a3 I'd strongly suggest you look at options in filament bias. One of these is the Bartola gyrator which is good with the 01A. 01a Preamp (Gen2) – Bartola(R) Valves

For those that don't want solid state I would go with a plate choke like the Lundahl LL1667 or LL1668. For 01A go with 5mA, for 26 go with 7mA, for 10Y go with 10mA/15mA, for 2P29L go with 15mA, for 4P1L go with 25mA.

I'll post my 10Y circuit later when it's up and running.
 
Member
Joined 2004
Paid Member
TRAM2 has mixed bias.
 

Attachments

  • Tram2.jpg
    Tram2.jpg
    51.5 KB · Views: 2,113
I wouldn't use the 2a3 for one simple reason - it can't be used in filament bias.
That's absolutely not true. A 4R/75W resistor would bring 10V of bias to a 2A3, which would work well for filament bias.
Filament bias for me is a must - I don't want to be using a cathode bypass cap.
Filament bias isn't the only way to ditch the cathode bypass cap. A TL31 could bias up a 2A3 for 50mA/10Vb with no cathode bypass cap. A 9V battery biasing the grid could do the job nicely also.


The issue with the 2A3 is the current required. Using a 6A3 would be much easier.
 
That's absolutely not true. A 4R/75W resistor would bring 10V of bias to a 2A3, which would work well for filament bias.

All kinds of things are possible but not necessarily desirable. The 4R cathode resistor would need to be mounted on a heatsink probably, and I'm not keen on the sound of those aluminium clad resistors anyway. Each filament would need something like a 15v 60VA transformer. And that's without choke input - good luck finding a commonly available 3A choke. Operating point looks something like 45mA at 100v. It's all possible, as you say, but not something I'd build when there are several easier options.
 
Some of the lowest inductance resistors that you can buy are housed in aluminum heatsinks, but they are ridiculously expensive. This "resistors in aluminum heatsinks sound bad" generality isn't super helpful, especially when there are so many resistors constructed in that manner from so many different manufacturers.

There are gobs of 160VA transformers with dual 15-18V windings. That will work nicely. Choke input isn't required in this application. A 3A CMC for filtering is $3 from Mouser.

These aren't overwhelming roadblocks. The 4P1L isn't the answer for everything.
 
Some of the lowest inductance resistors that you can buy are housed in aluminum heatsinks, but they are ridiculously expensive. This "resistors in aluminum heatsinks sound bad" generality isn't super helpful, especially when there are so many resistors constructed in that manner from so many different manufacturers.

There are gobs of 160VA transformers with dual 15-18V windings. That will work nicely. Choke input isn't required in this application. A 3A CMC for filtering is $3 from Mouser.

These aren't overwhelming roadblocks. The 4P1L isn't the answer for everything.

I didn't say aluminium clad resistors 'sound bad' - just that they're not my resistors of choice. I do prefer choke input - sounds smoother to me. But you're quite right - there are plenty of DHTs worth playing around with. My case is a bit different - I just use a 2 stage amp, fed directly from my DAC. The reason I use 4P1L for my outputs is the gain of around 11. That allows me to use 2 stages with an input tube of similar gain, around 10 if possible or nearest to that. Tubes like the 2a3 and 300b would interest me for a 3 stage amp, but I don't need that, fortunately, and that has saved me several hundreds in cash. I used to use 2a3, 6C4C and 300b and still have a few. Some day I'll pull them out and use them.
 
Here it is - my new 10Y preamp. This supersedes all DHT preamps I've ever made (quite a lot) - more clarity, more neutral, better bass. It just sounds completely smooth and natural. Quite a big increase in clarity, in fact. I find myself listening at lower volumes since it's all so clear.

The Hammond 126Cs seem to be working well in series. A pair of these is cheaper than the Lundahl plate chokes, and I imagine just as good from my memories of a pair of Lundahls I once had. The Hammond are bifilar wound - one of their better products. I take the end bells off and mount them under the chassis. You just need the primaries. They're big and solid items. With 212H available, a pair in series should also work for a 01A or 26. Coupling caps are two FT-2 0.1uF in parallel, or you could use FT-3.

I'm very happy with this preamp - it's not going anywhere. Apart from the 10Ys themselves, it's not too expensive to make, though there's quite a lot of work in it. I use a separate choke input filament supply and a separate PSU. Three chassis.
 

Attachments

  • 10Y Pre 126C x 2.png
    10Y Pre 126C x 2.png
    80.7 KB · Views: 1,198
Gosh! Can you do that? That would be brilliant. I never knew that was possible. I've attached the data. You mean you connect Blue to Black or Blue to Green?

If this works it would be awesome. It's gapped for 15mA. Will it still pass 15mA as a plate choke? How is it "over 400H" rather than 212H?
 

Attachments

  • Screen Shot 2017-11-21 at 14.19.57.png
    Screen Shot 2017-11-21 at 14.19.57.png
    103.8 KB · Views: 905