How many of you have used a $50,000 Vector Network analyzer to measure
the impedance of various coupling caps, over a frequency range of 10Hz to 1MHz?
I have.
How many of you have measured the DC Leakage current of various coupling caps,
when it is operating while in the circuit that it will be actually used in?
I have.
No PhD required.
Those old Black Beauty coupling capacitors from the 50s and 60s that were used on amplifiers like the Dyna Stereo 70 . . .
They are more leaky than a Naval Destroyer that has an 18 inch hole in its hull below the waterline.
Replace them, they are good for the Capacitor Museum, and for nothing else.
the impedance of various coupling caps, over a frequency range of 10Hz to 1MHz?
I have.
How many of you have measured the DC Leakage current of various coupling caps,
when it is operating while in the circuit that it will be actually used in?
I have.
No PhD required.
Those old Black Beauty coupling capacitors from the 50s and 60s that were used on amplifiers like the Dyna Stereo 70 . . .
They are more leaky than a Naval Destroyer that has an 18 inch hole in its hull below the waterline.
Replace them, they are good for the Capacitor Museum, and for nothing else.
Last edited:
Nearest I could go Summer is a £10,000 spectrum analyzer --and that was ex government .
No you don't need a PhD just intelligence one of my cousins got a good job because he had a good memory , my memory wasn't as good even though I had a higher IQ.
I agree about those old capacitors especially the paper ones from the 40,s /50,s.
I have a homemade small capacitance leakage tester that applies the correct working voltage to the capacitor and the period of flashes determines how leaky it is --still working since the 60,s .
Cost ?--- $3 to build.
No you don't need a PhD just intelligence one of my cousins got a good job because he had a good memory , my memory wasn't as good even though I had a higher IQ.
I agree about those old capacitors especially the paper ones from the 40,s /50,s.
I have a homemade small capacitance leakage tester that applies the correct working voltage to the capacitor and the period of flashes determines how leaky it is --still working since the 60,s .
Cost ?--- $3 to build.
Auricap XO, easy to source, price under 20$ each, pretty neutral/transparent, detailed, outer foil clearly indicated.
andyjevans,
you seem to follow sound principles. Teflon is a great dielectric, unlike paper, it has superior high frequency properties. You can`t use ceramic, silver mica or tantalum.
Interstage transformers tend to have a large inductance and a low R/L ratio. The less inductance in the signal path the better.
merlin el mago,
you should go for plate resistors. The resistance value needs to be empirically determined, not calculated.
Blaxshep,
PP usually denotes nonmetalized polypropylene.
duncan2,
I have learned that polychlorinated biphenyl (now banned) caused leakage current in motor capacitors (and transformers).
you seem to follow sound principles. Teflon is a great dielectric, unlike paper, it has superior high frequency properties. You can`t use ceramic, silver mica or tantalum.
Interstage transformers tend to have a large inductance and a low R/L ratio. The less inductance in the signal path the better.
merlin el mago,
you should go for plate resistors. The resistance value needs to be empirically determined, not calculated.
Blaxshep,
PP usually denotes nonmetalized polypropylene.
duncan2,
I have learned that polychlorinated biphenyl (now banned) caused leakage current in motor capacitors (and transformers).
Hi guys. If you're in the mood for some more coupling capacitor shootouts, could you give me some opinions on Russian caps or any other caps that are competitively priced, i.e. not boutique prices?
My standard coupling cap is a 0.1uF FT-2 teflon. I'm now looking for a larger value, like 0.47 to 1uF. I could parallel up some of the FT-2 caps, and I've done this, but I'm really looking for some larger values.
I rigged up a basic setup where I could add caps to the outputs of my 10Y driver stage into the 2a3 outputs. So these are added to the existing FT-2 caps. Just convenience. So far I compared no added cap to K40Y 0.22uF, 1000V and to KBG-MN 4uF, 600V. Not surprisingly no cap sounded the cleanest. The K40Y was rather shouty and not very pleasant on vocals, so I would dismiss this as a choice. The KBG-MN despite being much larger than necessary had a very pleasant sound I could live with. Vocals in particular were very natural. I wouldn't say the KBG added warmth - more that it expanded the soundstage which is more focussed with FT-2. Piano was a little softer, smoothed out just a touch. Nothing unpleasant though. These KBG-MN caps are huge. Out of the question for existing equipment, though possible for a new build.
