I understand some of the basic theory of power supplies: rectification, how the filter caps smooth the ripple. Dropping resistors I'm a little hazy on. Can someone explain, or is there an article somewhere which explains, how the values of the capacitors and resistors in the PS circuit are chosen?
It often seems a very imprecise science - I see posts saying things like "add a 10K to 20K resistor" at some point in the PS. Well, which is it? One is double the other! Surely they would have very different results?
Many thanks!
Power supply component values
Moderators: pompeiisneaks, Colossal
Re: Power supply component values
What voltage do you want at each node on the high voltage string? Use ohm's law and the current draw at each node (remembering that all the nodes downstream are also pulling current) and calculate the resistor values. Then go to the accountant and see how much money you are allowed to spend and that will most likely determine the value of the caps. The smallest value possible (cheapest) and still get away with an acceptable level of hum and hopefully no motorboating.
Larger value resistors will give better filtering but also more of a voltage drop. If the resistors are too large and the caps too big, the time constant gets pretty long for charging the caps. Not a terrible problem on the high voltage but you certainly don't want it to happen on the bias supply.
People have gotten used to the squishy sound of an old fender amp, but quite honestly, when you look at the schematic, it sure looks like it was designed to save money... the filter caps are as small and as few as could be gotten away with. Lower gain amps, like fenders, are more forgiving of under-designed power supplies. High gain amps are not.
It's a trade-off, more capacitance generally equates to less hum and more stability (all things being equal) yet is a stiffer feel, more solid, in your face, a solid low end, a perception of more power (because of the current in reserve for transients).
It's all about what you want, what you are trying to do, what you can get away with, what the accountant will let you spend.
Larger value resistors will give better filtering but also more of a voltage drop. If the resistors are too large and the caps too big, the time constant gets pretty long for charging the caps. Not a terrible problem on the high voltage but you certainly don't want it to happen on the bias supply.
People have gotten used to the squishy sound of an old fender amp, but quite honestly, when you look at the schematic, it sure looks like it was designed to save money... the filter caps are as small and as few as could be gotten away with. Lower gain amps, like fenders, are more forgiving of under-designed power supplies. High gain amps are not.
It's a trade-off, more capacitance generally equates to less hum and more stability (all things being equal) yet is a stiffer feel, more solid, in your face, a solid low end, a perception of more power (because of the current in reserve for transients).
It's all about what you want, what you are trying to do, what you can get away with, what the accountant will let you spend.
Re: Power supply component values
I have access to a bunch of 3300uFD 450V electrolytics......I wonder if those would have any use?
Re: Power supply component values
Thanks Jana. That all makes sense, especially the accountant part!
Can you point me to where there may be a practical example of the process? Y'know along the lines of "I'm building an amp with 2 EL84s, a 2 x 12AX7 preamp and a 12AX7 PI, so therefore I need XXX volts on the power tubes, XXX on the PI, XXX on the preamp, and to get that I calculate the current draw like this... blah blah" etc, etc.
Thanks!
Can you point me to where there may be a practical example of the process? Y'know along the lines of "I'm building an amp with 2 EL84s, a 2 x 12AX7 preamp and a 12AX7 PI, so therefore I need XXX volts on the power tubes, XXX on the PI, XXX on the preamp, and to get that I calculate the current draw like this... blah blah" etc, etc.
Thanks!
IanG
Re: Power supply component values
I don't think I would want to flip the standby switch with those installed! I'm thinking the inrush current would be... um... massive.I have access to a bunch of 3300uFD 450V electrolytics......I wonder if those would have any use?
No, I can't, sorry. What I know is gleaned from pouring over schematics, years of tinkering, and finding what sounds best to me. I would start with a known schematic of something that is similar to what you want to build and go from there.Can you point me to where there may be a practical example of the process? Y'know along the lines of "I'm building an amp with 2 EL84s, a 2 x 12AX7 preamp and a 12AX7 PI, so therefore I need XXX volts on the power tubes, XXX on the PI, XXX on the preamp, and to get that I calculate the current draw like this... blah blah" etc, etc.
Some of my rules of thumb are, keep voltages around 300 to 320 for EL 84's cathode biased. Less than about 350 to 370 for 6v6 cathode biased (yeah, I know, some fenders run them really hot, like 420 or so, but what the heck, it's only tubes, right?).
