• 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.

6SN7 based tube buffer

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

I have a cheap 6SN7 based tube buffer as the attachment but the mid/low area sounds quite muffled and the highs are not extended enough to my taste as well. Also during the playback of some loud passages, it could sound compressed and somewhat distorted.

I can see C24 and C34 are WIMA black box caps and C26 and the other paralleled caps are WIMA MKS4 and Philips MKT. I don't know much about electronics but can a simple tweak like replacing the C26 C24 C34 to some audio grade capacitors improve the sound quality?

P. Sugano


  • buffer.gif
    8.1 KB · Views: 1,296
What are the ( resistor) component values like and what voltage do you have at the plate ?

C29 at the input will cut HF . What value do you have and the resistors with it ?
Do you need the extra gain that you get with parallel triode use. If not then a single half might give better sound. With 240 volts at the supply , you should get clean sound.
The 6SN7 is a good tube.
The resistor values will help a lot in giving you some suggestions.
Muffled, huh? And then people tell you paralleled sections sound just as good but with half the output impedance.

If you are reluctant to undertake such an extensive surgery upgrading the black boxes will still bring some improvement. There is no need for the input cap assuming you use quality sources with no or low dc.
Your main problem with this circuit is a high source impedance. Even with paralleled sections, the 6SN7 is not a good candidate for this topology. It is probably the best tube out there for a real buffer circuit, but this circuit isn't one.

If you were to redesign, unparallel the sections, use the first as a common-cathode amp, then the second as a cathode follower.
I agree with Stuart.

The 6SN7 has plate impedance around 7K; so a T-barred bottle will give 3.5K in parallel with the plate load (which should also take account of the AC load on the cold side of the output cap.)

Ergo, with maybe a plate load of 3rp, say 10k, you'd have a Zout of no less than 2.6K. With the wrong interconnects and/or a low Z amp, this could sound rather indistinct, agree, but this might also be exacerbated by sag in the B+ supply.

There is an attempt in the circuit to trade gain for lower Zout, by use of the resistive divider, but this may not solve the problem either even though it will decouple the tube plate from the load.

The better option is to direct couple a CF to a plate loaded voltage amplifier as per Stuart's suggestion.

Some subtlety is necessary here, too. Choose around 22K plate load (giving you, with a cathode bypass cap, an Av around 75% of mu, or 14), then use 25K as the CF tail resistor. This will mean that as the VA draws more current, the CF draws less, but in a slight imbalance which will promote some H2/H3, or sweetness. A gain of 14 is far more than you need, in which case you should first delete the cathode bypass cap on the voltage amp, then perhaps further attenuate the input to the voltage amp with a simple resistive divider. Zout at the CF should be around 200R, though it is asymmetrical, but this should not cause problems.

You should run around 7.5mA in the first section, which mandates a supply of at least 300V, and perhaps even 320V.

As you can see, this is a very different design configuration, but it will sound much better and enable use of the same tube.


Hugh, it's even worse than you make it out to be. Quantifying the source impedance is impossible without circuit values, but the unbypassed cathode resistor guarantees that the source impedance will be higher than 7K.

To add insult to injury, the B+ is mighty low for this tube. A 6SN7 wants to see minimum 180V on the plate.
Thanks for the pre amp circuits!

As the original buffer is fed by the output of a PCM2702 USB DAC, how do I replace the old buffer with the circuits you uploaded? I know these are preamp circuits so the volume pot has to go but to make the new circuit work with the pcm2702 output, what change do I have to make precisely?

Thank you
planet10 said:

Here are a couple variations Frank posted some time ago.


Hi Dave,

Thanks for uploading Frank's preamp circuit. Now what do I need to do to connect this circuit to the output of a PCM2702? It is going to be a USB DAC so I don't need a volume pot but what else do I need to do to make it work? Someone please help.

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