Some background first: Over the past couple of years I have on and off built tube amps from schematics taken from the web. Initially I built a 5687 pre-amp and a EL34 triode connected SE amp attached to a FE167 full range speaker with reasonable success. I used this set-up in my garage for a few months using a cheap MP3 player as the source and it all worked great. Then over the next few months I really got addicted and I built more power amps, Darling DC coupled, Bevois Valley EL84PP and finally a 6EM7 DC coupled amp.
Then I decided to move some of the kit into the house and that’s when the problems started. When I set-up the preamp + any of the power amps with my CD player (Naim CD3) as source I had way too much gain (at 8-9 o’clock volume was way too high). So I built a 12B4 pre-amp to replace the 5687 pre. This was slightly better but sadly I found that I still had too much output (at 9 o’clock on volume pot.)
I then tried just connecting the CD player through a 100K volume pot to the power amp but the sound was a bit ‘flat’ & lifeless.
After some searching on the internet I came across this article that seems to offer a potential solution to my problems.
http://www.tubecad.com/october99/page8.html
Based on the article it looks like I can adjust the gain using the ratio of resistors (R1 & R2) to suit my set-up.
I have a schematic from the web using a D3A tube connected as triode: B+ ~270V, Ra=5K6, R1 = 100k, R2 = 150k, Rk = 82R (no bypass cap) Rl = 1M & C1 = 0.1uF
However, the real question though is what is the recommended/best way to connect the volume pot (100K) and buffer?
#1: CDP > Buffer (anode follower) > Vol Pot > Power amp > Loudspeaker or
#2: CDP > Vol Pot > Buffer (anode follower) > power amp > Loudspeaker
Any help is appreciated.
Cheers
Colin
Then I decided to move some of the kit into the house and that’s when the problems started. When I set-up the preamp + any of the power amps with my CD player (Naim CD3) as source I had way too much gain (at 8-9 o’clock volume was way too high). So I built a 12B4 pre-amp to replace the 5687 pre. This was slightly better but sadly I found that I still had too much output (at 9 o’clock on volume pot.)
I then tried just connecting the CD player through a 100K volume pot to the power amp but the sound was a bit ‘flat’ & lifeless.
After some searching on the internet I came across this article that seems to offer a potential solution to my problems.
http://www.tubecad.com/october99/page8.html
Based on the article it looks like I can adjust the gain using the ratio of resistors (R1 & R2) to suit my set-up.
I have a schematic from the web using a D3A tube connected as triode: B+ ~270V, Ra=5K6, R1 = 100k, R2 = 150k, Rk = 82R (no bypass cap) Rl = 1M & C1 = 0.1uF
However, the real question though is what is the recommended/best way to connect the volume pot (100K) and buffer?
#1: CDP > Buffer (anode follower) > Vol Pot > Power amp > Loudspeaker or
#2: CDP > Vol Pot > Buffer (anode follower) > power amp > Loudspeaker
Any help is appreciated.
Cheers
Colin
Hi,
I wanted to use the anode follower as I could achieve a small amount of gain by adjusting R2, whereas with a cathode follower, as I understand it, gain is usually <1.
I take your point about the d3a tube though but as I already have these tubes available I'm good to go with these.
I guess the answer is build the anode follower try the vol pot in front of the buffer then switch to the other option and decide which suits best. I guess I was hoping some-one would already have experience of both options and could save me some time.
Cheers
Colin.
Why would you use an anode follower (and an expensive tube!)for this application instead of the more generally recommended cathode follower?
I wanted to use the anode follower as I could achieve a small amount of gain by adjusting R2, whereas with a cathode follower, as I understand it, gain is usually <1.
I take your point about the d3a tube though but as I already have these tubes available I'm good to go with these.
I guess the answer is build the anode follower try the vol pot in front of the buffer then switch to the other option and decide which suits best. I guess I was hoping some-one would already have experience of both options and could save me some time.
Cheers
Colin.
OK, well then, just a few comments.
1) IMHO, never build a preamp with a 0.1 uF coupling cap and a 1M output resistor, it's basically unusable except for with a homebuilt high input impedance amp. Even then 0.1 uF is really small.
2) With those values you listed, gain is 0.42 by my calcs. A cathode follower doesn't look so bad
3) With those values, I get about a 1400 output Z. I wouldn't just swap that 100K pot you plan to try on the input into that position.
1) IMHO, never build a preamp with a 0.1 uF coupling cap and a 1M output resistor, it's basically unusable except for with a homebuilt high input impedance amp. Even then 0.1 uF is really small.
2) With those values you listed, gain is 0.42 by my calcs. A cathode follower doesn't look so bad
3) With those values, I get about a 1400 output Z. I wouldn't just swap that 100K pot you plan to try on the input into that position.
Hi leadbelly
Thanks for sparing the time to go through the calculations. It looks like I'm going in the wrong direction with this one. I think I'd better start doing some background reading on cathode followers.
