• 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.

6N3 Tube Preamp with DC-DC converter

Whoa...
There is a lot of fuzzy questions and fuzzy answers in this thread!

The buffer (yes, as someone stated above, there is no gain, hence it is a 1x buffer) psu gives +/- 30Volts and that is 60V over the tube, fairly good but does not give room for the ultimate dynamics, do NOT use a 15 or 18VAC supply with the original components, the capacitors might pop.
The 7805 filament regulator gets really hot with 6N3, the filament voltage is too low with only one 1N400x diode, use two or better three diodes in series, or replace with a 7806 reg. with one diode.
The sound is surprisingly good when the filament is not starved and you have done all soldering correctly, considering this being a $20 kit from china...try to buy the same parts in the US for $20... ;)
I use a 2C51 tube and is quite satisfied with the sound.
 
  • Like
Reactions: 1 user
hello

if i am not mistaken this board uses 12 volt ac and retifies it with a bipolar power supply, -30 for the cathode and +30 for the plate. if you look at the schematic there is a 78l05 voltage regulator for the heater (5 volt regulator).

Input impedance : 100kOhm
Frequency response:10Hz-40KHz +/- 1dB
THD: >0.005%
S/N:95dB
Gain:˜0dB
Input voltage:AC12V/1000MA
Size:L 113MM W 65MM H 63MM
weight:100g

this is the specs the vendor sent to me on ebay..
 
hello

if i am not mistaken this board uses 12 volt ac and retifies it with a bipolar power supply, -30 for the cathode and +30 for the plate. if you look at the schematic there is a 78l05 voltage regulator for the heater (5 volt regulator).

Input impedance : 100kOhm
Frequency response:10Hz-40KHz +/- 1dB
THD: >0.005%
S/N:95dB
Gain:˜0dB
Input voltage:AC12V/1000MA
Size:L 113MM W 65MM H 63MM
weight:100g

this is the specs the vendor sent to me on ebay..

12VAC is correct, the data also seems to be in quite in order, the frequency response is as stated, the only thing i really dont believe is the distortion figure, my old distortionmeter says about 0,02% but it is all a matter of calibration (or lack thereof..)

I designed a similar PSU with a bit better regulation and a filament regulator that *do* give about 6,3V as the original one does only give 5,6V or something like that = starved filaments!

An externally hosted image should be here but it was not working when we last tested it.
 
I bought two of these and like them. I use one as an input stage to a pair of monoblock LM3886 chip amps, the other is a spare. The combo sounds quite nice. The only available schematic – the one that was posted here earlier -- is from an earlier revision. So, if anyone has a current schematic that’d be nice. I made the following revisions:

1. Swapped out all the caps in the power supply, nine 220uf and one 2200uf, with better quality caps.
2. Removed the caps from the input.
3. Subbed the regulator for a 7806.
4. Added some resistance in series with the LED illuminating the tube. (I like the effect, but it was too bright.)

I sort of hate the power supplies and the chip amp uses a compatible bipolar supply, so I am going to replicate just the buffer amp part of the circuit. I’ll power it with the chip amp supply and a filament transformer and wrap the whole thing up into a nice package.

Attached is a pic of my breadboard:

Here are the ebay links

Monoblocks:

6N3 Buffer amp:

Best,

John
 

Attachments

  • 6n3.png
    6n3.png
    130.3 KB · Views: 1,042
Greetings: NOOB here! I have made a pcb layout of the buffer section and one for the power supply. If someone would like to try these out I will post the layouts. I have made several LM3886 amps and this is my first attemp at a tube buffer. I am building several with 12au7 and 12ax7 buffers just for fun.
Best Regards, be well
Daniel
 
This little amp project is pretty strange. Being unity gain, it certainly shouldn't have any hum/noise problems from the filament, so running a 12 V tube right off the transformer (with a series resistor if it puts out a little too much) would seem to make more sense.

Some with access to thrift stores may find discarded 12V to cell-phone adaptors for the car for little or nothing. They contain a little step-down switching regulator that typically puts out about 3.6V. Change the feedback resistor and make the output 6 or whatever for a preamp tube. It'll take the rectifer 18 VDC or whatever is there and draw much LESS current than the tube filament does keeping things cool and minimizing the load on the little transformer.

I wouldn't expect a cathode follower to give much of a "tube" sound. Also, with low supply voltages and the fairly large 12K resistor in there, it won't be much good for higher current loads (like 600 Ohm headphones??). So as-is, I'm not sure I see the point of the circuit. I suppose a high-impedance input might be needed if one had one of the older tube-style tone-control network with really high resistances right ahead of it, but it would seem unlikely to have that outside of tube gear that already has a power supply.

