I did a fair amount of simulation and a little bit of bench measurement that seemed to point to fixed screen voltage supplies resulting in rather large variations in transconductance tube to tube. The RC circuit I used in the actual design is quite effective at reducing the sensitivity to tube parameters through some level of current feedback in the screen circuit - the gain channel to channel matches very well with only very casually (if at all) matched tubes..
It has mostly been a learning process, and the cap that works well in this location was just a couple of dollars - I just didn't have anything decent in the requisite capacitance range. The capacitor in there now is just an inexpensive metalized polypropylene, ironically had I had some of my usual parts on hand or recognized its significance in the first place I might never really have learned how sensitive this node is to cap quality.. Apparently this is the most sensitive location in the whole design.
Low loss and low inductance. And they're actually just $1.12 in single quantities..
DF 0.0001%
Tolerance 3%
Diameter 12mm
Length 20mm
4um pp film/0.03um aluminum metalization
250VDC rating..
Probably not the ultimate capacitor for this location, but it is the first reasonable one I have heard in this spot. It might take a lot more denaro to better it in this specific application.
DF 0.0001%
Tolerance 3%
Diameter 12mm
Length 20mm
4um pp film/0.03um aluminum metalization
250VDC rating..
Probably not the ultimate capacitor for this location, but it is the first reasonable one I have heard in this spot. It might take a lot more denaro to better it in this specific application.
Maybe the low inductance parameter counts there. It could be some couple more EMI/RFI or resonate a bit there hence the particularly objectionable subjective traits with some.
Attachments
No, I haven't. I tend to buy either Russian caps from vendors in the Ukraine or Bulgaria, or American made film caps from local reps.. I didn't know coffin was in the capacitor business.
Hi,
Other than the classic MKP tone? No.
You can find those caps under all sorts of brand names in Europe as well. Mainly carried by shops that specialise in speakers and accessories.
They were mainly designed for x-over use so the pulse behaviour is extremely important there.
And, not unimportantly, they're very reasonably priced. Especially compared with what some tube shops ask for boutique caps.
@Piano3: Is your preamp SUT input or valve input?
Cheers, 😉
Other than the classic MKP tone? No.
You can find those caps under all sorts of brand names in Europe as well. Mainly carried by shops that specialise in speakers and accessories.
They were mainly designed for x-over use so the pulse behaviour is extremely important there.
And, not unimportantly, they're very reasonably priced. Especially compared with what some tube shops ask for boutique caps.
@Piano3: Is your preamp SUT input or valve input?
Cheers, 😉
Frank, it is valve input; because I need to change stylus constantly depending on the age of the record, only MM is really practical. Again, the differential topology helps by halving the input capacitive load due to the valve. I have several MC cartridges, including a Technics 305 and a spare arm mounting capability on my Garrard 401 so when I have finished the 2nd degree I'm currently studying for in September a MC capable system is near the top of my priorities list. Therefore, I will be asking the same question, valve, semi-conductor or SUT input. I built a MC phono amp (again differential) for my father 4 years ago using 6F12P with the triode section as the input and the pentode section trioded. Combined with a 20x step up xformer from Sowter this provided more than enough gain. I admit though, that it would be fun to at least try to do an MC with valves only
Hi,
Thanks for the reply.
Everyone of them presents its own set of pros and cons. I like to try it with valves as that's where my heart is and...It's a real challenge.
If only the quality/price ratio of current valves wasn't skyrocketing I'd have done it already.
I know the SE MC stage works very well. It's pretty easy to convert it to a fully balanced stage as well. The main problem is how to make it in such a way that it won't overload the MM stage (assuming a standard SE MM phono stage).
From that perspective it's much easier to start from a fresh slate and design a complete balanced MC/MM phono stage, then go unbalanced and buffer out to the line stage level.
BTW, that Russian triode/penthode is a fine candidate for a mu-follower with the penthode section as the current source... Just an idea. 😎
A penthode balanced phono isn't that daft an idea either as most of the penthode's nasties would be cancelled out by the inherent PP nature of the topology. Plenty of ideas here. 😀
Ciao, 😉
Thanks for the reply.
Everyone of them presents its own set of pros and cons. I like to try it with valves as that's where my heart is and...It's a real challenge.
If only the quality/price ratio of current valves wasn't skyrocketing I'd have done it already.
I know the SE MC stage works very well. It's pretty easy to convert it to a fully balanced stage as well. The main problem is how to make it in such a way that it won't overload the MM stage (assuming a standard SE MM phono stage).
From that perspective it's much easier to start from a fresh slate and design a complete balanced MC/MM phono stage, then go unbalanced and buffer out to the line stage level.
BTW, that Russian triode/penthode is a fine candidate for a mu-follower with the penthode section as the current source... Just an idea. 😎
A penthode balanced phono isn't that daft an idea either as most of the penthode's nasties would be cancelled out by the inherent PP nature of the topology. Plenty of ideas here. 😀
Ciao, 😉
Last edited:
Actually Frank, I did draw out a 6F12P mu follower but I didn't actually build it since I was concerned about the heater cathode potentials. I suppose I could have compromised and just set it about 75 volts positive.
modeling help
can someone explain where i am making my mistake. when i run the attached with an .ac oct 256 10 20k i get nothing approaching the riaa curve. thanks for any help
can someone explain where i am making my mistake. when i run the attached with an .ac oct 256 10 20k i get nothing approaching the riaa curve. thanks for any help
he did have offerings many years back.......http://www.diyaudio.com/forums/swap-meet/24554-super-pp-cap-taiwan.html
Oop.....
