Maxkracht and pdf64 make valid points.
I have to disagree on one point though. The power supply is the audio path.
Your signal enters V1's grid and goes....nowhere. No connection to any other element of the amp. It only influences the stream of power supply current flowing from cathode to plate. It recreates a simile of the input. The guitar modulates the power supply current to get amplification. In addition to the guitar signal, every component in the supply influences that current. Can you hear the difference..maybe, maybe not. The Hifi guys say yes, but they buy into tone stones and cable lifters.
But you'll not have to replace them regularly, and they "heal" when a overvoltage spike passes through them, instead of the hard short failures that electrolytics have, risking transformers.
I use MF caps, usually Solen for the larger values, and Audyn, Solen or Dayton for cathode bypass caps.
Practically though, the sizes make use in standard chassis challenging, at least for the main filters. Obbligato does make can versions of their caps to help with mounting, but they are large. Here's a pair on this 10"x17" chassis. I can't find an image of the top, but I think they were 6" high. I had to recess them into the chassis over an inch to keep the head box reasonable as the chassis was already 3 inches tall.
Electrolytic VS Film Power Supply Filter Capacitors!!!
Moderators: pompeiisneaks, Colossal
Re: Electrolytic VS Film Power Supply Filter Capacitors!!!
You do not have the required permissions to view the files attached to this post.
Tube junkie that aspires to become a tri-state bidirectional buss driver.
Re: Electrolytic VS Film Power Supply Filter Capacitors!!!
Film capacitors are technically superior to electrolytics of same/similar rating in all ways except:
1. Size and weight for equivalent capacitance and voltage ("CV product")
2. cost per unit of CV.
Of course, size/weight and cost per CV are so very, very important to manufactured products that electrolytics are the first choice for every application where they can compete at all. It's only when people can hand craft things (like the audience here) that electros even get questioned.
In no particular order,
"the most important topic to me, the main reason that I want to upgrade my guitar tube amp, is to make a major upgrade in performance, to upgrade the reaction to the nuances of the pick attack. "
It is very, very difficult to construct a technical case for the power supply filters being responsible for an audible difference in nuances of pick attack. Yes, everything affects everything else, as the hifi people are so wont to think, but still...
"An easy way to get an electrolytic to respond closer to a film cap is to put a film cap in parallel with it"
Certainly this it truer as frequency goes up. Every capacitor has internal series resistance and inductance. Electros, CV per CV, have more of these than film caps. An ideal cap has an impedance that decreases with increasing frequency *forever*. Real caps reach a minimum impedance at the frequency where the pure capacitance of the thing has an impedance equal to the rising impedance of the self-inductance. At higher frequencies the "capacitor" looks like an inductor. Paralleling any cap with another cap of lower self-inductance helps the combined impedance keep decreasing with higher frequencies. This concept is practically a mantra with RF designers, for nearly a century.
As with all real-world stuff, you can rapidly reach a point of diminishing returns. A capacitor's impedance just needs to be "low" compared to whatever is attached after the capacitor. This is more often a low frequency issue than a high frequency one; your ears can't hear voltage change speeds faster than would Fourier transform to greater than 20kHz, even if your speakers could reproduce it. And you're trying NOT to amplify MHz frequencies.
"A more noticeable way to lower power supply sag/get a tighter response would be to add bigger filter caps all together."
Yes, that. The more charge stored in a capacitor bank, the more can be delivered in toto, and the more slowly its voltage ramps down with the same load. In frequency domain terms, the bigger the capacitance, the lower the frequencies it will pass.
"Would too much power supply filtering hurt the frequency response of a guitar amp?"
Frequency response? No. Some people *like* the sound of the B+ sagging under heavy, prolonged high power use, calling it "sag". These people would claim the amp was impaired by having too big filter caps. I.e. what maxkracht said.
