Que tal, Celsius. Miller Effect increases with gain, so I think a CCS load (usually providing higher gain) will probably lower the frequency.
Cmiller = Cg-k + Cg-p * (stage gain + 1)
f = 1 / (2 * pi * Cmiller * Z)
Z here would be the output impedance of the previous stage. Maybe if you buffer the input (and lower Z) you can avoid the roll off.
Cmiller = Cg-k + Cg-p * (stage gain + 1)
f = 1 / (2 * pi * Cmiller * Z)
Z here would be the output impedance of the previous stage. Maybe if you buffer the input (and lower Z) you can avoid the roll off.
Bear in mind that pentodes also generate partition noise and that the screen supply in that first stage is pretty critical or sound quality may be adversely affected.
I ended up with cascodes using 6S3P and 6N23P/ECC88 instead, no partition noise and no problem with screen supply quality.
I designed a pentode based phono stage around the 6ж9п (6Z9P/6J9P)details are in this thread:
http://www.diyaudio.com/forums/analogue-source/246756-muscovite-mini-6-9-6z9p-phono-stage.html
The next evolution here:
http://www.diyaudio.com/forums/analogue-source/254641-muscovite-mini-ii-6n14p-phono-stage.html
And the final form here:
http://www.diyaudio.com/forums/analogue-source/273817-muscovite-mini-iii-6n23p-phono-stage.html
You might have greater success with pentodes than I did, but I never got it to sound right even with D3A which in pentode connection performed very similarly to the 6ж9п which incidentally is not a bad tube.
I ended up with cascodes using 6S3P and 6N23P/ECC88 instead, no partition noise and no problem with screen supply quality.
I designed a pentode based phono stage around the 6ж9п (6Z9P/6J9P)details are in this thread:
http://www.diyaudio.com/forums/analogue-source/246756-muscovite-mini-6-9-6z9p-phono-stage.html
The next evolution here:
http://www.diyaudio.com/forums/analogue-source/254641-muscovite-mini-ii-6n14p-phono-stage.html
And the final form here:
http://www.diyaudio.com/forums/analogue-source/273817-muscovite-mini-iii-6n23p-phono-stage.html
You might have greater success with pentodes than I did, but I never got it to sound right even with D3A which in pentode connection performed very similarly to the 6ж9п which incidentally is not a bad tube.
FWIW, I too have been intrigued by the idea of a 6AC7 in the 1st gain block of a phono preamp. Clearly, regulated g2 B+ is desirable. I suspect a snubbed 0D3/VR150 would be adequate in feeding both channels.
The issue of partition noise will not go away. OTOH, the high gm provides hope that S/N performance will be, at least, on a par with that obtained from a 6SL7 or 12AX7 section. Adverse Miller capacitance interaction with some MM carts. or SUTs is definitely a non-issue.
Obviously, heed the data sheet and use DC heating.
The issue of partition noise will not go away. OTOH, the high gm provides hope that S/N performance will be, at least, on a par with that obtained from a 6SL7 or 12AX7 section. Adverse Miller capacitance interaction with some MM carts. or SUTs is definitely a non-issue.
Obviously, heed the data sheet and use DC heating.
Thanks Hego.
Kevinkr, I have seem you room in your links, small speakers ha ha. Well is difficult to know that is the best solutions to the fist stage.
1 medium gain ecc88 or pentode strapped as triode ( d3a, e280f )
2 Cascode as you have now or hybrid fet+triode
3 Pentode pure
The 6sl7 to have a flat response need a load of 50k with 40gain, 40 x 40-10 ( RIAA) = 1200 maybe enough but need to cathode follower. Kevinkr thanks for you schematic very interesting
Kevinkr, I have seem you room in your links, small speakers ha ha. Well is difficult to know that is the best solutions to the fist stage.
1 medium gain ecc88 or pentode strapped as triode ( d3a, e280f )
2 Cascode as you have now or hybrid fet+triode
3 Pentode pure
The 6sl7 to have a flat response need a load of 50k with 40gain, 40 x 40-10 ( RIAA) = 1200 maybe enough but need to cathode follower. Kevinkr thanks for you schematic very interesting
Last edited:
I dont measure with 50k, only I calculated on formula for 6sl7 Cmiller I use 177pf 3,2+60+1*2,8 gain is the true on load.
About the grid have 47k to ground
CCS will slightly increase miller, but its not the roll-off of higher frequencies that is the problem.
The problem is if the phono stage is to be used for an MM cartridge. You did not say if the stage is for MM or MC.
If you are building for MM then I hate to tell you this - you are probably sunk.
