PelliX
Well-Known Member
- Joined
- Jun 16, 2020
- Messages
- 7,295
- Reaction score
- 14,774
I think you mean the harmonics, so 60, 120, 240. Could well be. If it's 120Hz it's coming from the rectifier, so I'd check the filter caps.
Does capacitor brand really make any difference?
- Thread starter ThreeChordWonder
- Start date
This site may earn a commission from merchant affiliate links like Ebay, Amazon, and others.
I think you mean the harmonics, so 60, 120, 240. Could well be. If it's 120Hz it's coming from the rectifier, so I'd check the filter caps.
I'm quite an evidence based person.
I've done a few recaps of old 30+ year old design hifi amps and other equipment, some solid state and some tube based (including an old 1957 receiver).
There's a few things that get lost in the discussion:
Specifications - specifically the DC and AC ratings. Most people seem to look at a single rating but actually an AC rating is as important. You may have a 400V DC rated film cap that then has a sub-200V AC rating. If your cap is dealing with a high gain output where the signal is driving a larger tube, then you may find that AC rating is going to cause you issues. A lower AC rating will end up distorting the signal at higher amplitudes (which could contribute to the NOS/vintage tone).
Specifications also indicate you're going to get a batch of 10 caps that are similar enough to substitute - this is where boutique starts annoying me, specifically where they don't provides a specification sheet, specifically showing the frequency vs ESR, etc
Now specifications, from experience don't make sound/tone.
Reason I know this is from replacing old Jamicon electrolytics in the signal path with Panasonic film caps only to find the sound being dark and muddy. I then switched for WIMA MKP10, not impressed but then WIMA MKP4 which gave clarity.
Now the reason I say - make tone/sound - is that the design (power supplies/components/signal design) all vary the environment that the substituted cap is placed in. The result can cause variation between amps, especially those that have had some drift due to thermal cycling or temperatures.
After switching the Jamicons to CDE (power), Panasonic low ESR organics for pre-stage decoupling and moving to film for the signal coupling caps, lastly switching out the cooked underwattaged resistors for larger metal oil Dale etc, it increased clarity.. that then ended up changing out the opamp pre ... etc etc.. The amp had lost a chunk of it's old character of tonally muddy class AB (heavily biased to operate the majority of it's time in class A), to be come a clear sounding with nuances that showed up the media being used.
Good power caps - back to the specs - that give better ESR at 50/60Hz are going to be better suited than those that are tuned for higher frequencies. The larger the cap, typically the better the ESR. Old caps tended to have a wider applicability for purpose, new caps (by design) tend to be more specialised.
I've directly Bode tested the caps for Panasonic, WIMA and Vishay, and the low voltage bode test does not show any statistically significant difference. What is important here is to place cap in the amp and then give it the full DC and AC through the amp and then check the bode plot.
Having said that a considerable amount of tone is made from so many unspecified components it's quite difficult to define the tone.
* resistors - thermal noise, etc
* caps - DC + AC effects like compression and distortion, blocking, phase shift at frequency etc that can change with temperature.
* tubes - different noise and transconductance with frequency based on previous state (ie a DC, vs AC vs pulse will have different outcome based on the thermal levels off the cathode/heaters, capacitance etc in the tube itself).
* diodes ..
... and the list goes on with power transformers and output transformers.
* line wall voltage and the noise that causes additional distortion effects
Manufacturers like Marshal and Fender are always looking for a statistical similarity but they know they'll never make two amps the same. You'll end up with a stunning amp, a plain Jane and Friday-PM special.. just by the component lotto.
Back to subject.
Caps I like
* Vishay MKT and MKP film caps - not had an issue and sound good.
* WIMA film caps - the massive FKP1s are clear as a bell but they're not audio caps! The MK4 are great sounding for hifi (630Vdc, 400Vac are the ones I like!).
* Panasonic low ESR organics for close in power decoupling
Caps I have no real preference or dislike for
* Panasonic power electrolytics
* CDE power caps
* F&T electrolytics (450V and 500V) although they seem a little slow.
Caps I'm less enamoured by:
* Panasonic film caps
* WIMA MK10
Caps I've not tried and don't have experience of
* Sprague
* IC
* Boutiques like Mundorf, Jansen, MOD
* RS NOS (any NOS electrolytics is very very likely to be way past it's life/shelf life)
So you can see - I don't see a guitar amp as an alchemist/witches brew but something that can be characterised (although with large numbers of variables). I don't mind boutiques IF they can provide visibility of the magic sauce that gives them the ingredient within the tone and they can provide that on a consistent basis with each batch.
If you're not giving a datasheet then.. I start asking uncomfortable questions.
I've done a few recaps of old 30+ year old design hifi amps and other equipment, some solid state and some tube based (including an old 1957 receiver).
