• 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.

5842 or EC8010 with OPT!

Hi Hauke,

I have no experience with the Euridice, but with transformer coupled preamps.

I personally think a tube like the EC8010 or 5842 are no good choices for a linestage, unless your sources have very low levels
and you need the amplification of these tubes.

Even with a step down transformer you will have something like
20 dB gain. Quite hefty for a linestage! That means you will have
to throw away a big part of the signal with the volume pot, which usually compromises sound.

If you want to get the maximum benefit from a transformercoupled linstage with low Zout, you need to place a low impedance volume control (i.e. 600) at the output.

But then you will have the problem of overload at the tubes' grids.
typical line output level of a CD player is 2V RMS, which is close to
6V pk-pk. These tubes are usually biased around 2-3V, which
would be barely enough and mean that you overload them permanently. Even if you stress it and run them at 4V bias and low current you have a mere 3dB of headroom and the tubes opwerate in a fairly non linear region.

My sugestions:

Use a tube which runs at higher (preferably much higher) grid bias. I.e. 27, 56, 76, 6J5, 6SN7 or better yet a directly heated
tube like 26 or 10/801 (my favorite)

Alternatively you can step down the voltage with an 4:1 input transformer. This would give you two benefits: The tube sees a very low DC resistance to ground (high gm tubes like that) and
you can drive a low impedance volume control which by this input transformer.

Just some thoughts, I can elaborate further if you're interested.

Viele Gruesse nach Hamburg aus dem sonnigen Muenchen


Hello Thomas,
thanx for your fast reply! First my introducion why i need a linestage with opt: I want to listen with my HD 600 earphones and i wanna drive several amps (biamping in future). So i need a lower output R. I have some #26,#76 in stock. I also looked at the wot amp:http://hjem.get2net.dk/aaholm_audio/projects/26Preamp/26Pre.htm
but what transformer should i buy? The Bartolluccis are very expensive, and i wanna be flexible in future.
Have you got schematics with the 6SN7?
Greetings Hauke

I suggest that you first determine the gain you need. This obviously depends on the output level of your sources and on the sensitivity of your power amp(s) and to some degree on
the sensitivity of your speakers.

Then you should get a clear picture of the impedance you
need to drive with your pre. I.e. 600 Ohms for the headphone ?
Based on that you can set a target for the Zout (preferably one third or less of the impedance you want to drive).

Based on that you can make a good choice of tube and transformer.

The best line output transformer I came across so far is the Lundahl LL1660.

I have no schematic with 6SN7, but trannie coupled linestages are very simple. You can work with the Euridice schematic and make the necessary adaptions.



2002-01-07 11:52 pm
Is this correct?

Hi Thomas and all

Also plan use 5842 in my setup.Can you check is it correct and logical on added ´´schema´´.
I wont try the Sovtek 2A3(have pair)as pseudo 45.



  • 5842 2a3.jpg
    5842 2a3.jpg
    45 KB · Views: 1,631

I can't see anything obvioulsy wrong with your circuit.
I think it will work.

However I have my reservations against the use of the 5842 as driver tube as I mentioned already. But this is a matter of taste, or lets say philosophy.

In order to get full power output, you will drive the driver tube at it's limit. That means it will contribute considerably to the total distortion. Now some people say that the 5842's distrotion spectrum harmonises well with that of a typical DHT.

However I see that as playing with distortion and the outcome is not really predictable. Since the driver tube needs to work quite hard, slight changes in bias will change it's distortion spectrum, which will produce a different blend with the output tube.

You have virtually zero headroom in the driver.

My philosophy is to have just enough gain in the system, but plenty of headroom. I shoot for a minimum of 12dB in the driver stage. This way the driver contributes very little to the harmonic spectrum of the amp. Its dominated by the output tube. This way the result is more predictable and needs less voicing. With such a headroom I can vary the driver bias a lot without noticably changing the sonic character of the amp.

I too use a DAC with resistor I/V conversion and a step up transformer. However my experience is that such current output DACs start to sound nasty if there is a large voltage swing on the current output. I shoot for something like 20mV (in my case a PCM1704). But that requires an additional gain stage. I see that you tried to do the whole amplification with as little stages as possible. I learned not to try to get the maximum out of each stage and rather use 3 stages which don't work so hard and have plenty of headroom.

It's years since I used the TDA1541 (I still have some gold crowns lying around). So I don't remember if they have considerable DC offset. You might want to check that. Even a few micro Amps can saturate the core of such small signal trannies. check the datasheet of the TDA1542 for the max offset and check with the trannie winder how much DC it can tolerate.

Please don't see my comments as critique, I just described my design philosophy. It doesn't mean that your circuit is necessarily bad sounding. In fact it can well be that it's exactly to your taste.

Try it, if it doesn't work out for you, than try some mods.

