I was wondering if having two 220uF caps in series as the first node would result in a 50 watt Fender amp sounding stiff and lifeless.
I realise that I’m generalising as the regulation of the power transformer will contribute as well.
I don’t have any experience in changing cap values and I was curious what people’s findings are?
Thanks for your time and replies.
Mojo talk - still power supplies?
Moderators: pompeiisneaks, Colossal
Mojo talk - still power supplies?
Yours Sincerely
Mark Abbott
Mark Abbott
-
Stevem
- Posts: 5144
- Joined: Fri Jan 24, 2014 3:01 pm
- Location: 1/3rd the way out one of the arms of the Milkyway.
2 others liked this
Re: Mojo talk - still power supplies?
What are you looking for out of an amp?
Then amount of filtering you have in the first power supply node will really have little effect on sound and play feel only once you have the amp cranked up loud enough to get the power supply voltage to start to sag.
The other effect of large amounts of filtering is a lowering of the 120 hz noise floor in the amp.
Then amount of filtering you have in the first power supply node will really have little effect on sound and play feel only once you have the amp cranked up loud enough to get the power supply voltage to start to sag.
The other effect of large amounts of filtering is a lowering of the 120 hz noise floor in the amp.
When I die, I want to go like my Grandfather did, peacefully in his sleep.
Not screaming like the passengers in his car!
Cutting out a man's tongue does not mean he’s a liar, but it does show that you fear the truth he might speak about you!
Not screaming like the passengers in his car!
Cutting out a man's tongue does not mean he’s a liar, but it does show that you fear the truth he might speak about you!
Re: Mojo talk - still power supplies?
I have heard some techs remark that if a supply is over filtered the amps clean tones can sound stiff and two dimensional. You are correct that the load on the power supply will affect the sound of the amp. I’m wondering if filter caps valuesis another wives tale or is there some merit to it?Stevem wrote: ↑Tue Jan 07, 2025 11:54 am What are you looking for out of an amp?
Then amount of filtering you have in the first power supply node will really have little effect on sound and play feel only once you have the amp cranked up loud enough to get the power supply voltage to start to sag.
The other effect of large amounts of filtering is a lowering of the 120 hz noise floor in the amp.
People building AC-30 clones seem to still to the original values of filter cap apart from Ken Fischer etc.
As mentioned, I don’t have much experience with filter caps effecting the sound of an amp and wondered what observations people had made in building and repairing amps?
Yours Sincerely
Mark Abbott
Mark Abbott
Re: Mojo talk - still power supplies?
Big filter caps will affect the low end feel. You will notice that a Bass Guitar will sound cleaner on that low E and a Guitar will sound a little over powering, but it depends on what speakers are used.
A clean low end works good for a Bass Amp, not so much for a guitar amp.
A clean low end works good for a Bass Amp, not so much for a guitar amp.
Re: Mojo talk - still power supplies?
There is a technical case to be made for the size of the filter caps affecting tone. I put the "sterile sound" description down to not enough technical investigation more than mojo.
If you feed a tube amp a B+ that's heavily contaminated with ripple, you can detect the 2x mains harmonics in the output at some level. Exactly how big the harmonics are depends on how much the amp rejects ripple, either through careful circuit design or by feedback suppressing it. Ripple on a push-pull output stage is first-order canceled by the transformer while both tube/sides are conducting, so class A biased amps reject ripple pretty well. Class AB less well because one or the other shuts off for part of an audio cycle. Part of the audio cycle matters because the ripple is rejected differently during a ripple cycle time, and the audio is not synchronized to that. I expect that careful enough evaluation would turn up some intermodulation between the audio and ripple.
Historically, musicians have preferred music made with more harmonics, even intermod harmonics, when they're not too big or too high-order. Certainly the audio sounds like more is going on in there with some amount of extra frequencies.