There's a few things that get lost in the discussion:
Specifications - specifically the DC and AC ratings. Most people seem to look at a single rating but actually an AC rating is as important. You may have a 400V DC rated film cap that then has a sub-200V AC rating. If your cap is dealing with a high gain output where the signal is driving a larger tube, then you may find that AC rating is going to cause you issues. A lower AC rating will end up distorting the signal at higher amplitudes (which could contribute to the NOS/vintage tone).
Specifications also indicate you're going to get a batch of 10 caps that are similar enough to substitute - this is where boutique starts annoying me, specifically where they don't provides a specification sheet, specifically showing the frequency vs ESR, etc
Now specifications, from experience don't make sound/tone.
Reason I know this is from replacing old Jamicon electrolytics in the signal path with Panasonic film caps only to find the sound being dark and muddy. I then switched for WIMA MKP10, not impressed but then WIMA MKP4 which gave clarity.
Now the reason I say - make tone/sound - is that the design (power supplies/components/signal design) all vary the environment that the substituted cap is placed in. The result can cause variation between amps, especially those that have had some drift due to thermal cycling or temperatures.
After switching the Jamicons to CDE (power), Panasonic low ESR organics for pre-stage decoupling and moving to film for the signal coupling caps, lastly switching out the cooked underwattaged resistors for larger metal oil Dale etc, it increased clarity.. that then ended up changing out the opamp pre ... etc etc.. The amp had lost a chunk of it's old character of tonally muddy class AB (heavily biased to operate the majority of it's time in class A), to be come a clear sounding with nuances that showed up the media being used.
Good power caps - back to the specs - that give better ESR at 50/60Hz are going to be better suited than those that are tuned for higher frequencies. The larger the cap, typically the better the ESR. Old caps tended to have a wider applicability for purpose, new caps (by design) tend to be more specialised.
I've directly Bode tested the caps for Panasonic, WIMA and Vishay, and the low voltage bode test does not show any statistically significant difference. What is important here is to place cap in the amp and then give it the full DC and AC through the amp and then check the bode plot.
Having said that a considerable amount of tone is made from so many unspecified components it's quite difficult to define the tone.
* resistors - thermal noise, etc
* caps - DC + AC effects like compression and distortion, blocking, phase shift at frequency etc that can change with temperature.
* tubes - different noise and transconductance with frequency based on previous state (ie a DC, vs AC vs pulse will have different outcome based on the thermal levels off the cathode/heaters, capacitance etc in the tube itself).
* diodes ..
... and the list goes on with power transformers and output transformers.
* line wall voltage and the noise that causes additional distortion effects
Manufacturers like Marshal and Fender are always looking for a statistical similarity but they know they'll never make two amps the same. You'll end up with a stunning amp, a plain Jane and Friday-PM special.. just by the component lotto.
Back to subject.
Caps I like
* Vishay MKT and MKP film caps - not had an issue and sound good.
* WIMA film caps - the massive FKP1s are clear as a bell but they're not audio caps! The MK4 are great sounding for hifi (630Vdc, 400Vac are the ones I like!).
* Panasonic low ESR organics for close in power decoupling
Caps I have no real preference or dislike for
* Panasonic power electrolytics
* CDE power caps
* F&T electrolytics (450V and 500V) although they seem a little slow.
Caps I'm less enamoured by:
* Panasonic film caps
* WIMA MK10
Caps I've not tried and don't have experience of
* Sprague
* IC
* Boutiques like Mundorf, Jansen, MOD
* RS NOS (any NOS electrolytics is very very likely to be way past it's life/shelf life)
So you can see - I don't see a guitar amp as an alchemist/witches brew but something that can be characterised (although with large numbers of variables). I don't mind boutiques IF they can provide visibility of the magic sauce that gives them the ingredient within the tone and they can provide that on a consistent basis with each batch.
If you're not giving a datasheet then.. I start asking uncomfortable questions.
Last edited:
PelliX
Well-Known Member
- Joined
- Jun 16, 2020
- Messages
- 7,295
- Reaction score
- 14,774
ThreeChordWonder
Well-Known Member
- Joined
- Jan 31, 2021
- Messages
- 517
- Reaction score
- 613
All very nice chaps, and I'm sure you're all correct, but I, for one, am not chasing perfection (whatever that is), just good enough.
Rabbit hole?
Rabbit hole?
That's either a Jamicon or CapXon cap.. Rifa was square bodied
PelliX
Well-Known Member
- Joined
- Jun 16, 2020
- Messages
- 7,295
- Reaction score
- 14,774
Aaww, it's thirsty! It needs electrolytes. They're what
To the OP, I will suggest this when choosing filter caps. Look at the schematic for expected DC voltage and select your replacement caps to have a voltage rating 30%-35% higher than this expected voltage. So, if the circuit is expecting 100V, then you want to select a cap with a voltage rating of 130V at a bare minimum, but since this isn't a standard value, then 150V would be perfectly fine. Headroom is just as important as temperature when it comes to longevity.