A quick search on the forums here as turned up plenty of reading so I'll start with these.
Cheers
Colin.
Thanks for sparing the time to go through the calculations. It looks like I'm going in the wrong direction with this one. I think I'd better start doing some background reading on cathode followers.
A quick search on the forums here as turned up plenty of reading so I'll start with these.
Cheers
Colin.
D3a Linestage build
Hi,
I didn’t have a great deal to do to-day so I out of curiosity I went ahead and built the D3A anode follower as described in post #1. It worked but probably not correctly.
I’ve attached a pdf file that shows the oscilloscope traces for 6Vpk-pk 10kHz square wave input signal and the corresponding output from the buffer circuit and output measured after volume pot at different settings of volume pot.
Strange thing is the buffer circuit didn’t appear to invert phase of signal and also circuit produce gain. This doesn’t fit with calculations carried out by leadbelly.
As I don’t really understand how these circuits work I’m unable to interpret the data from the oscilloscope traces although it looks like as the volume pot is increased the square wave is less well defined – probably a really bad thing.
Anyway, I guess I’ll just move onto a cathode follower now – it seems to be the recommended way forward. On the other hand I suppose ought to spend more time ready Morgan Jones instead of building more amps!
Cheers,
Colin.
Hi,
I didn’t have a great deal to do to-day so I out of curiosity I went ahead and built the D3A anode follower as described in post #1. It worked but probably not correctly.
I’ve attached a pdf file that shows the oscilloscope traces for 6Vpk-pk 10kHz square wave input signal and the corresponding output from the buffer circuit and output measured after volume pot at different settings of volume pot.
Strange thing is the buffer circuit didn’t appear to invert phase of signal and also circuit produce gain. This doesn’t fit with calculations carried out by leadbelly.
As I don’t really understand how these circuits work I’m unable to interpret the data from the oscilloscope traces although it looks like as the volume pot is increased the square wave is less well defined – probably a really bad thing.
Anyway, I guess I’ll just move onto a cathode follower now – it seems to be the recommended way forward. On the other hand I suppose ought to spend more time ready Morgan Jones instead of building more amps!
Cheers,
Colin.
Attachments
Hi,
I used Winzip 12 to compress the file - downloaded from here
winzip.
Perhaps I set the compression to the wrong setting? I found it quite difficult to get the file size down to the below the size limit set for posting here.
I'm at work now so I can't re-post it until tonight as the file is saved on my home PC only
Sorry about that!
Cheers
Colin.
I used Winzip 12 to compress the file - downloaded from here
winzip.
Perhaps I set the compression to the wrong setting? I found it quite difficult to get the file size down to the below the size limit set for posting here.
I'm at work now so I can't re-post it until tonight as the file is saved on my home PC only
Sorry about that!
Cheers
Colin.
Hi,
Here is another attempt at attaching a zip file. Hopefully this one will unzip successfully. Thanks for sticking with me on this
At the moment I haven't attempted to build a cathode follower yet- I'm still considering the possibilities. My 12b4 line amp was just a simple common cathode pre-amp based on info from this link:
http://www.diyaudio.com/forums/showthread.php?s=&threadid=68267
A good thread on cathode followers is here
http://www.diyaudio.com/forums/showthread.php?s=&threadid=34335
I guess you may have seen these before - if so sorry for wasting your time!
Cheers
Colin.
Here is another attempt at attaching a zip file. Hopefully this one will unzip successfully. Thanks for sticking with me on this
How's the CF turn out? Me too having problem with too much gain and have spare 12B4. Thx
At the moment I haven't attempted to build a cathode follower yet- I'm still considering the possibilities. My 12b4 line amp was just a simple common cathode pre-amp based on info from this link:
http://www.diyaudio.com/forums/showthread.php?s=&threadid=68267
A good thread on cathode followers is here
http://www.diyaudio.com/forums/showthread.php?s=&threadid=34335
I guess you may have seen these before - if so sorry for wasting your time!
Cheers
Colin.
Attachments
here
thread number 133602.
Learnings from the valvewizard and ideas from Joel and Eli.
THe valvewizard website is great - I find the math there easy to understand for a change, and applying hte logic was easy. I've no doubt that there are specifics around hte design I used that could be optimised, but for learning it is greatand easy.
Give it a go!
thread number 133602.
Learnings from the valvewizard and ideas from Joel and Eli.
THe valvewizard website is great - I find the math there easy to understand for a change, and applying hte logic was easy. I've no doubt that there are specifics around hte design I used that could be optimised, but for learning it is greatand easy.