Oh one last comment: The output impedance of the 1:15 with the 20 Ohm at its input is something like 4500 Ohm. Don't know how the Amplimo is spec'ed, But I would use a lower impedance pot, like 10kOhm. The 5842 has a high Miller capacitance which forms a high pass filter with the 50 kOhm pot.

Also add a 'safety' resistor from the wiper to ground, say 470k. Just in case the wiper looses contact, so that the grid still has a DC path to ground.

Best regards



2002-01-07 11:52 pm
tubestage vice 1:15 transformer?

Thanks Thomas for good comment!

Maybe is better dont use any transformer at all?!
and use an extra stage with 5842 or 6C45P-E (have pair of both)like Lesha. I did so before got Amplimo TM-3 and remember sound was somehow warmer.After adding 1:15 transformers sound changed a little agressive. For 2A3(maybe some good nos 45 in future)driver use some suitable DHT. Any idea or schema where 45 used with DHT driver, what DHT?


nothing against the transformer. In fact I'm a big fan of
transformercoupling. But I try not to use excessively
high winding ratios. Except for MC step up and OPT I
try to stick with 1:4 or less. Higher ratios usually need
compromises in other areas.

In addition the transformer provides some 'natural'
filtering of the HF garbage which is present at the DAC
output. Plus and this is a big plus, it offers the opportunity
to isolate the DAC electrically from the amp (isolated grounds).

I do exactly that in my DAC which uses the BB PCM1704 with
resistor I/V and Lundahl LL7903 1:4 step up. Followed
by a EC8020 gain stage with Lundahl LL1660 4.5:1 step down
line output.

A nice DHT driver for the 45 would be the 26.



2002-01-07 11:52 pm
How to run 26?

Is LL1667 / 15mA 270H anode choke good load for 26 driver?
How do you recommend to run the 26 as 45 driver?
Are batterys for 26 filaments must be?
Did found from net RCA 26 only. What manufacturers
26 did you recommend?

Sorry for so many questions.

Re: How to run 26?

tõnu said:
Is LL1667 / 15mA 270H anode choke good load for 26 driver?

I'm not going to do your homework for you :)
I guess you can figure that out yourself...

How do you recommend to run the 26 as 45 driver?

What exactly is your question ?

Are batterys for 26 filaments must be?

Either battery or very very well filtered DC from
current source or passively filtered supply

Did found from net RCA 26 only. What manufacturers
26 did you recommend?

I'm not a tube roller and I don't compare tube brands



J Epstein

2002-02-08 7:24 pm
OK, I spotted this thread and I want to add my 2 pfennigs:

The transformer coupled line stage I am using I built almost as a joke, to see what the effect of microscopic source impedance would be on my system. It has remained in my system for over a year now with the only modification being an upgrade to the output transformers.

The line stage uses 6C45Pi, with a grid choke, loaded by a 5K:8 step-down (25:1 step-down) transformer such as would be used at the output of a 45 power amp. In fact the line stage can drive speakers directly - it's a "spud" amp. Bias is 2 AA batteries in the cathode for about 2.3V. I agree with Vinylsavor that this is just barely enough bias but with the grid choke it seems to work OK in practice.

The line stage feeds a transformer volume control of the type sold for controlling speakers in multi-room setups, and this feeds the power amplifier input.

The line stage source impedance is something like 5 ohms, and this allows you to use an 8 ohm nominal volume transformer which is very cheap to buy. I like the sound of this combination and it seems to perform well enough that I can reliably hear any and all changes in the equipment downstream from it, which is always being tweaked and changed.

One advantage of the low impedance is that cables seem to be less of an issue and you can run a simple twisted pair to and from the transformer volume control and not worry about hum.

I can't say I have really done "shoot outs" against other preamps but this is one way to go that I have used and enjoyed. Vinyl savor, you might remember we tried this ugly little spud preamp for a while in Arhus.

I have recently thought about regulating the power supply with gas regulators since it is so easy to try. Currently it is a simple CLCLC passively filtered supply.
Hey Epsteinerin! ;)

Good to have you in this thread.

2.3V bias on the tube ? What output level do your
sources have ? I assume your digital source has the
usual 2V RMS ? If so, the linestage clips at 0dB output! :eek:

I have done some measurements of grid current on various
high gm tubes, including this Ruskie firecracker.
Postive grid current starts anywhere between -0.5 to -1V bias.

One can argue about the sonic significance of grid current
effects, but for sure it adds non linearities.

For a linestage I would prefer to operate the tube at a minimum
of 4V bias to avoid grid current effects even at full output.
However with this tube then there is not enough headroom
on the negative side, which means it runs into cut off at

That's why I wouldn't really recommend this tube for line level

Hi J.

I guess the grid choke helps the grid current issue, but
if you have regular 2V RMS sources (quite sure) then the
pk voltage is +/- 2.8V. So in transients the Rooskie's grid needs to be driven 0.5V positive. This requires power, which your CDplayer or DAC most likely can't deliver. So the peaks will be rounded off. probably not to badly clipped, but still rounded.