For fixed bias, EL34's in a 50 watt plexi sound really good at about 380 to 400 volts, 6l6's sound like dog pooh at those voltages. 6l6's at about 450 to 470 is a nice place to be. El34's at 470 to 480 are quite a bit different sounding animals than at 380.
I never go over 480 on any amp, simply because I don't want to mess with the caps (having to tie two caps in series to get enough voltage rating).
Watch the voltage on preamp tubes, especially if you have a cathode follower. An Ax7 is not supposed to have over 100 volts difference between the cathode and the heaters, although marshall has been doing it for years and gets away with it.
These are all my opinions, my preferences. Someone else may say I'm full of crap, maybe I am, but it's my crap and I live with it, lol.
The Duncan Power Supply Simulator is a nifty tool, check it out.
Re: Power supply component values
Probably easiest to start with an existing PS schematic and see how the voltages come out for your build. Then tweak the resistor values as needed.
I recently wanted to increase the voltage on my PI but not anywhere else. I wanted the PI to clip later. I reduced the value of the power resistor feeding it's node and increased the next power resistor to keep the same total. This doesn't work perfectly and it's guesswork on how much to increase/decrease. I think I went from: >1k>18k>9k>9k to >1k>12k>15k>9k or something close.
I was able to get the PI's voltages up and the rest stayed about the same. Now the PI doesn’t clip until after the output tubes start to clip on the volume dial. Mission accomplished.
I forget if the Duncan PS calculator shows voltages on each node. Those Duncan calculators are fantastic either way.
I recently wanted to increase the voltage on my PI but not anywhere else. I wanted the PI to clip later. I reduced the value of the power resistor feeding it's node and increased the next power resistor to keep the same total. This doesn't work perfectly and it's guesswork on how much to increase/decrease. I think I went from: >1k>18k>9k>9k to >1k>12k>15k>9k or something close.
I was able to get the PI's voltages up and the rest stayed about the same. Now the PI doesn’t clip until after the output tubes start to clip on the volume dial. Mission accomplished.
I forget if the Duncan PS calculator shows voltages on each node. Those Duncan calculators are fantastic either way.
If it says "Vintage" on it, -it isn't.
Re: Power supply component values
Does this come close? http://www.freewebs.com/valvewizard1/smoothing.htmlnee wrote:is there an article somewhere which explains, how the values of the capacitors and resistors in the PS circuit are chosen?
Bascially your dropping resistors determine the voltage at each node, according to Ohm's law, as has been mentioned already. Ideally you want the smoothing caps to smooth all frequencies, at least down to 1Hz, so you want the time constant of each RC filter to be at least 0.16, or in other words,
C=1/(2 pi R)
The larger the dropping resistor, the smaller you can make the cap.
Guitar amps, especially older designs, often scrimp by using rather puny caps. This can lead to motorboating, unless the bass response of the amp is kept down. With modern caps being so small and cheap, there's really no excuse for not having a stiff power supply.
Re: Power supply component values
Thanks, Merlinb, that's exactly the kind of things I was after. BTW that's an excellent site all round for valve guitar amp goodness.Merlinb wrote:Does this come close? http://www.freewebs.com/valvewizard1/smoothing.htmlnee wrote:is there an article somewhere which explains, how the values of the capacitors and resistors in the PS circuit are chosen?
IanG
Re: Power supply component values
A secondary function of the caps and resistors in the dropping string is de-coupling. If I understand this correctly, the voltage swing across the plate load resistor can also cause swing across the dropping resistor, which can cause the next or previous stages to interact.
Re: Power supply component values
Good point Bob-I. That is why it is commonly said to have two gain stages on each node; they will be out of phase with each other and the voltage swing from one will help to cancel out the other. Sometimes this "rule" is followed, sometimes not. It would be a helpful point to keep in mind if one is trying to get away with as low a value capacitance as possible.
Re: Power supply component values
Hey y'all kind of new here and by no means anything more than a beginner as an amp builder. I have built maybe 20 or so amps and many of them just never worked right, but when i got them right I thought they sounded great, for my style of playing.
I consider myself more of a player and do this for a hobby.
As a player that's how I approach anything I build. First and foremost how does it sound and does it let my fingers do what they want to do.
I have noticed one of the major issues with amps and tone and the whole letting my fingers work is the filtering. Especially the choke.
