Modifying 6H30 Mu-follower line-stage.
I have a mu-follower line stage that (unfortianly) is single ended. I have been thinking about buying a Pass Labs X250 power-amp, which i was going to run together with this tube-pre.
The pass labs does only have 11k input impedance, so i thought i would try to lower the output impedance on the linestage i have today.
I'm quite new to modifying / building stuff myself so i have tried to understand how everything works to get lower Z-out, and better linearity of the tubes.
Today the amplifier looks like this:
http://img707.imageshack.us/img707/4094/6h30orig.jpg
The tubes are running at ~60V and the bias is -4.2v (~3mA)
And i wouldn't want to modify/Change the PSU, so my question is. would it be possible to change the cathode resistors and the "middle" resistor to the following values?
http://img820.imageshack.us/img820/2842/6h30mod.jpg
This would make the tubes run at 75V and biased at approx -3.7v (~9mA)
And if i have understand it right, the tube curves - Tube Tester Files - 6N30P - 6H30Ï
If running at around 8mV the Rp(Ra?) would be around 1.8K instead of approx 3.4K when running 3mA.
and as i have understand it, the Ra/Ra is directly related to output impedance, and this would make Z-out ~123ohm?
What would be negativ by doing this?!
Any help will be appreciated, since i'm really a noob on this. thanks!
I have a mu-follower line stage that (unfortianly) is single ended. I have been thinking about buying a Pass Labs X250 power-amp, which i was going to run together with this tube-pre.
The pass labs does only have 11k input impedance, so i thought i would try to lower the output impedance on the linestage i have today.
I'm quite new to modifying / building stuff myself so i have tried to understand how everything works to get lower Z-out, and better linearity of the tubes.
Today the amplifier looks like this:
http://img707.imageshack.us/img707/4094/6h30orig.jpg
The tubes are running at ~60V and the bias is -4.2v (~3mA)
And i wouldn't want to modify/Change the PSU, so my question is. would it be possible to change the cathode resistors and the "middle" resistor to the following values?
http://img820.imageshack.us/img820/2842/6h30mod.jpg
This would make the tubes run at 75V and biased at approx -3.7v (~9mA)
And if i have understand it right, the tube curves - Tube Tester Files - 6N30P - 6H30Ï
If running at around 8mV the Rp(Ra?) would be around 1.8K instead of approx 3.4K when running 3mA.
and as i have understand it, the Ra/Ra is directly related to output impedance, and this would make Z-out ~123ohm?
What would be negativ by doing this?!
Any help will be appreciated, since i'm really a noob on this. thanks!
Hi. Don't really have time at the moment for anything deeper but here are generally my thoughts.
First i really do not understand why someone will design a power amp with 11k input impedance. Could it be just to make most tube preamps miserable driving it?
Second, you may be slightly confused with the importance of output impedance. While desirable, it is by no means sufficient. However, you are on the right path in raising quiescent current - it will definitely help driving low resistance loads. And if any tube can do it, a 6H30 certainly can.
And finally, running a 6H30 at 3mA is really a waste. Even without challenging loads it only starts sounding alive above 10-12mA.
So, yes, raising the current is the best you can possibly do in the situation.
First i really do not understand why someone will design a power amp with 11k input impedance. Could it be just to make most tube preamps miserable driving it?
Second, you may be slightly confused with the importance of output impedance. While desirable, it is by no means sufficient. However, you are on the right path in raising quiescent current - it will definitely help driving low resistance loads. And if any tube can do it, a 6H30 certainly can.
And finally, running a 6H30 at 3mA is really a waste. Even without challenging loads it only starts sounding alive above 10-12mA.
So, yes, raising the current is the best you can possibly do in the situation.
Tried that and it looks pretty good with R3=2k.
Approx 84V over the tubes and -4.4v bias, around 12mA.
Is it best to run there or would it be better with R4=R6=220 Ohm, R3=1,5k -4.2v bias at approx 17mA? This considering it will drive such low-impedance poweramp as the pass labs..
Also output impedance on this Z-out=Ra/Mu? R=U/I (R=87/0.017) = 5111Ohm
then Z-out is 5111/15=340 ohm. Have i calculated this right?