Re: fancy named caps
Economics is against boutique caps. Making capacitors is a capital-expense and equipment dependent undertaking. I suspect that few or none of the fancy-named cap labels make their own caps. Maybe one or two, but I would doubt it until proven. At best I'd bet on a them buying special-specification runs of caps with their desired specs; at worst, they would buy custom labeled ordinary caps. The boutique naming is a great thing to hang internet-expert testimony for advertising purposes. Once your special La-Di-Da(tm) caps get a reputation going, you can really rake in the bucks by claiming superiorities to other caps that are not measurable.
"What are the benefits of putting a resistor in parallel on each filter capacitor?
...
How do you calculate the proper resistor value to be put in parallel for each filter capacitor?"
It pulls down the high voltage on the cap more quickly when the power is turned off; this is a real advantage to techs, as they can work on a recently turned off amp without worry about being shocked by the remaining stored charge. In the Workhorse amps I put an LED in series with the bleeder resistors on the first filter caps so you could SEE whether the caps were bled down enough.
A secondary use is a resistor in parallel with each cap in a series stack of caps. This forces the caps to share the DC voltage equally, overwhelming the tendency of the lowest-leakage caps to have to stand most of the voltage.
As to how to calculate such resistors, I personally decide how much power I can stand to waste in the resistors, and use the power and voltage to calculate the resistor value. For instance, a cap has 250V max across it. If I had to keep power under 1W (for a 2W rated resistor) I would calculate 1W = 250*250/R and then R = 250*250/1 = 62.5K. That's a minimum value. I would probably use a 100K resistor, and it would dissipate P = 250*250/100,000 = 62,500/100,000 = 0.625W. If I had to use a 1/2W resistor, then the smallest resistor would be 0.5W = 250*250/0.25 = 250K. I'd probably use a 270K.
"Is there a specification for the Dissipation Factor for an electrolytic capacitor?"
Yes.
https://en.wikipedia.org/wiki/Dissipation_factor
1. Size and weight for equivalent capacitance and voltage ("CV product")
2. cost per unit of CV.
Of course, size/weight and cost per CV are so very, very important to manufactured products that electrolytics are the first choice for every application where they can compete at all. It's only when people can hand craft things (like the audience here) that electros even get questioned.
In no particular order,
"the most important topic to me, the main reason that I want to upgrade my guitar tube amp, is to make a major upgrade in performance, to upgrade the reaction to the nuances of the pick attack. "
It is very, very difficult to construct a technical case for the power supply filters being responsible for an audible difference in nuances of pick attack. Yes, everything affects everything else, as the hifi people are so wont to think, but still...
"An easy way to get an electrolytic to respond closer to a film cap is to put a film cap in parallel with it"
Certainly this it truer as frequency goes up. Every capacitor has internal series resistance and inductance. Electros, CV per CV, have more of these than film caps. An ideal cap has an impedance that decreases with increasing frequency *forever*. Real caps reach a minimum impedance at the frequency where the pure capacitance of the thing has an impedance equal to the rising impedance of the self-inductance. At higher frequencies the "capacitor" looks like an inductor. Paralleling any cap with another cap of lower self-inductance helps the combined impedance keep decreasing with higher frequencies. This concept is practically a mantra with RF designers, for nearly a century.
As with all real-world stuff, you can rapidly reach a point of diminishing returns. A capacitor's impedance just needs to be "low" compared to whatever is attached after the capacitor. This is more often a low frequency issue than a high frequency one; your ears can't hear voltage change speeds faster than would Fourier transform to greater than 20kHz, even if your speakers could reproduce it. And you're trying NOT to amplify MHz frequencies.
"A more noticeable way to lower power supply sag/get a tighter response would be to add bigger filter caps all together."
Yes, that. The more charge stored in a capacitor bank, the more can be delivered in toto, and the more slowly its voltage ramps down with the same load. In frequency domain terms, the bigger the capacitance, the lower the frequencies it will pass.
"Would too much power supply filtering hurt the frequency response of a guitar amp?"
Frequency response? No. Some people *like* the sound of the B+ sagging under heavy, prolonged high power use, calling it "sag". These people would claim the amp was impaired by having too big filter caps. I.e. what maxkracht said.
