diyAudio (
-   Tubes / Valves (
-   -   line preamp using 6SN7 instead of 12AU7 (

JPS 22nd May 2007 09:58 AM

line preamp using 6SN7 instead of 12AU7
I want to make this preamp with 6SN7 instead of 12AU7.
Could someone here please draw the schematic with part value suitable for 6SNTGT ?

ashok 22nd May 2007 10:14 AM

Why +/- supply for V2 ?
This is a line level preamp and so I guess need not supply more than 10 V into a load . A single supply can do that easily . Just seems to complicate matters .

So the input runs wide open and the volume control is in the middle. It will run quieter ( noise) at listening levels than with the volume control at the input. But then distortion will be lower with a lower input signal . Does it matter ? Maybe !

JPS 22nd May 2007 10:45 AM


A single supply can do that easily .
Simplified circuit with one power supply will be better.

ilimzn 22nd May 2007 11:06 AM

6SN7 will work fine without any other changes.
A single supply version will need another coupling cap to isolate DC from the pot, or, if you put the pot in front, the first and second stages can be DC coupled, and only one cap is needed at the output.
Since the amp is using current sources, it may be possible to reduce the supply voltage to about 200V or so. In case a DC coupled version is made, the heater supply will have to be elevated to half the supply voltage or 1/4 of the supply voltage if a silicon diode is connected between G and K of V2 - keep in mind that most 6SN7s have a maximum 100V limit for heater to cathode potential.

ErikdeBest 22nd May 2007 11:56 AM

Here is a nice article written by Broskie where the points by ilimzn are, besides explained, also illustrated!

This line stage could be theoretically improved upon with the addition of CCS on both the first and second tube, as shown in the schematic you have posted . To do this easily, you could buy the diyaudio CCS PCB's (which do have a better performance than the CCS on the schematic). SY posted an illustrative schematic of such a linestage somewhere, but basically it is about substituting R2 and R5 with CCS's.

Good luck


JPS 22nd May 2007 02:39 PM

Is it necessary to use CCS on both the triodes, as shown in the schematic?

Can we use CCS on one triode and still get good sound from this preamp ?

CCS on one triode and single power supply will be ideal.


ErikdeBest 22nd May 2007 03:18 PM


To keep things simple for the beginning, I would even recommend that you start building the line stage using solely resistors as loads and a single supply (as on the line stage of John Broskie). This is the most straightforwarded setup, that will already allow good sound and the chance for you to get other more fundamental things right, such as the PS, inputs, volume control, etc.

When you are done with this, listened a while and ready for more DIY action, you can replace the load resistors with CCSs. This may, indeed, be done for either of the triodes, or for both. And, most important, without having to change anything to the rest of the circuit. Just allow space in the chassis to accomodate the PCB's or veroboard for the components of the CCSs.

hope this helps


edit: do a search for the diyadio CCS boards, there is a pdf available there that treats the maths for creating and adjusting them. The diyaudio CCS boards with cascaded transistors will present 'theoretically' better performance than the CCS in your first schematic, as they present higher loads to the tube.

ilimzn 22nd May 2007 03:26 PM

Use of a CCS in the first stage (common cathode) makes for a very linear amplification stage, but also with the amplification factor equal to mu of the tube, in the case of the 6SN7 this up to 20, rather high for a line stage. The actual amplification factor depends on the loading by the pot, and is maintained at about 4 by the local NFB.

The construction of the preamp with the pot in the middle between stages is also a problem as for high level inputs the input stage is required to provide a large-ish voltage swing. In order to provide it with good linearity, a CCS is used, and a rather high power supply voltage. For a high level (2V RMS) input such as comes from a CD player, 60Vpp is swung by the first stage with no load, in reality this will be smaller due to loading by the pot. So, in essence, this is a question of the chicken and the egg - a CCS would not be so advantageous, if the pot was moved to the input. Also, direct coupling between stages could be used in that case. On the other hand, the pot in the middle approach has the advantage of the bandwidth being essentially unaffected by the pot position (which in the usual case it's not due to miller capacitance), however, in both cases wthis should be more than sufficient.

You could construct a 6SN7 preamp with the two stages and DC coupling if the volume pot is moved to the input. If the two stages have the same DC current draw, as in broskies 'constant current amp', a certain amount of nonlinearity compensation will occur and the complete amplifier will have a lower distortion than each of the satges separately. Assuring these conditions is however a matter of compromise as the actualload of the second stage is it's cathode resistor in parallel to the external load, which is an unknown. Such an amp could operate from a single power supply, with bias set up to have the plate of the first stage at about half of the supply voltage, and the heaters biassed at about 1/4 of the supply voltage. The diode between G and K of the second stage is mandatory. All of this has been discussed at length at John Broskies website.

EC8010 22nd May 2007 07:09 PM

A 6SN7 cathode follower could operate from the +250V rail (losing the +125V rail). The more significant problem (which others have highlighted) is that the first stage has too much gain; 12B4A (mu = 6) would make more sense.

Miles Prower 23rd May 2007 01:27 AM


Originally posted by ashok
Why +/- supply for V2 ?

Theoretically it eliminates the need for an output coupling capacitor. However, if you don't maintain a DC offset of 0Vdc at the cathode of the second triode, you risk poofing whatever it's driving.

All times are GMT. The time now is 01:38 AM.

vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2017 DragonByte Technologies Ltd.
Copyright 1999-2017 diyAudio

Content Relevant URLs by vBSEO 3.3.2