Thanks alot for your help so far!
Approx 84V over the tubes and -4.4v bias, around 12mA.
Is it best to run there or would it be better with R4=R6=220 Ohm, R3=1,5k -4.2v bias at approx 17mA? This considering it will drive such low-impedance poweramp as the pass labs..
Also output impedance on this Z-out=Ra/Mu? R=U/I (R=87/0.017) = 5111Ohm
then Z-out is 5111/15=340 ohm. Have i calculated this right?
Thanks alot for your help so far!
Okay, seems like i have understand everthing wrong with output impedance (Rp) how is it possible to calculate everything around that? is there any formula for that?
Because i see at Tube Tester Files - 6N30P - 6H30Ï that when the tubes are running around 20mA the impedance seem to vary around 1.1-1.3K, so how can i calculate it at my voltage/current/bias?
Because i see at Tube Tester Files - 6N30P - 6H30Ï that when the tubes are running around 20mA the impedance seem to vary around 1.1-1.3K, so how can i calculate it at my voltage/current/bias?
You have a couple of compromises going on here, due mainly to the relatively low power supply voltage for a mu follower. The low distortion characteristics of the mu follower are due largely to the middle resistor being multiplied by the mu of the top tube thus providing a very high effective ra.
The trouble is, as soon as you raise the current, the voltage drop across this resistor increases and you are tempted to reduce its value, as indeed has already been suggested, and you really no longer have a mu follower. The distortion increases and the gain drops as the linearising effect is lost. The normal way to increase the current in a mu follower is to use a higher supply voltage but this is not an option for you.
I have two suggestions:
Do nothing. The output impedance is already low enough to drive an 11K load; the real question is how hard do you want to drive it? Contrary to popular belief, the mu follower is not single ended and with the existing 3mA idle current it can in theory drive 6mA peak to peak into a load which equates to about 20V rms into 11K. I am guessing your power amp needs 2V rms at the most, so this should be quite within the limits of the current circuit.
Alternatively, if you really want to increase the current significantly then change the topology to an SRPP and do without the middle resistor altogether. You will get higher distortion and the stage gain will be about half that of the mu follower but you are moving in that direction anyway by reducing the middle resistor.
Cheers
Ian
The trouble is, as soon as you raise the current, the voltage drop across this resistor increases and you are tempted to reduce its value, as indeed has already been suggested, and you really no longer have a mu follower. The distortion increases and the gain drops as the linearising effect is lost. The normal way to increase the current in a mu follower is to use a higher supply voltage but this is not an option for you.
I have two suggestions:
Do nothing. The output impedance is already low enough to drive an 11K load; the real question is how hard do you want to drive it? Contrary to popular belief, the mu follower is not single ended and with the existing 3mA idle current it can in theory drive 6mA peak to peak into a load which equates to about 20V rms into 11K. I am guessing your power amp needs 2V rms at the most, so this should be quite within the limits of the current circuit.
Alternatively, if you really want to increase the current significantly then change the topology to an SRPP and do without the middle resistor altogether. You will get higher distortion and the stage gain will be about half that of the mu follower but you are moving in that direction anyway by reducing the middle resistor.
Cheers
Ian
Don't quite agree with that one.
When gain approaches mu because of the high impedance load, the stage will be single ended.
The SRPP functions as a push pull stage.
Everything else in between will function somewhere between SE and PP.
When raising high voltage supply is no option R3 might be replaced by a solid state
eek current source (DN2540 for instance). That should work with low loss of voltage across the CS. That will give a pure mu follower.Most important I think is to see what the stage must do in terms of gain and output impedance.
Thanks alot for your answer Ian!
I wouldn't it be ok to run the Cathode resistors at ~220ohm and then drop the middle resistor to about 1k8 - 2k2?
I found another interesting thread with distortion figures about this (although with 12ax7) which seemed to favor lower-value middle resistor, at least when running without the lower valve cathode bypass, and for 2nd harmonics distortion. 3rd order was worse, but isn't the 2nd order more important to keep low?
