Member
Joined 2003
Hi,
I am currently working on my second amplifier project, consisting of Twisted Pear Audio kits: Mauro's My_Ref rev.C and the kookaburra. The amp and preamp is being built with the parts kit sold by Twisted Pear Audio, apart from the 1W resistors have been replaced with 2W 2% NTE "flameproof" resistors.
I have searched through various threads, and it seems as though there are a lot of different preferences to coupling capacitor choice. The My_Ref kit comes with Wima MKS4 1.0uF 63V coupling caps, but I also have some Solen MKP-FC 2.2uF 250V caps I could use.
Not taking the size difference into account, since I'm sure it has a slight effect on sound, what is your preference? A polyester or a polypropelyne coupling cap? I will be testing both types once the amp is complete, I am just wondering what to expect.
Thanks.
I am currently working on my second amplifier project, consisting of Twisted Pear Audio kits: Mauro's My_Ref rev.C and the kookaburra. The amp and preamp is being built with the parts kit sold by Twisted Pear Audio, apart from the 1W resistors have been replaced with 2W 2% NTE "flameproof" resistors.
I have searched through various threads, and it seems as though there are a lot of different preferences to coupling capacitor choice. The My_Ref kit comes with Wima MKS4 1.0uF 63V coupling caps, but I also have some Solen MKP-FC 2.2uF 250V caps I could use.
Not taking the size difference into account, since I'm sure it has a slight effect on sound, what is your preference? A polyester or a polypropelyne coupling cap? I will be testing both types once the amp is complete, I am just wondering what to expect.
Thanks.
Member
Joined 2003
Thanks for your comments. I think I shall use the Solen caps, and maybe start shopping for this "Vitamin Q" type capacitor.
Member
Joined 2003
After using the Solen coupling caps for a month, and very much enjoying the sound, I tried bypassing them with Vitamin Q capacitors. Solen 2.2uF 250V MKP "fast cap" were bypassed with 0.1uF Sprague 300VDC Vitamin Q capacitors.
After listening with these caps installed for over a week now, I don't like them. They've taken some of the "life" out of the amp. For those who like a warmer, smoother sound might like these caps, but for me I feel that the detail or the "life" has been removed. I know it is hard to describe a sound in words, but my verdict is that I am sticking with the Solen caps and am not going to mess around with it anymore.
After listening with these caps installed for over a week now, I don't like them. They've taken some of the "life" out of the amp. For those who like a warmer, smoother sound might like these caps, but for me I feel that the detail or the "life" has been removed. I know it is hard to describe a sound in words, but my verdict is that I am sticking with the Solen caps and am not going to mess around with it anymore.
Hahhhhaaa! The Solen's are great, aren't they! So transparent, crisp, pacey... "fast" indeed! And with very good tonal transparency. So much for anyone who tried to knock 'em down. 😀
BTW, I'm using the same value of 2.2uF 250V. Awesome coupling caps, and impressively better than the Wima MKS2.
Member
Joined 2003
This sure is an old thread! I generally just stay away from the cap and cable discussions these days. But since we're here...for amplifier coupling I've moved away from the Solen caps to Epcos MKP. Check PN 495-2940-ND at Digikey (I'm fairly certain this is what I've purchased previously). They are inexpensive and very good. Recommended.
I really just used the Solen caps because that's what I had on hand at the time (left overs from crossover build). They do the job quite well anyway, but I really like the Epcos.
I really just used the Solen caps because that's what I had on hand at the time (left overs from crossover build). They do the job quite well anyway, but I really like the Epcos.
Do you mean this, BTW?
Epcos | Passivi | Condensatori | Condensatori film plastico | Radiale 100°C |B32654A3225K
Epcos | Passivi | Condensatori | Condensatori film plastico | Radiale 100°C |B32654A3225K
Last edited:
Member
Joined 2003
I'd have to look back at my purchase orders to know exactly what I have, but yes they're the blue box Epcos MKP. I seem to remember them being rated as "high pulse strength" or something to that effect.
Ok, good to know. I'll try them when I have the occasion. However, the axial Solen's are a much better fit to my DAC's tiny enclosure.
Also, of the Solen I like the audiophile-ish (tinned) OFC copper leads.
Also, of the Solen I like the audiophile-ish (tinned) OFC copper leads.
My newbie search finding this old thread, I thank those who came before me.
Specifically, I have 5 large commercial amps with decent FET output stages, which I want to improve:
The input to the driver stage (output from the XLR balanced input stage) has an inter-stage coupling cap labelled on the schematic as NP 25/25 (sheet says to assmume microfarad) which I assume is internally two 25 microfarad electrolytic caps back-to-back to make a non-polar electrolytic. This coupling cap is in series with a 2.2K resistor, then a 470 pf cap to ground then 16K resistor to ground. It looks like a good candidate for an upgrade.