Give it a go!
a nice low gain line amp can be made with a 6S19P, gain will be in the 2-3x range. output Z will be extremely low. they even work nicely below 50V plate voltages. much easier to work with than say a 6AS7 (microphonics, heater hum, channel seperation etc). and they're dirt cheap
cheers,
mb4b
cheers,
mb4b
Hi,
makeb4break:
This sounds like a good option for me - right amount of gain. I'll investigate this further. I think member sorenj made a line amp using 6S19P - I'll carefully read his thread later. Thanks!
aardvarkash10:
Thanks for the tip re: thread number 133602 - again more reading for me.
It's a shame about the anode follower that I made with the D3A - it appears to work but the oscilloscope traces do not make much sense (to me) and I don't have the right know-how to sort this out. I guess a small amount of knowledge is sometimes dangerous! At least I have a breadborded psu I can probably use for future line amp projects.
Cheers
Colin.
makeb4break:
a nice low gain line amp can be made with a 6S19P, gain will be in the 2-3x range.
This sounds like a good option for me - right amount of gain. I'll investigate this further. I think member sorenj made a line amp using 6S19P - I'll carefully read his thread later. Thanks!
aardvarkash10:
Thanks for the tip re: thread number 133602 - again more reading for me.
It's a shame about the anode follower that I made with the D3A - it appears to work but the oscilloscope traces do not make much sense (to me) and I don't have the right know-how to sort this out. I guess a small amount of knowledge is sometimes dangerous! At least I have a breadborded psu I can probably use for future line amp projects.
Cheers
Colin.
How to convert D3a pentode to triode?
This is a simple question in connection with the anode follower buffer (D3a) I have been playing with.
According to the D3a datasheet: g3 goes to cathode & g2 to anode.
Reading Morgan Jones ‘ Valve amplifiers 3rd Edition’, p119, he states g2,g3 to anode.
I’m confused here – which is the right way?
BTW, I have connected it as MJ recommendations at the moment.
Cheers
Colin.
This is a simple question in connection with the anode follower buffer (D3a) I have been playing with.
According to the D3a datasheet: g3 goes to cathode & g2 to anode.
Reading Morgan Jones ‘ Valve amplifiers 3rd Edition’, p119, he states g2,g3 to anode.
I’m confused here – which is the right way?
BTW, I have connected it as MJ recommendations at the moment.
Cheers
Colin.
Something may be escaping me here, but we are talking about ADDING A STAGE because there is TOO MUCH GAIN?
Get a grip. This is fundamentally 4ss-backwards. Reduce the input to the PA with a resistive divider or attenuator. This will have no perceptible effect on the sound quality.
w
'Flat' and 'lifeless' are just another way of saying 'quieter'. They have no other meaning in this context.
Get a grip. This is fundamentally 4ss-backwards. Reduce the input to the PA with a resistive divider or attenuator. This will have no perceptible effect on the sound quality.
w
'Flat' and 'lifeless' are just another way of saying 'quieter'. They have no other meaning in this context.
Hi
Just to answer a couple of points from recent posters.
Yes I agree this is the easiest way forward. I would have to do this on all power amps but I guess it's no big deal. Only problem I can see is that when I want to use my MP3 player/vinyl as source - I think I'm going to be lacking sufficient gain. Therefore I don't think this is a satisfactory solution unless I've misunderstood.
I was hoping to use an anode follower say with a gain of x 2 that will then allow me to use both CDP and vinyl (I have a VSPS phono stage) as a source.
Cathode follower is also a possible way forward but with a gain of <1 may not be sufficient for the vinyl. (Output from my MM cartridge is 3mv, gain of VSPS ~ x100 gives 0.3V input into preamp)
I should have made this clear up front I guess!
Thanks for all the replies,
Cheers
Colin
Just to answer a couple of points from recent posters.
Reduce the input to the PA with a resistive divider or attenuator. This will have no perceptible effect on the sound quality.
Yes I agree this is the easiest way forward. I would have to do this on all power amps but I guess it's no big deal. Only problem I can see is that when I want to use my MP3 player/vinyl as source - I think I'm going to be lacking sufficient gain. Therefore I don't think this is a satisfactory solution unless I've misunderstood.
I was hoping to use an anode follower say with a gain of x 2 that will then allow me to use both CDP and vinyl (I have a VSPS phono stage) as a source.
Cathode follower is also a possible way forward but with a gain of <1 may not be sufficient for the vinyl. (Output from my MM cartridge is 3mv, gain of VSPS ~ x100 gives 0.3V input into preamp)
I should have made this clear up front I guess!
Thanks for all the replies,
Cheers
Colin
So what is the problem using a anode follower with your different sources?
You just need to switch the feedback resistor (R2?)to get the gain you need. This is the way Mullard did it with EF86 in the old days.
About resistor divider this is a not a good idea as you will loose dynamics that way.
You just need to switch the feedback resistor (R2?)to get the gain you need. This is the way Mullard did it with EF86 in the old days.
About resistor divider this is a not a good idea as you will loose dynamics that way.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
- Home
- Amplifiers
- Tubes / Valves
- Tube Buffer – question?