For most MM cartridges, you want to get the total input capacitance (including miller, cables, wire, socket, etc.) below something like 200pF. Some MM carts like to see as little as 180pF, and your cable might alone be 120pF - that does leave you much room, does it?
See the problem? so.. is it for MM or MC?
If you are building for MM, you will likely get way more simbilance than you will desire. It will sound crap and you will blame the MM cart. However MM carts are FANTASTIC when correctly loaded... 99.99% of the time they are NOT.
There are solutions though. Look to the work of Allen Wright (r.i.p) for some inspiration.
Of course if you are doing MC then there is a ton of other problems to solve....
Ian
The problem is if the phono stage is to be used for an MM cartridge. You did not say if the stage is for MM or MC.
If you are building for MM then I hate to tell you this - you are probably sunk.
For most MM cartridges, you want to get the total input capacitance (including miller, cables, wire, socket, etc.) below something like 200pF. Some MM carts like to see as little as 180pF, and your cable might alone be 120pF - that does leave you much room, does it?
See the problem? so.. is it for MM or MC?
If you are building for MM, you will likely get way more simbilance than you will desire. It will sound crap and you will blame the MM cart. However MM carts are FANTASTIC when correctly loaded... 99.99% of the time they are NOT.
There are solutions though. Look to the work of Allen Wright (r.i.p) for some inspiration.
Of course if you are doing MC then there is a ton of other problems to solve....
Ian
Last edited:
No, I was not speaking about the grid stopper resistor. Two reasons for this:SemperFi said:
1. when I and others say 'grid resistor' we mean 'grid resistor' (also known as 'grid leak resistor') not 'grid stopper resistor'.
2. 47k would be rather high for a grid stopper (except in a guitar amp), but could be a grid resistor for an MM input. Celsius specififcally said he had 47k to ground.
This is exactly what I was thinking.
If that's the case, forget HF roll-off.... that's not going to be the problem.
This is my first experience with a phono preamp, before I never need to study the miller effect, in my two lines preamp I use 6e5p as triode and 6sn7, when you see the square wave at 20khz is perfect but when I begun to play with high gain 6sl7 the square wave is not square and I begin to study the Miller effect.
My first step is to know what is the optimum load resistor without roll off, because the 6sl7 is a very linear tube, Later I will decide if use MM 47k o MC 220R, I have seen that with 220R grid to ground the roll off is less and the rest of frequency is flat, my idea is 6sl7 x 6sl7 x 6e5p ( ccs low Z), this weekend I begin to test with 40k at load.
The 6sl7 with ccs have a big roll off, I didn t know that a phono preamp is not easier as seem.
My first step is to know what is the optimum load resistor without roll off, because the 6sl7 is a very linear tube, Later I will decide if use MM 47k o MC 220R, I have seen that with 220R grid to ground the roll off is less and the rest of frequency is flat, my idea is 6sl7 x 6sl7 x 6e5p ( ccs low Z), this weekend I begin to test with 40k at load.
The 6sl7 with ccs have a big roll off, I didn t know that a phono preamp is not easier as seem.
As have been said by Kevin (trough his circuits) and Ian, for MM input your choice must be a cascode, it lowers input capacitance about one order of magnitude.
See here for detailed calculations
How to Gain Gain - A Reference Book on Triodes in Audio | Burkhard Vogel | Springer
I assume that you do not want a big capacitor on the input, so grounded grid is out of discussion.
A phono stage is always difficult to implement, the 6SL7 seems too noisy for MC. IMHO.
See here for detailed calculations
How to Gain Gain - A Reference Book on Triodes in Audio | Burkhard Vogel | Springer
I assume that you do not want a big capacitor on the input, so grounded grid is out of discussion.
A phono stage is always difficult to implement, the 6SL7 seems too noisy for MC. IMHO.
Last edited:
Many do use an input cap to limit LF-woble (or what is it called?) from warped records.
I've been using battery bias on my preamp tubes and justified the input cap by the many warped records I have. If it hadn't been for that I'd probably not use bat-bias on first tube.
If I decide use MC ( Benz Micro Gold ) what will be the steps to design the preamp without step-up transformer.
6sl7 is not mandatory I have many e280f e180f e810f 6e5p 6ac7
My preamps have a gain 24db ( 6sn7) and 29db ( 6e5p)
Benz Micro Gold
- Output: .5mV
- Stylus: .3x.7mil
- Impedance: 20 ohms
- Loading range: >200 ohms
- Weight: 5.7g
- Compliance: 15cu
- Tracking force: 1.7-1.9g
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.