Then there is the obvious compression. Using more power out of the amp makes the B+ sag, and that both makes the gain of some stages lower and makes the ripple higher and more prominent.
I've read a lot of Golden Age technical books on amp design - in fact, I have hard copies of the MIT Radiation Lab amplifiers textbook, third and fourth editions of the Radiotron Designers Handbook, the Audio Cyclopedia, as well as reprints of a lot of older booklets and 50s-60s amplifier textbooks. The issues were probably addressed somewhere in there, although since I haven't read them for a decade or so I can't recall it. The 50s and 60s stuff would be heavily contaminated by hifi viewpoints; I don't recall the "too much cap is sterile" from before the 1980s. But I may just have missed that.
My best guess, absent more technical measurements, is that there is something there, AND that it's gotten somewhat exaggerated by people looking for subtle differences to hand their hat on. Just a guess. More info needed.
If you feed a tube amp a B+ that's heavily contaminated with ripple, you can detect the 2x mains harmonics in the output at some level. Exactly how big the harmonics are depends on how much the amp rejects ripple, either through careful circuit design or by feedback suppressing it. Ripple on a push-pull output stage is first-order canceled by the transformer while both tube/sides are conducting, so class A biased amps reject ripple pretty well. Class AB less well because one or the other shuts off for part of an audio cycle. Part of the audio cycle matters because the ripple is rejected differently during a ripple cycle time, and the audio is not synchronized to that. I expect that careful enough evaluation would turn up some intermodulation between the audio and ripple.
Historically, musicians have preferred music made with more harmonics, even intermod harmonics, when they're not too big or too high-order. Certainly the audio sounds like more is going on in there with some amount of extra frequencies.
Then there is the obvious compression. Using more power out of the amp makes the B+ sag, and that both makes the gain of some stages lower and makes the ripple higher and more prominent.
I've read a lot of Golden Age technical books on amp design - in fact, I have hard copies of the MIT Radiation Lab amplifiers textbook, third and fourth editions of the Radiotron Designers Handbook, the Audio Cyclopedia, as well as reprints of a lot of older booklets and 50s-60s amplifier textbooks. The issues were probably addressed somewhere in there, although since I haven't read them for a decade or so I can't recall it. The 50s and 60s stuff would be heavily contaminated by hifi viewpoints; I don't recall the "too much cap is sterile" from before the 1980s. But I may just have missed that.
My best guess, absent more technical measurements, is that there is something there, AND that it's gotten somewhat exaggerated by people looking for subtle differences to hand their hat on. Just a guess. More info needed.
"It's not what we don't know that gets us in trouble. It's what we know for sure that just ain't so"
Mark Twain
Mark Twain
Re: Mojo talk - still power supplies?
Although there is a mountain of data in the books you sited, I don't think there is much on the sound of an amp driven to hard clipping the way a Guitar amp is. I built a 130W Bass amp with a Pi filtered power supply with about 1V pk to pk ripple on the power supply at full rail to rail square wave. For smaller gigs I built a 50W version.
Re: Mojo talk - still power supplies?
Analysis vs Application
Depending on your sonic goals, you might like a lot of filter capacitance, or you might prefer very little filter capacitance. Whether you hear an effect & how much depends on whether you blast the snot out of the amp, or if you play at quiet TV volume at home. The only thing that makes sense is for you to tack a ton of capacitance into an amp you know & play it, then form an opinion. At least you will know the outcome is relevant to you use-case, rather than mine, or someone else's.
On another forum, plenty of folks use attenuators, load boxes, or re-ampers and might blast a 50- or 100-watt amp at levels they would never stomach if they were going Amp ---> Speaker. Especially when recording the output for quiet playback, they suddenly notice all kinds of crazy subharmonics, ghost-noting, and intermodulation distortion (but those words mostly describe the same thing). You'll hear it in this snippet where there's an "oblique bend" where one note is held steady while the bend of another note is released.
