Super Simple Single Stage Tube PreAmp

QUAD 33-303 Pre & Power Amplifier
B

bibin3210

Well-Known Member
Joined
Feb 1, 2012
Messages
463
Points
93
Location
N.Paravur, Cochin
Super Simple Single Stage Tube PreAmp (4S Universal Preamplifier for 12A*7 Tubes)

After completing my TubeLab SSE Power Amp I thought about building a tube preamp to mate with it. But I don't want any complex design and I want it to be as simple as possible. Mark Houston's Universal Tube Preamp catches my attention and its schematic looks pretty simple to build.

To bring the cost down without compromising on the sound quality I decided to use NOS 6N2P (Voskhod) tubes and rest all components from local market.

I had completed populating the PCB few weeks back, but due to time constraints I didn't got a change to test it. Today I connected it to my SSE and powered it and the sound was just amazing. Now the sound is more open and detailed...let it burn in for some more time.....:licklips:

Attaching few pictures of the completed PCB:







Thanks,
Bibin
 
The input is wide open. So all input signals will be amplified by the stage gain. If you input a DAC signal typically 1 to 2 V, the output will be 2x28 V ( 12AX7) = 46 V. The volume control will have a lot of voltage across it and you might have trouble setting it as a small rotation will produce a lot of signal voltage. The 12AU7 would be preferable with a gain at about 10. The 12AX7 output impedance is also very high . Only acceptable in some situations . With a typical 400pF load ( connecting cable plus connectors ) and 50 K load you will start having HF roll off starting by 10 Khz. Besides this with a 1 volt input the signal will clip as the stage is running wide open. The pot could be placed at the input though it leads to several possible disadvantages. Poorer s/n ratio and very sensitive pot position ( as earlier).
With a 12AU7 all problems are reduced and there will be no clipping even with a 2 volt DAC signal direct at the output. However running the stage wide open does have higher distortion than at low signal level if the pot is at the input.
If this stage is used after a phono preamp the 12AU7 would be fine.
Adding a cathode follower would be a good move though you need 2 tubes for stereo. Alternatively you can use a MOSFET follower stage which would have a slightly poorer performance.
To have low gain in a triode stage ( without negative feedback) you need other tubes and there aren't many with very low gain. It's a sticky situation !

The supply line rejection is only about 20dB and hence requiring very clean supplies. You could however get an additional 20 dB rejection at mains frequency (hum) by splitting the plate resistor or adding another small resistor like 2.2K in series with the plate resistor and the dc supply. Take the junction of the plate resistor and the 2.2K to ground via a 22uF capacitor.
 
Last edited by a moderator:
Fantastic said:
The input is wide open. So all input signals will be amplified by the stage gain. If you input a DAC signal typically 1 to 2 V, the output will be 2x28 V ( 12AX7) = 46 V. The volume control will have a lot of voltage across it and you might have trouble setting it as a small rotation will produce a lot of signal voltage. The 12AU7 would be preferable with a gain at about 10. The 12AX7 output impedance is also very high . Only acceptable in some situations . With a typical 400pF load ( connecting cable plus connectors ) and 50 K load you will start having HF roll off starting by 10 Khz. Besides this with a 1 volt input the signal will clip as the stage is running wide open. The pot could be placed at the input though it leads to several possible disadvantages. Poorer s/n ratio and very sensitive pot position ( as earlier).
With a 12AU7 all problems are reduced and there will be no clipping even with a 2 volt DAC signal direct at the output. However running the stage wide open does have higher distortion than at low signal level if the pot is at the input.
If this stage is used after a phono preamp the 12AU7 would be fine.
Adding a cathode follower would be a good move though you need 2 tubes for stereo. Alternatively you can use a MOSFET follower stage which would have a slightly poorer performance.
To have low gain in a triode stage ( without negative feedback) you need other tubes and there aren't many with very low gain. It's a sticky situation !
Click to expand...

Hi Fantastic,

I agree with you, this has got high out-put impedance and will not match with many of the amps. Matt has discussed this in detail in his forum and has recommended to used 250K pot at the output and has recommended to use a buffer like Pass B1 after the pot.

Dual stage with one cathode follower is a better option and I am working on that too and also a single stage tube buffer is in pipeline.