Still got some but here is not commercial section, so please talk about this somewhere else....
cheers
Coffin
Still got some but here is not commercial section, so please talk about this somewhere else....
cheers
Coffin
He was recommending your caps and that is actually OK even on a thread such as this, now if you were promoting them that would be a different matter.. 😀
Kevin, do not forget Siemens MKV (or Epcos or Siemens+Matsushita) B25834 (radial) or B25839 (axial leads).
I use both types at critical positions. It's hard to beat them.
25839 is pricey (currently on ebay 15 USD), but 25834 can be found for 4-5 USD/pc from German sellers.
If the dimension is not a problem, that would be my choice.
I use both types at critical positions. It's hard to beat them.
25839 is pricey (currently on ebay 15 USD), but 25834 can be found for 4-5 USD/pc from German sellers.
If the dimension is not a problem, that would be my choice.
I have just gone back and had a look at my simulations and I realized that the .step command did not execute correctly, and in fact as I originally stated in simulations at least the circuit is quite sensitive to the value of C7, and 1uF causes a pretty big roll off in the bass. Can't win for loosing I guess, that said I will need to remeasure the RIAA response but I expect to see the same huge error at 20Hz.. I will do this next week..
I tried a number of substitutions in your circuit and added a laplace transform controlled voltage source to provide the inverse RIAA. The tale is told...
Subjectively there is no lack of bass, and it is actually sounding good - so my suspicion is a few identical cheap caps in parallel may do the trick if the simulation results are borne out in practice. (I expect they will be)
Attachments
OK I figured out what I did wrong with the .step param directive in spice and fixed it.. Duh..
Anyway I have attached the corrected capture below and you can see quite clearly the effect of 1uF vs 20uF at C7.
It appears a big honkin' cap is actually necessary if you want that last octave within reasonable bounds of RIAA accuracy. My system has no usable output below 30Hz so I can probably get away with it..
Anyway I have attached the corrected capture below and you can see quite clearly the effect of 1uF vs 20uF at C7.
It appears a big honkin' cap is actually necessary if you want that last octave within reasonable bounds of RIAA accuracy. My system has no usable output below 30Hz so I can probably get away with it..
Attachments
Hi,
In the context of this preamp these are (IMHO) best relegated to the PS right after the rectifier.
Mind you, they're fine coupling caps too but at their voltage rating (900VDC?) they just can't perform to their best in a penthode circuit were voltages are rather low.
To perform at their best caps need to be run close to their voltage rating.
BTW, Epcos was a Siemens company which was sold to TKD(?). Together with Siemens own MKV B258 series these are the only polypropylene in oil motor start caps I know of.
Similar tech was/is used by ASC but I've no idea how the two compare.
The B258 series are becoming hard to come by and well worth it if you need to tame some penthode harshness....
There are a myriad of ways to soften the penthode's signature, the Japanese, French way is to address the PS and do it the old fashioned way using old fashioned components. Which have their virtues albeit it more by omission than by absolute truth.
It all depends on the "chef"' 's taste, right? 🙂
Come to think of it, replacing that zener reference diode in the reg for a gas tube already make a difference too. And no, nobody will hear it's noise.
Then again, maybe I should stick to my own "pots" and "kettles"...😀
Ciao, 😉
In the context of this preamp these are (IMHO) best relegated to the PS right after the rectifier.
Mind you, they're fine coupling caps too but at their voltage rating (900VDC?) they just can't perform to their best in a penthode circuit were voltages are rather low.
To perform at their best caps need to be run close to their voltage rating.
BTW, Epcos was a Siemens company which was sold to TKD(?). Together with Siemens own MKV B258 series these are the only polypropylene in oil motor start caps I know of.
Similar tech was/is used by ASC but I've no idea how the two compare.
The B258 series are becoming hard to come by and well worth it if you need to tame some penthode harshness....
There are a myriad of ways to soften the penthode's signature, the Japanese, French way is to address the PS and do it the old fashioned way using old fashioned components. Which have their virtues albeit it more by omission than by absolute truth.
It all depends on the "chef"' 's taste, right? 🙂
Come to think of it, replacing that zener reference diode in the reg for a gas tube already make a difference too. And no, nobody will hear it's noise.
Then again, maybe I should stick to my own "pots" and "kettles"...😀
Ciao, 😉
Actually I went from gas tube back to zener for a couple of reasons, one was it limited the OLG of the error amplifier and I was throwing away at least 20dB of the available gain by virtue of the dynamic impedance of the 5651 which when measured was around 1K vs 100 ohms for 100V zener. The second issue was noise, a zener can be shunted with meaningful levels of capacitance and large numbers of them can also be placed in series and since they are essentially gaussian noise sources each doubling in number increases noise by 3dB and voltage by 6dB for a net improvement of 3dB. Four in series are actually about 6dB quieter than a single zener..
The cooking analogy is a good one however, and I have talked a lot about this in conjunction with this design. I wonder how well the design would accommodate a 4.7uF Pulse X cap..
The cooking analogy is a good one however, and I have talked a lot about this in conjunction with this design. I wonder how well the design would accommodate a 4.7uF Pulse X cap..
- Status
- Not open for further replies.