I forget how I stumbled across PETP caps, but there is this thread from a while back:
http://www.diyaudio.com/forums/lounge/143675-petp-capacitors-one-best.html
These are still plentiful and reasonably priced on eBay, so I decided to give them a try. I've been tweaking my Bryston B60 integrated and have found that the coupling cap on the preamp output predictably has a strong influence on the sound. The stock part is a 100uF Roederstein NP electrolytic, which is musically pleasant but hardly transparent. I've tried a few different things - WIMA MPK10, leftover Black Gate BG-N, Mundorf eCap Plain. The Mundorf has been the best so far but the search continues.
I ordered (10) K73-16 10uF 63V and (5) K73-17 4.7uF 63V from eBay seller "bird_sr71a" in Ukraine. These are PETP dielectric which I understand is the same as Mylar. The parts came pretty quickly (about a week to CA, USA) and were well packed and in good (NOS) condition as promised.
The mil-spec parts come individually wrapped in wax paper. The physical construction seems very good. The smaller dipped K73-17 isn't "OTK" mil-spec, but they're quite small so I got it as a backup. I can't tell what the lead material is - it's definitely not tinned copper (cross-section is uniform in color) and it seems too stiff to be silver, but it is non-magnetic, so probably tin? There was a little bit of tarnish which I sanded off with 600-grit paper.
I repurposed an old plastic project box to conduct a listening test. The switch allows me to bypass the cap for A/B test, and I can test directionality by reversing the plugs. I placed the cap between the Bryston amp and source - either a Cambridge StreamMagic 6 streamer or Monarchy NM24 tube DAC. Speakers are Merlin TSM-MX with REL T9 subwoofer. Cabling is DH Labs and Cardas.
Initial casual listening test with the Cambridge streamer with no break-in was extremely promising. I was expecting a noticeable degradation off the bat with the insertion of the test box, but it sounded really good! Remarkably transparent when comparing with the straight wire bypass - I had a hard time telling them apart immediately. With a bit more listening, my wife and I both felt the cap was just a little bit "less" of everything - bit less treble extension, bit more constrained dynamics, bit less soundstage depth, bit less fullness in the midrange and bass. But it was all minor sins of omission, with no glaring weakness - in fact, the cap had less character than most interconnects I've tried.
I let the caps break in for a day with some crappy pop internet radio into a 1k ohm load, then tested with real music played on the Monarchy DAC. It opened up and sounded even closer to the straight wire bypass. The previous observations still applied, but it was probably within 95% of the fidelity. I will need to spend some more time with them to ascertain for sure, but I feel more than confident enough to go through the trouble of trying them in the Bryston.
Incidentally, I did briefly listen for directionality. I don't have a lot of patience for such things, but I *think* it may have sounded incrementally fuller and more present with the signal flowing opposite the direction of the label writing. Who knows if the labelling is even consistent in the manufacturing process - I'm certainly not going to test them all individually and I don't have a good way of measuring (aluminum can prohibits inductive measurement) so I'm just going to go with that.
I still need to try out the K73-17 green drops. I may do that before I try installing the larger ones in the Bryston, where it may be a tight fit.
Russian PETP K73 capacitors - Album on Imgur
http://www.diyaudio.com/forums/lounge/143675-petp-capacitors-one-best.html
These are still plentiful and reasonably priced on eBay, so I decided to give them a try. I've been tweaking my Bryston B60 integrated and have found that the coupling cap on the preamp output predictably has a strong influence on the sound. The stock part is a 100uF Roederstein NP electrolytic, which is musically pleasant but hardly transparent. I've tried a few different things - WIMA MPK10, leftover Black Gate BG-N, Mundorf eCap Plain. The Mundorf has been the best so far but the search continues.
I ordered (10) K73-16 10uF 63V and (5) K73-17 4.7uF 63V from eBay seller "bird_sr71a" in Ukraine. These are PETP dielectric which I understand is the same as Mylar. The parts came pretty quickly (about a week to CA, USA) and were well packed and in good (NOS) condition as promised.
The mil-spec parts come individually wrapped in wax paper. The physical construction seems very good. The smaller dipped K73-17 isn't "OTK" mil-spec, but they're quite small so I got it as a backup. I can't tell what the lead material is - it's definitely not tinned copper (cross-section is uniform in color) and it seems too stiff to be silver, but it is non-magnetic, so probably tin? There was a little bit of tarnish which I sanded off with 600-grit paper.