Having read
http://www.ecp.cc/cap-notes.html
F = 1/(2 * pi * C * R)
Approximanting, 1/(2 * pi * .000025 * 16000) = .4 hz as the 3db down frequency, which I assume is just a big cap and passes infrasonic bass and keeps the phase changes subsonic, mostly because large-value non-polar electorlytics are incredibly cheap.
That site gives a rule-of-thumb value of 4.7uf for a 10K input, or 1uf for a 50K input. So I'd probably get away with a high-quality 4.7 microfarad. Power rails are + and - 87 volts, so voltage ratings are not a problem.
So if I were a cheap B**tard I might bypass the existing non-polar caps with something better, but instead should replace them with a quality cap of at least 4.7 microfarads.
Now the hard part is choosing specifics...anything new in the last 5 years I should know about (like current sales of preferred sources??).
Specifically, I have 5 large commercial amps with decent FET output stages, which I want to improve:
The input to the driver stage (output from the XLR balanced input stage) has an inter-stage coupling cap labelled on the schematic as NP 25/25 (sheet says to assmume microfarad) which I assume is internally two 25 microfarad electrolytic caps back-to-back to make a non-polar electrolytic. This coupling cap is in series with a 2.2K resistor, then a 470 pf cap to ground then 16K resistor to ground. It looks like a good candidate for an upgrade.
Having read
http://www.ecp.cc/cap-notes.html
F = 1/(2 * pi * C * R)
Approximanting, 1/(2 * pi * .000025 * 16000) = .4 hz as the 3db down frequency, which I assume is just a big cap and passes infrasonic bass and keeps the phase changes subsonic, mostly because large-value non-polar electorlytics are incredibly cheap.
That site gives a rule-of-thumb value of 4.7uf for a 10K input, or 1uf for a 50K input. So I'd probably get away with a high-quality 4.7 microfarad. Power rails are + and - 87 volts, so voltage ratings are not a problem.
So if I were a cheap B**tard I might bypass the existing non-polar caps with something better, but instead should replace them with a quality cap of at least 4.7 microfarads.
Now the hard part is choosing specifics...anything new in the last 5 years I should know about (like current sales of preferred sources??).
The 25/25 probably means 25 uF at up to 25 Volts.
But a non-polar 25 uF would be made by placing two 50 uF caps in series (not two 25 uF), back to back, usually with a large resistance in parallel with each cap.
Yes, it looks like the high-pass cutoff frequency would be 0.4 Hz. But what about the low-pass cutoff frequency, for the 2200 Ohms with 470 pF to ground? That looks like it gives about 154 kHz. I might replace the 470 pF with 330 pF, to raise that to 219 kHz.
Anyway, replace the 25 uF with polypropylene. How cheap are you? A 4.7 uF with the 16k would give a high-pass cutoff of about 2.12 Hz. But lower would be better.
Madisound has a 10 uF 100V polypropylene for $0.60. Or, if you can stand to spend a few dollars per cap, you might be able to get better ones.
POLYPROPYLENE CAPACITORS from Parts Express ship same day and come with 45 day money back guarantee. Free Shipping Available. Order free 10,000 product catalog.
https://www.madisound.com/store/adv..._in_description=1&inc_subcat=1&page=1&sort=4a
Mouser Electronics - Electronic Component Distributor
Electronic Components Distributor | DigiKey Corp. | US Home Page
With film caps, one main thing to look for is the construction. One way they could make them would be to attach the leads to the ends of two long strips of thin foil with a thin strip of PP under each strip of foil, then roll those up together. That way could result in a lot of inductance. The better way is to let the foil extend out past the edge of the PP all the way along one side (with the two strips hanging out on opposite sides), and then rolling them. Then end plates can contact one entire edge of each foil strip and the leads can be attached to the endplates. That way you don't get much inductance.
But a non-polar 25 uF would be made by placing two 50 uF caps in series (not two 25 uF), back to back, usually with a large resistance in parallel with each cap.
Yes, it looks like the high-pass cutoff frequency would be 0.4 Hz. But what about the low-pass cutoff frequency, for the 2200 Ohms with 470 pF to ground? That looks like it gives about 154 kHz. I might replace the 470 pF with 330 pF, to raise that to 219 kHz.
Anyway, replace the 25 uF with polypropylene. How cheap are you? A 4.7 uF with the 16k would give a high-pass cutoff of about 2.12 Hz. But lower would be better.
Madisound has a 10 uF 100V polypropylene for $0.60. Or, if you can stand to spend a few dollars per cap, you might be able to get better ones.
POLYPROPYLENE CAPACITORS from Parts Express ship same day and come with 45 day money back guarantee. Free Shipping Available. Order free 10,000 product catalog.
https://www.madisound.com/store/adv..._in_description=1&inc_subcat=1&page=1&sort=4a
Mouser Electronics - Electronic Component Distributor
Electronic Components Distributor | DigiKey Corp. | US Home Page
With film caps, one main thing to look for is the construction. One way they could make them would be to attach the leads to the ends of two long strips of thin foil with a thin strip of PP under each strip of foil, then roll those up together. That way could result in a lot of inductance. The better way is to let the foil extend out past the edge of the PP all the way along one side (with the two strips hanging out on opposite sides), and then rolling them. Then end plates can contact one entire edge of each foil strip and the leads can be attached to the endplates. That way you don't get much inductance.