So that's as far as I've got. I need to buy something for a couple of DHT stages I'm building. Any opinions on these:
And any recommendations for good value non-Russian caps?
My standard coupling cap is a 0.1uF FT-2 teflon. I'm now looking for a larger value, like 0.47 to 1uF. I could parallel up some of the FT-2 caps, and I've done this, but I'm really looking for some larger values.
I rigged up a basic setup where I could add caps to the outputs of my 10Y driver stage into the 2a3 outputs. So these are added to the existing FT-2 caps. Just convenience. So far I compared no added cap to K40Y 0.22uF, 1000V and to KBG-MN 4uF, 600V. Not surprisingly no cap sounded the cleanest. The K40Y was rather shouty and not very pleasant on vocals, so I would dismiss this as a choice. The KBG-MN despite being much larger than necessary had a very pleasant sound I could live with. Vocals in particular were very natural. I wouldn't say the KBG added warmth - more that it expanded the soundstage which is more focussed with FT-2. Piano was a little softer, smoothed out just a touch. Nothing unpleasant though. These KBG-MN caps are huge. Out of the question for existing equipment, though possible for a new build.
So that's as far as I've got. I need to buy something for a couple of DHT stages I'm building. Any opinions on these:
- MBGCH
- KBG-MP
- K71-4 polystyrene.
And any recommendations for good value non-Russian caps?
Thanks for the quick reply! Hmm.... I didn't like the K40Y much. It did come after a FT-2 not instead of it. That could make a difference. Have you heard any of the other Russians I mentioned? Or any other caps in general, to get some yardstick for comparisons?
I have listened to the following caps through out the years. They sound consistently in different amps, so to my thinking their sonic signature is amp independent to some degree. Here are my subjective observations in order of preference:
1. Russian K40Y (.22uF, not 0.47uF). Helps will adding some texture, great tonality (violin/cello) and smoothness. Could be too much of a good thing if your amp is already smooth, polite. I use this as a bypass cap in parallel with a WIMA DCLink in a ultrapath position.
2. Duelund CAST Copper. Natural sounding. But, not enough tonality and texture if your system is too aggressive. Won't impact the sound adversely in either direction.
3. Mundor Silver/Gold Oil. Huge soundstage. I personally don't care for the mids/highs tonality. To etchy, not real. BUT, this might the best cap if the amp/system is already smooth or too polite.
I have also had Jupiter Copper Wax, Riken, Wima, others. These did not improve my sound and in one way or another, degraded the sound.
Hope this helps.
1. Russian K40Y (.22uF, not 0.47uF). Helps will adding some texture, great tonality (violin/cello) and smoothness. Could be too much of a good thing if your amp is already smooth, polite. I use this as a bypass cap in parallel with a WIMA DCLink in a ultrapath position.
2. Duelund CAST Copper. Natural sounding. But, not enough tonality and texture if your system is too aggressive. Won't impact the sound adversely in either direction.
3. Mundor Silver/Gold Oil. Huge soundstage. I personally don't care for the mids/highs tonality. To etchy, not real. BUT, this might the best cap if the amp/system is already smooth or too polite.
I have also had Jupiter Copper Wax, Riken, Wima, others. These did not improve my sound and in one way or another, degraded the sound.
Hope this helps.
We seem to have similar preferences. No matter how long or by what method they're broken in to my ear Russian PIOs stay slush bombs. Good for masking ugliness in legacy rebuilds when you don't want to redesign the circuit maybe.
Under 200 volts I find the Russian FT-1 is terrific. Maximum value is 0.022 uF which works very well as a bypass for caps in the normal coupling range. My current test mule employs a single 0.022 uF into a 1 meg grid resistor. Completely halted the coupling cap neurosis merry-go-round for now. Caveats, that's with ferrites between the grid and grid resistor instead of resistor grid stoppers. The logic behind selecting a cap with 0.0001 ohm series internal resistance to series with 1000/10,000 ohms never made sense to me. Also the FT-1 case (all non-magnetic) floats and at least in this circuit builds up a charge. For now I'm just wrapping one high value resistor lead around it with the other end grounded.