Thanks,
Bibin
 
Last edited:
@Bibin: any plans for the PSU for this pre? Perhaps something like the tube and choke regulated one you linked?
 
jls001 said:
@Bibin: any plans for the PSU for this pre? Perhaps something like the tube and choke regulated one you linked?
Click to expand...

Solid State PSU is already there in the board. Tube rectifier is avoided to bring the cost down. Choke is optional (but highly recommended) and can be added any time by removing one 150R/5W resistor.

Thanks,
bibin
 
Bibin,
The RC at the plate as I mentioned is very effective. You can use an even higher capacitor ( 220uF ) and get much larger supply rejection. Much less expensive than a choke which was typically used before large electrolytic caps became available.
The amp will be warm sounding ( liked a lot by some ) due to the high second harmonic content . However it should sound smooth ,due to the monotonically decreasing harmonic content with rising frequency...typical for such tubes circuits.

High output impedance could be used by careful design and sometimes could be useful to cut gain without resorting to additional resistors which also introduces noise.
But as I always say, if it sounds good to you, IT IS OK ! No matter what the measurements say.
Sometimes poorer measured amps sound better than amps with sparkling specs! The only final test is to listen to it. If you like it very much, don't even bother to test it ! :)
Main reason being you might find it hard to explain why it sounds better though it measure worse than some other equipment.
If you find it sounding nice just leave it as it is. Just make changes ( maybe add parts under the board ?) to improve the power supply rejection ratio.
I'm running a 2 tube preamp with ac heaters. There is hum right next to the speakers. From about 2 feet away you cannot hear any hum when the volume is set for loud listening levels.
Considering tube prices today, you should try to get by with as few tubes as possible. The MOSFET buffer is an alternative but I've tried it and it didn't sound very good. Acceptable but slightly inferior to the sound without it !
Simulations also shows higher distortion especially odd order harmonics. Not acceptable.
 
Fantastic said:
Bibin,
The RC at the plate as I mentioned is very effective. You can use an even higher capacitor ( 220uF ) and get much larger supply rejection. Much less expensive than a choke which was typically used before large electrolytic caps became available.
The amp will be warm sounding ( liked a lot by some ) due to the high second harmonic content . However it should sound smooth ,due to the monotonically decreasing harmonic content with rising frequency...typical for such tubes circuits.

High output impedance could be used by careful design and sometimes could be useful to cut gain without resorting to additional resistors which also introduces noise.
But as I always say, if it sounds good to you, IT IS OK ! No matter what the measurements say.
Sometimes poorer measured amps sound better than amps with sparkling specs! The only final test is to listen to it. If you like it very much, don't even bother to test it ! :)
Main reason being you might find it hard to explain why it sounds better though it measure worse than some other equipment.
If you find it sounding nice just leave it as it is. Just make changes ( maybe add parts under the board ?) to improve the power supply rejection ratio.
I'm running a 2 tube preamp with ac heaters. There is hum right next to the speakers. From about 2 feet away you cannot hear any hum when the volume is set for loud listening levels.
Considering tube prices today, you should try to get by with as few tubes as possible. The MOSFET buffer is an alternative but I've tried it and it didn't sound very good. Acceptable but slightly inferior to the sound without it !
Simulations also shows higher distortion especially odd order harmonics. Not acceptable.
Click to expand...

Fantastic,

Why do you prefer AC heaters in a Pre-Amp? I also observed that latest solid state rectifiers are good or better than tube rectifiers. In my SSE I have both kind of rectifiers on board and can switch between tube and solid state rectifier using a toggle switch, I couldn't hear any difference in sound. I added the solid state rectifier as a back up, in case if one fails I can switch to other.

Thanks,
Bibin
 
Well the very first reason reason to use ac heaters is it's 'simple' ! :)
The only other reason is because if you study the subject on heater voltage you will find that over time they say that using an ac signal prolongs the heater life. Now this is debated a lot but I'm inclined to listen to the real old timers who really were experts in tube matters and they preferred ac heaters.
A few people claim better sound too but I have no evidence for this. ( I tried to Google to locate where they had mentioned about heater life with regard to ac/dc supplies. As usual, you can't find something quickly when you really want it ! But I am certain about it as I have read about it many times and I've been into tube design for years and this topic has been discussed very many times. )
My vague recollection was that it had to do with the current reversing in the heater ( ac heater ) and hence ( being balanced !) avoiding some failure mechanism. Additionally in dc the current flowed only one way and there was a voltage gradient on it constantly only in one direction . But it doesn't matter. Use dc if the input signals are very low or you want the least amount of ( measured) hum no matter what !