I repurposed an old plastic project box to conduct a listening test. The switch allows me to bypass the cap for A/B test, and I can test directionality by reversing the plugs. I placed the cap between the Bryston amp and source - either a Cambridge StreamMagic 6 streamer or Monarchy NM24 tube DAC. Speakers are Merlin TSM-MX with REL T9 subwoofer. Cabling is DH Labs and Cardas.
Initial casual listening test with the Cambridge streamer with no break-in was extremely promising. I was expecting a noticeable degradation off the bat with the insertion of the test box, but it sounded really good! Remarkably transparent when comparing with the straight wire bypass - I had a hard time telling them apart immediately. With a bit more listening, my wife and I both felt the cap was just a little bit "less" of everything - bit less treble extension, bit more constrained dynamics, bit less soundstage depth, bit less fullness in the midrange and bass. But it was all minor sins of omission, with no glaring weakness - in fact, the cap had less character than most interconnects I've tried.
I let the caps break in for a day with some crappy pop internet radio into a 1k ohm load, then tested with real music played on the Monarchy DAC. It opened up and sounded even closer to the straight wire bypass. The previous observations still applied, but it was probably within 95% of the fidelity. I will need to spend some more time with them to ascertain for sure, but I feel more than confident enough to go through the trouble of trying them in the Bryston.
Incidentally, I did briefly listen for directionality. I don't have a lot of patience for such things, but I *think* it may have sounded incrementally fuller and more present with the signal flowing opposite the direction of the label writing. Who knows if the labelling is even consistent in the manufacturing process - I'm certainly not going to test them all individually and I don't have a good way of measuring (aluminum can prohibits inductive measurement) so I'm just going to go with that.
I still need to try out the K73-17 green drops. I may do that before I try installing the larger ones in the Bryston, where it may be a tight fit.
Russian PETP K73 capacitors - Album on Imgur
Initial listening in Bryston B60
I decided to give the K73-16's a go in the Bryston. Here's what it looked like inside with Mundorf E-Cap Plain 22uF, which sounded very good:
I forgot to take a picture with the PETP caps installed, but it was also a tight fit, and the leads were barely long enough. In retrospect I should have listened to the smaller K73-17's, and installed those if they sounded as good as the 16's.
After a dozen or so hours of break-in, the sound is... good, but not great (yet). On the plus side, it's very smooth and free of grain, with no harshness or glare whatsoever. Compared to the Mundorf though, it seems a bit muddy and confused, e.g. with intricate orchestral passages instruments tend to smush together. The Mundorfs are benefiting from a couple hundred hours of break-in - they did not sound so great either brand new - so I'll give the PETP's at least 100 hours before drawing any conclusions.
At the very least, I can say that the PETP's are very pleasant sounding and excellent for the money (way better than e.g. Solen). But the Mundorfs are also very nice-sounding caps (and electrolytic no less!). I'd recommend giving them a try in solid state gear where you need a largish-value non-polar at reasonable cost and size. I also have the "raw" version which is quite a bit smaller, but haven't tried it yet.
I decided to give the K73-16's a go in the Bryston. Here's what it looked like inside with Mundorf E-Cap Plain 22uF, which sounded very good:
I forgot to take a picture with the PETP caps installed, but it was also a tight fit, and the leads were barely long enough. In retrospect I should have listened to the smaller K73-17's, and installed those if they sounded as good as the 16's.
After a dozen or so hours of break-in, the sound is... good, but not great (yet). On the plus side, it's very smooth and free of grain, with no harshness or glare whatsoever. Compared to the Mundorf though, it seems a bit muddy and confused, e.g. with intricate orchestral passages instruments tend to smush together. The Mundorfs are benefiting from a couple hundred hours of break-in - they did not sound so great either brand new - so I'll give the PETP's at least 100 hours before drawing any conclusions.
At the very least, I can say that the PETP's are very pleasant sounding and excellent for the money (way better than e.g. Solen). But the Mundorfs are also very nice-sounding caps (and electrolytic no less!). I'd recommend giving them a try in solid state gear where you need a largish-value non-polar at reasonable cost and size. I also have the "raw" version which is quite a bit smaller, but haven't tried it yet.