Last edited:
Thank you for a response, I appreciate the dialogue!
Could definitely be correct, elsewhere sometimes they print the "V" or sometimes 10/100 does appear to mean 10 microfarads and 100 volt rating.
I hosted the schematics at:
http://mysite.verizon.net/gregorypinkowski/PR1800-schematic.pdf
please let me know if that works for you (big).
Ignore the last page, that's the input for the home model without balanced XLR inputs or the phase inverter for bridged mono operation.
driver page 3 (of 8), right at the input, capacitor C1 labelled NP 25/25 (note comment in title block to assume microfarads).
Uh, the low-pass (high-pass to ground), rule of thumb * 10 to avoid phase shift. Then again my ears probably don't go past 15K. Output from my ribbons rolled off above 10K, never did rule out these amps.
Yes, I've seen the same 10 mfd as Madisond on eBay. Two in parallel = 20 mfd would be closer to the original 25 mfd. Whaddaya think? Sure are cheap enough. Wish I really knew how to get value for my money.
What about the output filter on page 4. .047 at 250 volt. Should I open up an amp and see if it's an electrolytic?
What about any caps in the feedback circuit?
If they have a DC bias I'd just as soon leave whatever's there.
Any other good ideas?
Could definitely be correct, elsewhere sometimes they print the "V" or sometimes 10/100 does appear to mean 10 microfarads and 100 volt rating.
I hosted the schematics at:
http://mysite.verizon.net/gregorypinkowski/PR1800-schematic.pdf
please let me know if that works for you (big).
Ignore the last page, that's the input for the home model without balanced XLR inputs or the phase inverter for bridged mono operation.
driver page 3 (of 8), right at the input, capacitor C1 labelled NP 25/25 (note comment in title block to assume microfarads).
Uh, the low-pass (high-pass to ground), rule of thumb * 10 to avoid phase shift. Then again my ears probably don't go past 15K. Output from my ribbons rolled off above 10K, never did rule out these amps.
Yes, I've seen the same 10 mfd as Madisond on eBay. Two in parallel = 20 mfd would be closer to the original 25 mfd. Whaddaya think? Sure are cheap enough. Wish I really knew how to get value for my money.
What about the output filter on page 4. .047 at 250 volt. Should I open up an amp and see if it's an electrolytic?
What about any caps in the feedback circuit?
If they have a DC bias I'd just as soon leave whatever's there.
Any other good ideas?
Yes, I was able to download and view the schematics, with no problem.
I would replace ALL of the electrolytics, if possible, with new electrolytics (except for the input cap that we already talked about, where teflon or polypropylene would be best).
Some Caig DeOxit would be good for the switches and connectors. Caig.com or local electronics stores.
It would help if you could find a service manual, since it might have procedures for setting the biases and output offsets, etc.
You could check ebay.com for caps. Mouser.com is probably my favorite (no minimum order and very cheap shipping, plus very fast service).
I would replace ALL of the electrolytics, if possible, with new electrolytics (except for the input cap that we already talked about, where teflon or polypropylene would be best).
Some Caig DeOxit would be good for the switches and connectors. Caig.com or local electronics stores.
It would help if you could find a service manual, since it might have procedures for setting the biases and output offsets, etc.
You could check ebay.com for caps. Mouser.com is probably my favorite (no minimum order and very cheap shipping, plus very fast service).
Last edited:
Yes, I need some of that Caig DeOxit for the 1/4 input jacks ASAP.
I'll replace the big caps someday,,,but for now they work fine.
Yes, I've used Mouser before...many many years ago.
What about any other small caps?
I'll replace the big caps someday,,,but for now they work fine.
Yes, I've used Mouser before...many many years ago.
What about any other small caps?
There are lots of small electrolytics. There's a good chance that everything over a few tenths of a uF is electrolytic. I see quite a few more, in your schematic.
Electrolytics go bad fairly quickly. After 20 or 30 years they should probably just be replaced without a second thought, even in unused equipment, unless you're into "re-forming" them, by bringing the voltage up slowly. I much prefer just replacing them.
Electrolytics go bad fairly quickly. After 20 or 30 years they should probably just be replaced without a second thought, even in unused equipment, unless you're into "re-forming" them, by bringing the voltage up slowly. I much prefer just replacing them.
Most are not in the signal path. Many have some DC offset, in which case electrolytics are OK. Right now I'm looking for ones to upgrade with an opportunity to improve the sound, in the signal path or negative feedback path.
I can replace the other ones that work later. Right now they work fine.
I can replace the other ones that work later. Right now they work fine.
I should re-visit my preamp too. Lots of room for capacitor improvement there, and upgrading mediocre op-amps.
- Status
- Not open for further replies.