When the amp is cranked, and heavily taxing an under-filtered power supply, hum increases on the transients or during sustained loud sections. It's mostly no big deal when you're playing in a loud band, since extra hum or "stank" gets masked by everything else. At least one company has added a recreation of this occurrence to their dirt pedals.
But some folks won't like this unexpected contribution to their distortion. These folks are more likely to use a master-volume amp (usually with a stoutly-filtered power supply) and/or rely on pedals for dirt. They separate distortion-creation and compression/dynamics from Loudness.
Only you know your application & preferences. You're going to have to try it for yourself if you want an answer that's meaningful for you.
Re: Mojo talk - still power supplies?
Thank you everyone for responding to my post.
Yours Sincerely
Mark Abbott
Mark Abbott
Re: Mojo talk - still power supplies?
If you are up for a little experimentation try 220 || 220 but put a 25 watt 150 ohm resister in between the rectifiers and the caps. Adds a little sag and widens the conduction angle. Your power transformer will thank you.
Re: Mojo talk - still power supplies?
Hopefully this "50 watt Fender" has a solid-state rectifier. Many-µFs means popping a tube rectifier, unless series resistance is added to lower the peak current through the rectifier.
- Folks often cite "Maximum Capacitance" for a tube rectifier
- The rectifier has a "Peak Current Maximum" rather than a capacitance max; more-µFs increases peak-current, while more-R reduces peak-current
Re: Mojo talk - still power supplies?
Good point. Quiz question: which Fender '50 Watt' amp has a tube rectifier?
Re: Mojo talk - still power supplies?
That depends on how much rounding one does and/or how finely the hair is split.
Most onlookers are used to saying a "2x EL34 amp" is "50 watts." And so many guitarists will also say/think a 2x 6L6 Fender amp is "50 watts" including the following amps with tube rectifiers, even if they only measure "40 watts."
Re: Mojo talk - still power supplies?
Yes, a bit stiff esp. if you use a lot of NFB and/or high voltages. But usually 110uf is fine really on blackface/silverface type amps
No, not lifeless
www.myspace.com/20bonesband
www.myspace.com/prostitutes
Express, Comet 60, Jtm45, jtm50, jmp50, 6g6b, vibroverb, champster, alessandro rottweiler
4x12" w/H75s
www.myspace.com/prostitutes
Express, Comet 60, Jtm45, jtm50, jmp50, 6g6b, vibroverb, champster, alessandro rottweiler
4x12" w/H75s
Re: Mojo talk - still power supplies?
I’m still following this and listening intently.
Yours Sincerely
Mark Abbott
Mark Abbott
Re: Mojo talk - still power supplies?
It's one of those things like RG said. Sometimes people are looking for subtle shades of tonal grey to hang a hat on as the magic bullet.
When we use tube rectifiers, there's a limit to how much capacitance can be applied to the tube. It's usually listed in the tube datasheet as the maximum value for capacitance input power filter. The other way to filter a power supply is with an inductor on the fitler input coming off the rectifier. This results in lower voltage, but limits the current peaks the rectifier sees. If you remember ELI-the-ICE man, that's the clue. The capacitor looks like a short to ground for the peaks coming off the rectifier. For most common guitar amp rectifier tubes, that limit is usually around 40-50-microfarads. Some early tube bass amps employed paralleled tube rectifiers in order to provide enough current and avoid damage from a larger filter capacitor.
Solid state rectifiers offer more peak current than vacuum tubes. Hence, when high voltage silicon diodes became available, we started to see increasing levels of filter capacitors, which improved bass response on higher power amps in particular.
Two things are at play: The AC signal analysis of the power supply reveals that the larger the capacitor, the less resistance it offers to ground, so when the output section is producing lots of bass frequencies and drawing lots current, shunting that AC component to ground results in a stiffer DC rail to feed the power amp.