If one wants a little stage with some gain or even distortion, it would probably work much better to move the 12K resistors to be in series with the plates, and take the outputs from the plate (through caps of course, + side to plate if polarized). And the spot where the 12K's are now would just be jumpered. there's already a bias resistor so the triodes should be happy.

Using a single supply with that multiplier makes more sense if only running these triodes (couldn't introduce noise between cathode and grid then), but the supply shown would be good for powering op-amps. Note that it isn't a regulator. The first transistor basically buffers the voltage on the filter cap at its base. If the beta is 50, it's like having a cap 50 times bigger as a filter. The ripple would be reduced by that factor. It would not provide teh ability to deliver surges of 50 times the current however.
The second transistor is a current limiter. When the current through the base-emitter resistor is enough to develop around .67 Volt, the transistor conducts robbing base drive from the other transistor, reducing output as needed to limit the current. I'm guessing someone copied that circuit from somewhere. There's really no need for a current limiter in the supply for a simple preamp, but it could be handy for other uses. If not needed, omit that transistor and just short the points where the 33 and 330 Ohm resistors are.

There is no need for a big capacitor at the output of the 5V regulator. Just a .1uF disc or so to prevent oscillation is fine. Some might like to simplify the circuit by using the LED instead of the 1N4007 on the ground pin of the regulator. That should boost the output of the regulator by about 1.6 V. A resistor from the output to ground pins of the regulator can increase the LED current. 510 Ohms would add about 10 mA to the tiny current that comes through the regulator ground pin.
A way to get totally "free power for the LED: the LED and series resistor could instead be added between the 3-terminal regulator input and output pins. Then the current drawn by the LED would reduce the current through the regulator by the same amount. (not that 10 or 20 mA would make it much cooler when running a 300 mA filament)
 
I got the JAN-5670W to replace the original 6n3 driver tubes in my amp and got a wicked Hummm. I removed them, put the original 6n3s back in then tried one of the 5670W, then the other. Didnt matter what configuration, either one or both of thr JAN-5670W's huumed and this was just within seconds of turning the amp on with no source or volumn.
 
Last edited:
First thing i d try is increase the 470R to the cathode. This will increase the bias, (hopefully) moving the point of clipping higher.

Increase? A higher figure like 1k or a higher current like 220R?

P.S.
And that tube is running just 60v. The solid state power supply is easily altered, and thus I am still wondering what is the Optimal voltage for 6n3p?
 
Hi,

I also bought this pre-amplifier to experiment, I found it to be quite good, my take and mods are:

1. change the input caps, to film type Wima, 0.1uF
2. change the output caps to polypropylene, 1.0uF Kimber caps

both of which are left overs from my past projects.

Also I have a commercial preamp, Project retail price at $300 with 8.6dB gain, based on IC's. I think this preamp out performs the Project amp. This cheap preamp presents a wide and bigger sound stage. I don't think I can hear the distortion.

The limitations of this preamps are:

1. Anode voltage is too low, about 32Vdc in my case
2. Cathode current is too low about 1mA, for driving high impedance amps
is OK.

As it is I don't think it can be improved further. My plan for enhancements are:

1. The topology remains similar with these changes,
a. 6N3 tube change to 6DJ8 set cathode current to about 6mA
b. 6DJ8 Anode voltage set at about 85Vdc
c. the cathode resistor removed, replaced it with constant current fet

2. The tube powered from 12Vdc battery with DC to DC convertor to boost it to 85Vdc

I think the above can be done. This will be my next project.

See attachment, the 6N3 pre-amp is the one in the translucent box, with the Project commercial pre-amp on top.
 

Attachments

  • IMG_0796.JPG
    IMG_0796.JPG
    103.2 KB · Views: 934

Bas

diyAudio Moderator Emeritus
Joined 2003
I have only bought the bare pcb's and some 6N3P-E's from a Russian shop off Ebay. Can I use the tubes for this design or not? Does anyone have a better design for a cathode-follower with 6N3P's cause I've heard these tubes sound quite good.
 
Bas I have only bought the bare pcb's and some 6N3P-E's from a Russian shop off Ebay. Can I use the tubes for this design or not? Does anyone have a better design for a cathode-follower with 6N3P's cause I've heard these tubes sound quite good. [/QUOTE said:
Actually it is cheaper to buy as a kit, because the costs of all the components if bought separately would cost more than the kit's price.
Actually it is better to buy the kit and replace some components with better quality parts e.g. coupling caps, power caps, resistors, etc.