How Is Appropriate Size Determined For Filter Caps?
Moderators: pompeiisneaks, Colossal
How Is Appropriate Size Determined For Filter Caps?
I always base my filter cap values on what has been traditionally used in similar circuits. But I'd now like to learn how to properly size them.
The two obvious categories are power amp and preamp. If the cap is only supporting a single dual triode tube, is 8uf sufficient, 16uf, 22uf, etc? As more tubes are added, how should the filter cap go up in value?
What is the harm of having too much capacitance in the filter caps?
If there's a good article on this that anyone if familiar with, please direct me there.
Thanks guys!
The two obvious categories are power amp and preamp. If the cap is only supporting a single dual triode tube, is 8uf sufficient, 16uf, 22uf, etc? As more tubes are added, how should the filter cap go up in value?
What is the harm of having too much capacitance in the filter caps?
If there's a good article on this that anyone if familiar with, please direct me there.
Thanks guys!
Don
Unencumbered by Knowledge
Trapped inside this old man's body, is the mind of an old man
Unencumbered by Knowledge
Trapped inside this old man's body, is the mind of an old man
- LeftyStrat
- Posts: 3117
- Joined: Fri Jan 14, 2005 6:58 pm
- Location: Marietta, SC, but my heart and two of my kids are in Seattle, WA
Re: How Is Appropriate Size Determined For Filter Caps?
Here's an article about the baseman:
http://www.ampbooks.com/home/classic-ci ... AB-ripple/
It's a balancing act and mostly comes from experimentation. If anyone does have any other info, I'd sure love to hear it.
Heavier filtering tends to give stronger bass and a more 'stiff' amplifier. At the extreme it can make an amp lifeless.
Less filtering tends to add compression and a looser feel. At it's extreme it can lead to ghosting (and motor boating but I believe this has more to do with improper isolation of stages). It can result in flubby bass.
Experimentation is key. Take an old design and double the first couple of filter caps, and you can definitely hear and feel a difference.
I tend to try to find that fine line between just enough for the bass response you desire, and nothing more.
http://www.ampbooks.com/home/classic-ci ... AB-ripple/
It's a balancing act and mostly comes from experimentation. If anyone does have any other info, I'd sure love to hear it.
Heavier filtering tends to give stronger bass and a more 'stiff' amplifier. At the extreme it can make an amp lifeless.
Less filtering tends to add compression and a looser feel. At it's extreme it can lead to ghosting (and motor boating but I believe this has more to do with improper isolation of stages). It can result in flubby bass.
Experimentation is key. Take an old design and double the first couple of filter caps, and you can definitely hear and feel a difference.
I tend to try to find that fine line between just enough for the bass response you desire, and nothing more.
Last edited by LeftyStrat on Mon Mar 12, 2012 9:16 pm, edited 1 time in total.
It's never too late to have a happy childhood.
Re: How Is Appropriate Size Determined For Filter Caps?
Also, the Rectifier type is a starting point. Tube Rectifiers have a limit on the first cap value, in most circuits. There are creative ways to increase the upper limit.
Re: How Is Appropriate Size Determined For Filter Caps?
I agree with LeftyStrat's comments. Also be aware that if your amp uses a tube rectifier, the first (sometimes referred to "primary" or "reservoir")filter caps should be limited in capacitance, depending upon the tube rectifier used (5Y3GT, 5U4GT, 5AR4, etc.), otherwise the rectifier tube can blow. This info is available on most tube data sheets. For example, a 5AR4 tube is limited to 60uf of capacitance, a 5U4GB is limited to 40uf and a 5Y3GT is limited to 10uf. If the amp uses solid state rectifiers, you can use more capacitance. I've also read that excess filter capacitance in the preamp circuits is not a good thing; best to stay with 20uf or less.
-
Marioverkill
- Posts: 61
- Joined: Thu Jan 05, 2012 2:23 pm
- Location: NB,Canada
Re: How Is Appropriate Size Determined For Filter Caps?
Designing Power Supplies for Tube Amplifiers by Merlin Blencowe,chapter 4:Smoothing,Filtering and Decoupling.Also RDH4 chapter 31.
Mario
Mario
- LeftyStrat
- Posts: 3117
- Joined: Fri Jan 14, 2005 6:58 pm
- Location: Marietta, SC, but my heart and two of my kids are in Seattle, WA
Re: How Is Appropriate Size Determined For Filter Caps?
Oh yeah, I totally forgot to mention rectifier constraints, I was mostly thinking about silicon rectification.
And +1 on Merlin's books.
And +1 on Merlin's books.
It's never too late to have a happy childhood.
Re: How Is Appropriate Size Determined For Filter Caps?
Thanks, guys! Your replies are quite helpful. Enough so that I'm prepared to carry on, and select the values for my next build. This build is SS rectified.
I will order the Blencowe book on power supplies. It's one I don't have, but should.
I will order the Blencowe book on power supplies. It's one I don't have, but should.
Don
Unencumbered by Knowledge
Trapped inside this old man's body, is the mind of an old man
Unencumbered by Knowledge
Trapped inside this old man's body, is the mind of an old man
Re: How Is Appropriate Size Determined For Filter Caps?
Hi billyz-billyz wrote:Also, the Rectifier type is a starting point. Tube Rectifiers have a limit on the first cap value, in most circuits. There are creative ways to increase the upper limit.
What are some of the "creative ways"? I often use 5y3's and would like to use a higher mf input cap than the max value spec calls for. I use 20mf without problems yet but would like to go higher in small "clean" amps.
Thanks!
Re: How Is Appropriate Size Determined For Filter Caps?