Re: fancy named caps
Economics is against boutique caps. Making capacitors is a capital-expense and equipment dependent undertaking. I suspect that few or none of the fancy-named cap labels make their own caps. Maybe one or two, but I would doubt it until proven. At best I'd bet on a them buying special-specification runs of caps with their desired specs; at worst, they would buy custom labeled ordinary caps. The boutique naming is a great thing to hang internet-expert testimony for advertising purposes. Once your special La-Di-Da(tm) caps get a reputation going, you can really rake in the bucks by claiming superiorities to other caps that are not measurable.
"What are the benefits of putting a resistor in parallel on each filter capacitor?
...
How do you calculate the proper resistor value to be put in parallel for each filter capacitor?"
It pulls down the high voltage on the cap more quickly when the power is turned off; this is a real advantage to techs, as they can work on a recently turned off amp without worry about being shocked by the remaining stored charge. In the Workhorse amps I put an LED in series with the bleeder resistors on the first filter caps so you could SEE whether the caps were bled down enough.
A secondary use is a resistor in parallel with each cap in a series stack of caps. This forces the caps to share the DC voltage equally, overwhelming the tendency of the lowest-leakage caps to have to stand most of the voltage.
As to how to calculate such resistors, I personally decide how much power I can stand to waste in the resistors, and use the power and voltage to calculate the resistor value. For instance, a cap has 250V max across it. If I had to keep power under 1W (for a 2W rated resistor) I would calculate 1W = 250*250/R and then R = 250*250/1 = 62.5K. That's a minimum value. I would probably use a 100K resistor, and it would dissipate P = 250*250/100,000 = 62,500/100,000 = 0.625W. If I had to use a 1/2W resistor, then the smallest resistor would be 0.5W = 250*250/0.25 = 250K. I'd probably use a 270K.
"Is there a specification for the Dissipation Factor for an electrolytic capacitor?"
Yes.
https://en.wikipedia.org/wiki/Dissipation_factor
I don't "believe" in science. I trust science. Science works, whether I believe in it or not.
Re: Electrolytic VS Film Power Supply Filter Capacitors!!!
You are of course correct, but potentially semantic. My mind and mental models are too simple for that…
Is an amp a power supply being wiggled; Is it a wiggle being supplied power? Where can I buy a tone stone?
Re: Electrolytic VS Film Power Supply Filter Capacitors!!!
The only good approximation is that an amplifier is fundamentally a power supply that has some other junk attached to it to let some of the power out in a carefully controlled manner.
I don't "believe" in science. I trust science. Science works, whether I believe in it or not.
Re: Electrolytic VS Film Power Supply Filter Capacitors!!!
This is a bit off-topic, but I want to share the very best that I have been learning!
I reached out again to Nick Lucas from https://www.hificollective.co.uk. Asking him about the other component upgrades for my amp, this was the email response from Nick:
"Hi
Mix it up a little, use some 2W Allen Bradleys and the 2W AMRGs in the circuit. The Bradleys will add a slight bit of warmth and colouration and the AMRG will give the detail.
Regards
Nick "
This along with his suggestions of using this particular brand of electrolytic capacitors. This is from another email Nick sent me:
"Hi
See
CAP-100-R-22U-500V: 22uF 500Vdc Audio Note Kaisei POLAR Electrolytic Capacitor
and
CAP-100-R-50U-500V: 50uF 500Vdc Audio Note Kaisei POLAR Electrolytic Capacitor
Regards, Nick"
Thanks everyone for all the great explanations and details that were shared here. This is an amazing place to have people sharing their wisdom with others and provides an historical record to be used as educational material and as a future points of reference! This was so much fun!!!
Thanks again for sharing!!!
I reached out again to Nick Lucas from https://www.hificollective.co.uk. Asking him about the other component upgrades for my amp, this was the email response from Nick:
"Hi
Mix it up a little, use some 2W Allen Bradleys and the 2W AMRGs in the circuit. The Bradleys will add a slight bit of warmth and colouration and the AMRG will give the detail.