The other thing is the amount of energy storage in the filter bank. The energy in the capacitor in Joules is given by the formula W=1/2CV^2. The energy storage tends to help with transient response.
One can work through the math and simple analysis of the circuit and voltages and impedance to get an idea what it does.
The "recipe" question of how much capacitance do I need to get a particular flavor is more subjective. If you're using a tube rectifier, you have a limit, beyond which you'll shorten the life the rectifier. The usual practice is to employ the largest value consistent with rectifier life. But, the lower the value, the longer the tube lasts. So, lower power amps will often reduce the primary filter input capacity for reliability and cost reasons.
With a solid state rectifier, you have a lot more leeway. One can increase it the capacity to the limits of the silicon diode, or decrease it until the circuit behaves badly.
If you have a "50-watt" tube amp with a solid state rectifier, and it has 100uf or even more, feel free to experiment with the filter capacitance value. I would suggest maybe go as low as 10% of the design value, just to see an extreme example of what it does. As long as the voltage rating is sufficient, a 10uf filter would reveal the extreme low-limit behavior. Then increase it little in steps to hear the effect. Too little will produce objectionable artifacts such as hum and filter ripple, and some point, you'll hit a limit beyond which audible changes cease. There are many cases of unnecessary capacitor purchasing for tube amps.
More to it than just the filter, of course. A solid state rectifier will usually have lower source impedance, which probably also increases the sense of "tightness".
Experiment and see. Part of the fun.
When we use tube rectifiers, there's a limit to how much capacitance can be applied to the tube. It's usually listed in the tube datasheet as the maximum value for capacitance input power filter. The other way to filter a power supply is with an inductor on the fitler input coming off the rectifier. This results in lower voltage, but limits the current peaks the rectifier sees. If you remember ELI-the-ICE man, that's the clue. The capacitor looks like a short to ground for the peaks coming off the rectifier. For most common guitar amp rectifier tubes, that limit is usually around 40-50-microfarads. Some early tube bass amps employed paralleled tube rectifiers in order to provide enough current and avoid damage from a larger filter capacitor.
Solid state rectifiers offer more peak current than vacuum tubes. Hence, when high voltage silicon diodes became available, we started to see increasing levels of filter capacitors, which improved bass response on higher power amps in particular.
Two things are at play: The AC signal analysis of the power supply reveals that the larger the capacitor, the less resistance it offers to ground, so when the output section is producing lots of bass frequencies and drawing lots current, shunting that AC component to ground results in a stiffer DC rail to feed the power amp.
The other thing is the amount of energy storage in the filter bank. The energy in the capacitor in Joules is given by the formula W=1/2CV^2. The energy storage tends to help with transient response.
One can work through the math and simple analysis of the circuit and voltages and impedance to get an idea what it does.
The "recipe" question of how much capacitance do I need to get a particular flavor is more subjective. If you're using a tube rectifier, you have a limit, beyond which you'll shorten the life the rectifier. The usual practice is to employ the largest value consistent with rectifier life. But, the lower the value, the longer the tube lasts. So, lower power amps will often reduce the primary filter input capacity for reliability and cost reasons.
With a solid state rectifier, you have a lot more leeway. One can increase it the capacity to the limits of the silicon diode, or decrease it until the circuit behaves badly.
If you have a "50-watt" tube amp with a solid state rectifier, and it has 100uf or even more, feel free to experiment with the filter capacitance value. I would suggest maybe go as low as 10% of the design value, just to see an extreme example of what it does. As long as the voltage rating is sufficient, a 10uf filter would reveal the extreme low-limit behavior. Then increase it little in steps to hear the effect. Too little will produce objectionable artifacts such as hum and filter ripple, and some point, you'll hit a limit beyond which audible changes cease. There are many cases of unnecessary capacitor purchasing for tube amps.
More to it than just the filter, of course. A solid state rectifier will usually have lower source impedance, which probably also increases the sense of "tightness".
Experiment and see. Part of the fun.