That book and his Preamp Design book have helped me immensely.Marioverkill wrote:Designing Power Supplies for Tube Amplifiers by Merlin Blencowe,chapter 4:Smoothing,Filtering and Decoupling.Also RDH4 chapter 31.
Mario
It pretty technical reading but he explains it well enough to get a grasp even if you are new to it.
Very useful books.
Not all can learn by reading a book but it is very good reference material.
Tom
Don't let that smoke out!
Don't let that smoke out!
Re: How Is Appropriate Size Determined For Filter Caps?
One way would be to use the correct value for your tube rectifier , say 32mf or less as the first capacitor then a choke then use 100mf to feed the plates of your power tubes.tonewood wrote:Hi billyz-billyz wrote:Also, the Rectifier type is a starting point. Tube Rectifiers have a limit on the first cap value, in most circuits. There are creative ways to increase the upper limit.
What are some of the "creative ways"? I often use 5y3's and would like to use a higher mf input cap than the max value spec calls for. I use 20mf without problems yet but would like to go higher in small "clean" amps.
Thanks!
You may also need some form of current limiting device before the rectifier to limit inrush current at first switch-on. ie.a 20 ohm 20W heavy-duty WW resistor should suffice.This method will cause further voltage drop in the B+ supply when under load.
Some have used Thermistors as well.
Re: How Is Appropriate Size Determined For Filter Caps?
tonewood wrote:Fender 5B3 circuit uses 20mF on 5Y3 too. Just about to build one. There are originals with more than 50 years on them still running. I suppose 20mF on 5Y3 is safe.billyz wrote:I often use 5y3's and would like to use a higher mf input cap than the max value spec calls for. I use 20mf without problems yet but would like to go higher in small "clean" amps.
Re: How Is Appropriate Size Determined For Filter Caps?
MCK wrote:There aren't really any maximum capacitor specs for rectifier tubes. What we see quoted are usually from the "Typical Operation" examples given in the data sheets. The real limits on rectifier tube operation are Peak Plate Current (the absolute maximum current the plate can deliver in normal operation), Hot Switching Transient Plate Current (the absolute maximum the plate can withstand for up to .02 seconds), plus the AC Plate Supply Voltage and the Average DC Output Current (which are interelated and therefore shown graphically on a Rating Chart rather than as a number). The Transient Plate Current limit is very large (4.6A for a 5U4), but a VERY large capacitor or a shorted capacitor might exceed it and the rectifier tube will arc. Since Supply Voltage and Output Current are related, you can have more current if your supply voltage is lower or you can use a higher supply voltage your current demand is lower, as long as everything stays below the lines at the top of the chart.tonewood wrote:Fender 5B3 circuit uses 20mF on 5Y3 too. Just about to build one. There are originals with more than 50 years on them still running. I suppose 20mF on 5Y3 is safe.billyz wrote:I often use 5y3's and would like to use a higher mf input cap than the max value spec calls for. I use 20mf without problems yet but would like to go higher in small "clean" amps.
Interestingly, rectifier ratings originally assumed the tube might be operated off batteries (early radios, cars, airplanes) with AC provided by a vibrator, or with the plates directly supplied by 110VAC, so they didn't necessarily account for PT winding resistance and PT inductance. Crappy transformers (regulation of 15% or worse) are nicer to rectifier tubes because they have a higher effective plate supply impedance. You can add your own impedance by putting resistors (or thermistors) on the primary of secondary of the PT (according to Alex Dumble, some Tweed Fenders had these resistors wired inside the wrap of the PTs). PTs also have inductance so provide choke-like protection to the rectifier (unfortunately PT inductance varies with operation so it isn't a published number).
As long as you don't try to hang a pair KT88s off a 5Y3 (or a quad of anything off a single rectifier tube), I would suggest that can use an initial cap filter of up to 80uF on a 5U4, 60uF on a 5AR4 and 35uF on a 5Y3 (probably even 40uF, since some recent texts show up to 60).
If in doubt, add a little series resistance between the rectifier and the first cap to soften the demand.
Re: How Is Appropriate Size Determined For Filter Caps?
For power amp supply:
- if you have a tube rectifier, check maximum capacity allowed
- for diode rectifier, around 1-2µ2 / Watt for B+
- around 300n-1µ / Watt for screens (it depends on choke values too)
For preamp and PI:
- use the 1Hz low pass filter rule: 20k and 10µ, 10k and 22µ, 4k7 and 47µ and so on.
- if you have a tube rectifier, check maximum capacity allowed
- for diode rectifier, around 1-2µ2 / Watt for B+
- around 300n-1µ / Watt for screens (it depends on choke values too)
For preamp and PI:
- use the 1Hz low pass filter rule: 20k and 10µ, 10k and 22µ, 4k7 and 47µ and so on.
-
titser_marco
- Posts: 131
- Joined: Tue Mar 13, 2007 3:08 pm
Re: How Is Appropriate Size Determined For Filter Caps?
Hi! I came across this after searching for info on filter resistors. Could you please care to explain what this rule means as it relates to filter caps?roberto wrote:For power amp supply:
For preamp and PI:
- use the 1Hz low pass filter rule: 20k and 10µ, 10k and 22µ, 4k7 and 47µ and so on.
Re: How Is Appropriate Size Determined For Filter Caps?
From this page: http://www.valvewizard.co.uk/smoothing.htmltitser_marco wrote:Hi! I came across this after searching for info on filter resistors. Could you please care to explain what this rule means as it relates to filter caps?roberto wrote:For power amp supply:
For preamp and PI:
- use the 1Hz low pass filter rule: 20k and 10µ, 10k and 22µ, 4k7 and 47µ and so on.
"Each RC smoothing stage is a low-pass filter with a cut-off frequency of:
f = 1 / (2 pi R C)"
He who dies with the most tubes... wins