The tread can be found here - http://www.diyaudio.com/forums/tubes-valves/54871-distortion-analysis-mu-follower.html
I wouldn't it be ok to run the Cathode resistors at ~220ohm and then drop the middle resistor to about 1k8 - 2k2?
I found another interesting thread with distortion figures about this (although with 12ax7) which seemed to favor lower-value middle resistor, at least when running without the lower valve cathode bypass, and for 2nd harmonics distortion. 3rd order was worse, but isn't the 2nd order more important to keep low?
The tread can be found here - http://www.diyaudio.com/forums/tubes-valves/54871-distortion-analysis-mu-follower.html
Excpet the load is an active one and the stage can both sink and source current as witnessed my measuring its load driving capability.
Cheers
Ian
If that figure is 2.5V rms then it will certainly overload the tube and lead to grid current distortion. Not only that but that stage gain is about 30 times which means the output would be over 60V rms which I am sure is not what you intend. You really need a pot between the DAC and the mu follower stage.
Regarding the top CF stage you can bias it the same. Because of the near 100% negative feedback, the grid to cathode signal remains small.
Cheers
Ian
Ian,
I don't understand where your figures come from.
2.5 VRMS is 7 volts pk-pk.
-3.7 volt bias will allow 7.4 pk-pk before overload / grid current; no problem with the 2.5 VRMS.
Also don't understand your "stage gain of 30".
When the 6H30 stage is working as a pure mu-follower the gain will never be more than the mu, which is about 15 for the tube in question. The top tube is a unity gain stage at best.
I don't understand where your figures come from.
2.5 VRMS is 7 volts pk-pk.
-3.7 volt bias will allow 7.4 pk-pk before overload / grid current; no problem with the 2.5 VRMS.
Also don't understand your "stage gain of 30".
When the 6H30 stage is working as a pure mu-follower the gain will never be more than the mu, which is about 15 for the tube in question. The top tube is a unity gain stage at best.
Grid current usually begins to occur when the grid reaches about -0.5V and often before that. I usually design this sort of stage so that the signal never pushes the grid more positive than -1.0V. This means the peak input signal voltage must be kept below 3.7 - 1.0 = 2.7V for grid current not to occur. That's just under 2V rms.
Quite right. Don't know why but I had it in my head that this tube had a mu of over 30. Even so, a gain of 15 will give an output of as much as 37V rms which I am sure is much more than the OP intends.
Cheers
Ian
Yes you are right.
The Pass Labs amp needs about 1.5 VRMS for full output.
Maybe the OP likes to control volume in the 7 to 9 o'clock range.
When the tube stage is meant for sound reasons there are certainly better options.
I Know that this line stage have to much gain for the pass, but since i already own this pre-amp (consonance reference 50) which i don't want to completely scrap (in that case rather sell later on). But i found a great price for the Pass poweramp, and i thought i would make the best out this line-stage to match it.
I will probably build an attenuator with a 6h30 buffer and just input selectors later on, or maybe a low-gain tube stage since i have pretty much enough gain for everything today (maybe an exception for the vinyl).
I am confused. Are you saying the preamp circuit we have been discussing is the Consonance Reference 50? If it is, does that not already have a volume control at the input so the DAC signal level is not really an issue? And presumably you indulged in a little tube rolling by changing the supplied ECC82 for a 6H30?
Or have I got this completely wrong?
Cheers
Ian
Hello, and yes this is the consonance ref 50, so it does have a volume before the gain stage (so really signal well pretty much NEVER be 100%).
I didn't think about that when i wrote the above replay.
Anyway, this is an older (Ref50 Mk1, not MK2) and those did have 6h30 instead of ECC82 tubes.
I'm right now looking if there is any way i can convert this amp to a normal lower gain-stage with cathode follower, and perfered to have the volume between gain stage and cathode follower. but i'm not shure how i will do yet.
22dB (as i have right now) is way to much gain for my liking and i start to look if it is possible to modify the amp so i'll get a completley new circuit, or if i'll just build a new one separate. so i'll see what i do...
Last edited:
If you change it to an SRPP , (which is a straightforward modification) you will effectively halve the gain.
Cheers
Ian
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.