When would I use dc as a heater supply ? In a tube phono stage ( low signal levels) as apart from a line stage with large signal levels ( typically 1 to 2 volts). There is always some mains hum leakage through capacitive coupling with the heater. In a line stage the hum levels exist but are so low that in practice it isn't really audible during normal listening levels ( even loud levels!).
So it's a tradeoff. Poorer s/n ratio ,though acceptable, and longer heater life and simple implementation.

You might reflect a bit on that. Do you want the best measured s/n ratio or a longer heater life with unobtrusive ( though measured higher! ) hum levels ?
Very often audible hum levels are caused by the physical construction rather than due to the ac heater itself. Goes without saying that the heater wires need to be routed properly !
With dc heaters you should ensure that rectification spikes do not get into the supply. They will very easily show up in the output as they are much higher in frequency and find it much easier to jump the capacitive coupling !
 
Last edited:
Fantastic said:
Well the very first reason reason to use ac heaters is it's 'simple' ! :)
The only other reason is because if you study the subject on heater voltage you will find that over time they say that using an ac signal prolongs the heater life. Now this is debated a lot but I'm inclined to listen to the real old timers who really were experts in tube matters and they preferred ac heaters.
A few people claim better sound too but I have no evidence for this. ( I tried to Google to locate where they had mentioned about heater life with regard to ac/dc supplies. As usual, you can't find something quickly when you really want it ! But I am certain about it as I have read about it many times and I've been into tube design for years and this topic has been discussed very many times. )

When would I use dc as a heater supply ? In a tube phono stage ( low signal levels) as apart from a line stage with large signal levels ( typically 1 to 2 volts). There is always some mains hum leakage through capacitive coupling with the heater. In a line stage the hum levels exist but are so low that in practice it isn't really audible during normal listening levels ( even loud levels!).
So it's a tradeoff. Poorer s/n ratio ,though acceptable, and longer heater life and simple implementation.

You might reflect a bit on that. Do you want the best measured s/n ratio or a longer heater life with unobtrusive ( though measured higher! ) hum levels ?
Very often audible hum levels are caused by the physical construction rather than due to the ac heater itself. Goes without saying that the heater wires need to be routed properly !
With dc heaters you should ensure that rectification spikes do not get into the supply. They will very easily show up in the output as they are much higher in frequency and find it much easier to jump the capacitive coupling !
Click to expand...

I will also experiment this and let you know my feedback.

BTW the following is the initial draft schema of the two stage I am working on:



Instead of ECC83 I will be using 6N2P, please ignore the values mentioned for time being.

Thanks,
Bibin
 
Last edited by a moderator:
Is the resistor in the plate circuit of the output stage a mistake or you want it there ? Why use C1 ? Why a cap across the cathode in the first stage ?
You don't want too much gain and this will increase it. Without the cap distortion will go down further due to negative feedback. While Zout goes up for the first stage ( without the cap) , it's followed by a very high impedance load and so it does not matter.
This is the standard implementation and it reduces distortion as the load on the first stage is very low ( high Zin of second stage) and you get a very low Zout.
Funnily all tube circuits have been worked out by the 40's and 50's I think. You can use tubes only in certain (simple!) ways unlike solid state devices.
However selecting the operating conditions will affect the sound either by measurement or audibly. But tubes are quite forgiving ! :)
 
Dear Sir,

Ive been following this thread thus I am curious to know why you are preferring the ECC83 family? In the modern setting it is not the best tube for this application. What is your amp input imp? Anything below 50K could be problamtic for this design.

If I had to stick to indirectly heated triodes then ECC82 variants is relatively better for this job.

This ECC83 in CF topology is good but you can do better with ECC82 in a grounded grid design. You will get better driving current and lower imp.

If you decide to stick to ECC83 then plan to do a higher B+ of about 320VDC and plate resistor 220k. Rk around 1.2K and you should be in a good linear region + lowest distortion for this tube. You will actually be harnessing the higher gain so that could mean a good thing distortion wise. And I agree with Shri Fantastic that you dont need the bypass cap. You are still looking at a plentiful gain of >45. You will need a very well filtered power supply and DC filament.