Regards
Nick "
This along with his suggestions of using this particular brand of electrolytic capacitors. This is from another email Nick sent me:
"Hi
See
CAP-100-R-22U-500V: 22uF 500Vdc Audio Note Kaisei POLAR Electrolytic Capacitor
and
CAP-100-R-50U-500V: 50uF 500Vdc Audio Note Kaisei POLAR Electrolytic Capacitor
Regards, Nick"
Thanks everyone for all the great explanations and details that were shared here. This is an amazing place to have people sharing their wisdom with others and provides an historical record to be used as educational material and as a future points of reference! This was so much fun!!!
Thanks again for sharing!!!
Re: Electrolytic VS Film Power Supply Filter Capacitors!!!
This is off-topic, but it has value and I want to share what I have been learning. Listening to Bob Gjika talks about inrush current, I found a possible solution other than a Large Chock solution!
That would be Inrush Current Limiters & Sensing Thermistor Solutions from, https://www.ametherm.com/. A very interesting site with great tech support!
https://www.ametherm.com/inrush-current/
https://www.ametherm.com/datasheets/sl105r003
Thanks everyone for sharing. This site and the genius minds here have been so helpful!!!
That would be Inrush Current Limiters & Sensing Thermistor Solutions from, https://www.ametherm.com/. A very interesting site with great tech support!
https://www.ametherm.com/inrush-current/
https://www.ametherm.com/datasheets/sl105r003
Thanks everyone for sharing. This site and the genius minds here have been so helpful!!!
Re: Electrolytic VS Film Power Supply Filter Capacitors!!!
Inrush current limiters? Try this:
http://www.geofex.com/Article_Folders/m ... 0Clamp.pdf
The current clamp action stops all the current greater than its limit as set by the source resistor. This not only tames the inrush current at power-on time, but also prevents any pulse currents into filter caps under heavy loads. The side effect is removing any currents that could damage a tube rectifier if the first filter cap is bigger than the spec on the rectifier tube.
To a certain extent, it tames the AC-primary side inrush current too. The primary side no longer has to provide the big currents into the secondaries for power on.
http://www.geofex.com/Article_Folders/m ... 0Clamp.pdf
The current clamp action stops all the current greater than its limit as set by the source resistor. This not only tames the inrush current at power-on time, but also prevents any pulse currents into filter caps under heavy loads. The side effect is removing any currents that could damage a tube rectifier if the first filter cap is bigger than the spec on the rectifier tube.
To a certain extent, it tames the AC-primary side inrush current too. The primary side no longer has to provide the big currents into the secondaries for power on.
I don't "believe" in science. I trust science. Science works, whether I believe in it or not.
Re: Electrolytic VS Film Power Supply Filter Capacitors!!!
R.G., Bless you. This information is very valuable and needed! This site is amazing and a great place to share ideas, wisdom and people to lean form great minds what works.
Thanks again for sharing your years of wisdom with me and everyone here. RG, you are an inspiration to all!!!
Thanks again for sharing your years of wisdom with me and everyone here. RG, you are an inspiration to all!!!
Re: Electrolytic VS Film Power Supply Filter Capacitors!!!
I want to personally thank everyone for sharing their wisdom with us all. Especially, maxkracht, TUBEDUDE, and R.G. for taking the time to explain in details what is happening and why. This has great educational value for all generation to bring clarity to this topic.
I have not heard back from F&T Caps from Germany, as soon as I do, I will post their response. We had input from amp builders and from Nick Lucas at https://www.hificollective.co.uk. Well done everyone.
Thank you for sharing the names of manufactures of caps you use and like, F&T, Sprague, Panasonic. I believe they have the better specs for ESR, Ripple Current, Life Expectancy, Max Temp etc...
I have not heard back from F&T Caps from Germany, as soon as I do, I will post their response. We had input from amp builders and from Nick Lucas at https://www.hificollective.co.uk. Well done everyone.
Thank you for sharing the names of manufactures of caps you use and like, F&T, Sprague, Panasonic. I believe they have the better specs for ESR, Ripple Current, Life Expectancy, Max Temp etc...