PETP appears to be Mylar (as you say) which is polyester. Given a sufficiently large value these should be transparent as coupling caps, but smaller values (or unusually low load impedances) may give rise to a little distortion so some people prefer to use other more linear dielectrics. Note that the distortion, if present, is likely to be at quite a low level (and so not recognised as distortion) and mainly second-order (which may add some 'warmth' or 'richness' to the sound). Hence if you really can hear a difference this is almost certainly due to the extra distortion produced by these caps. Or it may be the change in capacitance value.
Last edited:
Yeah, the bypass test is just to see if there are any obvious sonic problems with the caps. Once we put it in an actual application, a lot will depend on the circuit tuning, both electrical and sonic. In the case of the Bryston, some of the "character" of the preamp is actually coming from the non-polar electrolytic which is definitely not going to be as linear as a film cap.
This is not an argument or is it 
https://youtu.be/kQFKtI6gn9Y
Marc - ignoring the threat of the capacitor police, I'm interested in hearing your thoughts on these caps. Always on the look out for good sounding, inexpensive film caps. I've used the PETP caps in power supplies but never thought to try them out as coupling caps.
---Gary
https://youtu.be/kQFKtI6gn9Y
Marc - ignoring the threat of the capacitor police, I'm interested in hearing your thoughts on these caps. Always on the look out for good sounding, inexpensive film caps. I've used the PETP caps in power supplies but never thought to try them out as coupling caps.
---Gary
Last edited:
Hi @GaryB - good to see you here again. Sorry I never replied to your PM several years back - I totally missed it and just read it!
Do you still have that Threshold FET 10? If so, I'd definitely recommend trying a K73-16 or 17 as a coupling cap. It's still breaking in but it's sounding pretty good. I've never tried anything super exotic and every circuit is different, but so far it is definitely more neutral than WIMA MKP10 or BG-N. I have a Jantzen Z-Superior in my tube DAC, totally different animal so can't say how they compare but it would be interesting. I find the Jantzen to sound very nice but it seems a bit on the smooth and sweet side... the PETP by contrast has very little character to it, just little sins of omission here and there, and it's opening up more with break-in. I'll give it some more time before listening more seriously.
p.s. I too finally tired of the grind of living in the Northeast and moved to Silicon Valley last year.
Do you still have that Threshold FET 10? If so, I'd definitely recommend trying a K73-16 or 17 as a coupling cap. It's still breaking in but it's sounding pretty good. I've never tried anything super exotic and every circuit is different, but so far it is definitely more neutral than WIMA MKP10 or BG-N. I have a Jantzen Z-Superior in my tube DAC, totally different animal so can't say how they compare but it would be interesting. I find the Jantzen to sound very nice but it seems a bit on the smooth and sweet side... the PETP by contrast has very little character to it, just little sins of omission here and there, and it's opening up more with break-in. I'll give it some more time before listening more seriously.
p.s. I too finally tired of the grind of living in the Northeast and moved to Silicon Valley last year.
No worries - I remember the email but was surprised to see that it was almost 6 years ago when I checked. Who says that email is faster than snail mail!
Yep - still use it in a 2nd system in the house. Wonder whatever happened to Marty and BFS. Do you still hear from him?
p.s. I too finally tired of the grind of living in the Northeast and moved to Silicon Valley last year.
Congrats on the move out west. While there are definitely pros and cons, I suspect it's a good move for your career assuming that you're in tech. I've been out here a decade so I've forgotten all the bad parts of living in NY and find myself getting nostalgic for the place. But it's more a case of wanting to visit a bit more as opposed to move back.
---Gary
p.s. Based on your updated location, I live next town over. Glad to share DIY tips some time.
Unfortunately I haven't heard from Marty in a few years now. He had last mentioned he was dealing with an illness and I'm afraid his health became very poor... beyond that I'm not sure.
Sounds like you're pretty close to me - happy to drop off some of the 10uF PETP's if you'd like to give it a go in the FET 10 or what not.
Sorry to hear about Marty - I always enjoyed reading his magazine.
Thanks for the kind offer on the caps. I actually have a small stash of the 10uf PETP caps that I've been using for PS bypassing, so no need to raid your stock.
Feel free to ping me if you need anything - I've got a decent amount of stuff in my parts bin and some test gear if you're in a pinch.
---Gary
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.