Im just quoting all this from memory thus take it FWIW . It will be helpful if you graph out the ECC83 data sheet for the best op point. However, dont loose your sanity in getting a precise op-point, accurate voltage etc, Tubes work happily within a broad range of deviations.

But whatever you do, I applaud the passion you bring to such things.

Best of luck

aks
 
From what I understand , Bibin is using the 6N2p just because he has it. I think he is quite aware of the restrictions. He also needs less gain and without adding external resistors there isn't too much he can do. However it will be a very interesting exercise to try and get what he wants out of the tube he has. He will certainly learn a lot from the exercise. ;)
 
At $50 that tube preamp is quite inexpensive. However you really won't know how good it is until you try it out.
Tubes start distorting 'more' the moment you load them ! That's how they work.
So a single tube preamp will have it's limitations and need not necessarily sound better than a solid state version. It will be more expensive also !
But it has it's 'glowing charm' ! :)

A well designed power amp doesn't really require a preamp ( if the gain is right!) since most sources today are DAC based and produce between 1 and 2 volts signal levels. But a 'buffer' meaning unity gain 'preamp' can often help with the sound by overcoming 'loading' issues with the source. A good tube buffer usually does very well here.
 
Last edited:
Fantastic said:
Is the resistor in the plate circuit of the output stage a mistake or you want it there ? Why use C1 ? Why a cap across the cathode in the first stage ?
You don't want too much gain and this will increase it. Without the cap distortion will go down further due to negative feedback. While Zout goes up for the first stage ( without the cap) , it's followed by a very high impedance load and so it does not matter.
This is the standard implementation and it reduces distortion as the load on the first stage is very low ( high Zin of second stage) and you get a very low Zout.
Funnily all tube circuits have been worked out by the 40's and 50's I think. You can use tubes only in certain (simple!) ways unlike solid state devices.
However selecting the operating conditions will affect the sound either by measurement or audibly. But tubes are quite forgiving ! :)
Click to expand...

That Rp2 is not added by mistake, I have something like Rp2 = half Rp1 in my mind when I add it there. Actually that is not required. The cap across Rk1 is actually an optional one for additional gain if needed.

I agree with you on the forgiving nature of tubes, I really experienced this characteristics while working.

I will do a simulation of these circuits and post the results here. However I don't have the .INC for 6N2P to simulate in LTSpice. If any of you have that data please share.

Thanks,
Bibin
 
Last edited by a moderator:
If I were you, I'd only spend time to simulate the circuit to confirm the dc operating conditions and check gain and overload limits and confirm it's frequency response.
Things like distortion content would have to be measured as most Spice data isn't very accurate for most tubes. They are just guidelines for basic operational information. Like, you will get some idea where you are without actually building it.
The FFT figures will not be accurate but you can see how it varies by altering operating conditions and the load. Then you can build it and modify parts to get what you want.....some call it 'fine tuning' ! :)

You could use the 12AX7/ECC83 Spice data to see where you are. DC conditions will vary but it's easy to change parts to get what you want eventually. The 6N2p isn't an exact equivalent but some parameters are similar. Resistor values could be calculated using the curves on the data sheet.
You'll may loose time chasing the 6N2p data. But you never know someone might have already put that up somewhere on the Net ! Maybe you can do that later ? :)
Here is a list of Spice data for tubes: 6N2P isn't there in this one !
http://ayumi.cava.jp/audio/pctube_1.11_win.zip
Look at the following thread on diyaudio.com . You might find the 6N2p buried in it somewhere .
http://www.diyaudio.com/forums/tubes-valves/243950-vacuum-tube-spice-models.html
 
Last edited:
Rega Turntables & Amplifiers
Order your Rega Turntables & Amplifiers from HiFiMART.com - India's reputed online dealer.

Similar threads

F
Build This Tube Amp With A Screwdriver
Replies
10
Views
8K
Amarendra
A
F
My 10 Year Old Granddaughter Builds a Tube Amp
Replies
10
Views
8K
linuxguru
L
F
Glo-Tone Boogie Factor 1626 Tube Preamp
Replies
1
Views
3K
FlaCharlie
F
S
Acoustic portrait Swara Tube buffer V2 (A review)
Replies
15
Views
5K
square_wave
S
R
ATC and tube amps
2
Replies
27
Views
14K
sajeet
S
Back
Top