RDF: A bit off topic - I dont use any grid stopper or grid leak resistors bteween the RCA input and Driver tube. And, I don't hear any issues. What am I missing? I just don't see/hear the purpose behind the grid leak or stopper resistors? Feel free to take this offline if you see appropriate.
If the circuit is stable and not oscillating at ultrasonic frequencies you're not missing anything. To my mind less is better. In my case the input stage is high gain with about 25 dB of loop feedback so just playing it safe. The output stage doesn't use any stoppers.
The only side case coming to mind is if upstream devices are spewing ultrasonic junk. I ran into this with a Topping DX3 Pro+ into a gainclone. Audible buzzing at the listening chair was only remedied by replacing the chipamp resistor input stoppers with ferrite beads.
The only side case coming to mind is if upstream devices are spewing ultrasonic junk. I ran into this with a Topping DX3 Pro+ into a gainclone. Audible buzzing at the listening chair was only remedied by replacing the chipamp resistor input stoppers with ferrite beads.
These sound great and you can't beat the price.
https://industrial.panasonic.com/cdbs/www-data/pdf/RDI0000/ABD0000C183.pdf
https://industrial.panasonic.com/cdbs/www-data/pdf/RDI0000/ABD0000C183.pdf
banpuku: "i don't use any grid stopper or grid leak resistors" - between amp input and tube grid
You are mixing two very different things: grid stopper is a resistor on tube socket to the tube input grid - to prevent oscillation on high to medium transconductance (gm) tubes. Your Aa input tube on your SE GM70 power amp is a low gm tube. (1mA/V) No grid stopper needed.
grid leak resistor: a resistor from the tube input grid to ground. There has to be a dc path to ground. On your circuit its the 100K on the Aa tube. If you remove it, you would destroy the Aa.
You are mixing two very different things: grid stopper is a resistor on tube socket to the tube input grid - to prevent oscillation on high to medium transconductance (gm) tubes. Your Aa input tube on your SE GM70 power amp is a low gm tube. (1mA/V) No grid stopper needed.
grid leak resistor: a resistor from the tube input grid to ground. There has to be a dc path to ground. On your circuit its the 100K on the Aa tube. If you remove it, you would destroy the Aa.
Lauscher - many thanks for the education on the 2 resistor types. I pulled the 100K grid leak resistor a couple of weeks ago and everything is fine (for now at least). Should I put it back in place and if so, how do I determine the minimum grid leak resistor value? I have a nice 17K resistor I would like to try, but not sure if its large enough resistance. Any insight would be appreciated.
As said, you need a resistance on the grid to ground for your circuit to work. If you took out 100K and the stage still worked, then you must be getting the resistance from the previous stage. It has to come from somewhere. If you have a resistance on the output of the previous stage, then this will be in parallel with the grid leak you insert. But to know what value is on the output of the previous stage you'd have to measure it with a multimeter. You'd probably be shooting for 50K to 100K total input resistance as a rule. But if you have no grid leak at all, your Aa stage would go nuts, so beware if you disconnect the preceding stage.
OK, so riddle me this Andy and Lauscher.
Resistance for DAC: 660ohm (measured at RCA jacks)
Resistance for elmaformer volume control set at unity (0db): 57 ohms
Resistance at Grid vs. Ground of power amp driver tube (with 26K grid leak resistor): 1.3ohms
When there is no grid leak resistor on the power amp drive tube, the sound is very open, dynamic and pleasing. With a 26K grid leak resistor (z-foil naked), the sound is muted, loss of air, loss of dynamics. Thoughts?
Resistance for DAC: 660ohm (measured at RCA jacks)
Resistance for elmaformer volume control set at unity (0db): 57 ohms
Resistance at Grid vs. Ground of power amp driver tube (with 26K grid leak resistor): 1.3ohms
When there is no grid leak resistor on the power amp drive tube, the sound is very open, dynamic and pleasing. With a 26K grid leak resistor (z-foil naked), the sound is muted, loss of air, loss of dynamics. Thoughts?
Some articles discussing capacitors attached. Also a power point by Bruce Hofer of Audio Precision at: https://www.dropbox.com/s/7363piqyu...- Bruce Hofer - Audio Precision - AP.ppt?dl=0