Personally, I don't care what the manufacturer put in the circuit in regards to brand. Whenever possible, I choose Nichicon, Rubycon, and Panasonic as a first choice, but we all know sometimes you just can't find your preferred brand in the values you need. Just stay away from the Chinese knock-off brands. They are riddled with problems and won't last long. For instance, if I see Jamicon or Samwha, these immediately get replaced. In my experience, these are garbage and never last long. Putting subpar caps into a circuit can and often does cause issues later on. When a cap fails, it can often send out a surge that tends to take out other components down the line, especially FETs that are highly susceptible to ESD and surges of any kind.
As for tonal differences, I've never noticed a difference, no matter what caps I've chosen and my customers have never complained about the choices of caps I've used. IC caps are not my first choice, but I've had to use them on many occasions and not had any issues. Again, the key is to make sure you properly select your voltage rating. Many manufacturers, especially from years ago, didn't allow for any headroom (or only allowed very little headroom) and this ultimately shortens the life of the capacitor, as does heat ratings. Like others have said, even if the temperature won't ever get up to 105C, it's still a good idea to use higher temperature rated caps simply for their longevity.
On another note, most major distributors will carry top brands at very reasonable prices. Do yourself a favor and don't buy caps from eBay/Amazon, unless it's an absolute last resort. You don't know how old those caps are or what state they're in. Buy from a legitimate distributor, because they'll be guaranteed. You might still get some bad caps here and there, but they'll be completely backed up and the company will replace them at no charge and no hassle. PLUS, you know you won't be getting any fakes that can open up a whole other can of worms.
Personally, I don't care what the manufacturer put in the circuit in regards to brand. Whenever possible, I choose Nichicon, Rubycon, and Panasonic as a first choice, but we all know sometimes you just can't find your preferred brand in the values you need. Just stay away from the Chinese knock-off brands. They are riddled with problems and won't last long. For instance, if I see Jamicon or Samwha, these immediately get replaced. In my experience, these are garbage and never last long. Putting subpar caps into a circuit can and often does cause issues later on. When a cap fails, it can often send out a surge that tends to take out other components down the line, especially FETs that are highly susceptible to ESD and surges of any kind.
As for tonal differences, I've never noticed a difference, no matter what caps I've chosen and my customers have never complained about the choices of caps I've used. IC caps are not my first choice, but I've had to use them on many occasions and not had any issues. Again, the key is to make sure you properly select your voltage rating. Many manufacturers, especially from years ago, didn't allow for any headroom (or only allowed very little headroom) and this ultimately shortens the life of the capacitor, as does heat ratings. Like others have said, even if the temperature won't ever get up to 105C, it's still a good idea to use higher temperature rated caps simply for their longevity.
On another note, most major distributors will carry top brands at very reasonable prices. Do yourself a favor and don't buy caps from eBay/Amazon, unless it's an absolute last resort. You don't know how old those caps are or what state they're in. Buy from a legitimate distributor, because they'll be guaranteed. You might still get some bad caps here and there, but they'll be completely backed up and the company will replace them at no charge and no hassle. PLUS, you know you won't be getting any fakes that can open up a whole other can of worms.
Guitar-Rocker
Well-Known Member
- Joined
- Mar 5, 2012
- Messages
- 3,251
- Reaction score
- 3,168
Ok I'll bite. I've built a few hundred amps so I got to be crazy. I prefer Vishay MKT's 1813 for clarity. Note I did not say for clean. If you don't understand the difference between clarity and clean, then quit reading. To me they are far more soulful / toneful. Yeah I said I'm crazy.
I feel original mustards are very smooth sounding, but always a bit darker. Mallorys display more grit /sizzle tones. Am NOT a fan of Sozos at all, yeah even broken in.
I have not had good experiences with JJ cap cans, noisy has been my take. I use F/T cap cans.
I feel original mustards are very smooth sounding, but always a bit darker. Mallorys display more grit /sizzle tones. Am NOT a fan of Sozos at all, yeah even broken in.
I have not had good experiences with JJ cap cans, noisy has been my take. I use F/T cap cans.
ThreeChordWonder
Well-Known Member
- Joined
- Jan 31, 2021
- Messages
- 517
- Reaction score
- 613
I must admit I'm suspicious of SoZo caps. Anyone who just took the logo of a certain Mr. J. Page of London W1 and messed it about... well, it might fool some of the people all of the time, etc., etc.
I'm also very suspicious of "NOS" stuff.
A lot of the rest seems to come down to your individual preferences, and I'm not sure my hearing is still good enough (I blame being in the audience at 80s Motorhead etc. gigs and motorcycling) to appreciate any difference.
I'm going to try some Vishay coupling caps, but beyond that...
Thanks anyway.
I'm also very suspicious of "NOS" stuff.
A lot of the rest seems to come down to your individual preferences, and I'm not sure my hearing is still good enough (I blame being in the audience at 80s Motorhead etc. gigs and motorcycling) to appreciate any difference.
I'm going to try some Vishay coupling caps, but beyond that...
Thanks anyway.
Similar threads