For instance, I built a BF Deluxe clone with a bandmaster choke. For the life of me it sounded like crap and I did not know why, I had built another using the same exact parts except for the choke. So I changed the choke to a BFDR one like the other clone. Well there went hours and hours of messing with this amp all for nothing, it was the choke.
It did exactly that, it choked off the notes. Not in the literal sense but it did not let my fingers work right.
If you want to see an perfect example of what I am talking about with too much filtering and having it just right go to youtube.
Put some Allman brothers on and see the difference in tone from Dickie Betts and Warren Haynes then later Derek Trucks, all great players but you can hear the modern sound and the old sound of Betts.
This is a exactly what I am talking about. Their tone is great but different. I prefer as little filtering as possible but keeping the bottom a little tight. When an amp has the sound I like it breaths, those other sounds do not breath.
If you want to hear breath listen to live albums of Cream, Johnny Winter and other players who use BF fenders and those late 60's early 70's Marshall's. Then listen to the post Van Halen players. Part of the transformation is not only in there pickups, pedals, speakers and the higher gain amps it's the design of those high gain amps in terms of their filtering.
You want to hear an amp breath listen to the John Mayall beano album.
They sing nicely, they are fat sounding with juicy tone but the newer amps don't breath.
If you listen to live cream you will hear an amp breath. Plenty of distortion, sustain but the notes jump out, they don't just stay there and keep on going. They breath, jump out and then die. It makes you work much harder on the guitar. I see the newer player play softly and with great dexterity, the old timers beat the crap out of their guitars, literally to get those sounds, the Who for instance. You don't have to do that anymore.
I prefer to beat my guitar up and work my fingers to the bone. That's when I play best, when it's harder, not easy to get great sustain. As long as it comes when I do beat it...ha ha ha.
Am I making sense to people?
Give me an old MKII 50 watter with a 4-12 greenback cab or a BFSR and I'm happy, even a good BFDR with an old Jensen C12N. Those are my sounds and you have to work a bit to get them to give it up.
So to me I use as little filtering as possible as long as the amp is working right for what I am trying to do with it. More is less in my book.
I consider myself more of a player and do this for a hobby.
As a player that's how I approach anything I build. First and foremost how does it sound and does it let my fingers do what they want to do.
I have noticed one of the major issues with amps and tone and the whole letting my fingers work is the filtering. Especially the choke.
For instance, I built a BF Deluxe clone with a bandmaster choke. For the life of me it sounded like crap and I did not know why, I had built another using the same exact parts except for the choke. So I changed the choke to a BFDR one like the other clone. Well there went hours and hours of messing with this amp all for nothing, it was the choke.
It did exactly that, it choked off the notes. Not in the literal sense but it did not let my fingers work right.
If you want to see an perfect example of what I am talking about with too much filtering and having it just right go to youtube.
Put some Allman brothers on and see the difference in tone from Dickie Betts and Warren Haynes then later Derek Trucks, all great players but you can hear the modern sound and the old sound of Betts.
This is a exactly what I am talking about. Their tone is great but different. I prefer as little filtering as possible but keeping the bottom a little tight. When an amp has the sound I like it breaths, those other sounds do not breath.
If you want to hear breath listen to live albums of Cream, Johnny Winter and other players who use BF fenders and those late 60's early 70's Marshall's. Then listen to the post Van Halen players. Part of the transformation is not only in there pickups, pedals, speakers and the higher gain amps it's the design of those high gain amps in terms of their filtering.
You want to hear an amp breath listen to the John Mayall beano album.
They sing nicely, they are fat sounding with juicy tone but the newer amps don't breath.
If you listen to live cream you will hear an amp breath. Plenty of distortion, sustain but the notes jump out, they don't just stay there and keep on going. They breath, jump out and then die. It makes you work much harder on the guitar. I see the newer player play softly and with great dexterity, the old timers beat the crap out of their guitars, literally to get those sounds, the Who for instance. You don't have to do that anymore.
I prefer to beat my guitar up and work my fingers to the bone. That's when I play best, when it's harder, not easy to get great sustain. As long as it comes when I do beat it...ha ha ha.
Am I making sense to people?
Give me an old MKII 50 watter with a 4-12 greenback cab or a BFSR and I'm happy, even a good BFDR with an old Jensen C12N. Those are my sounds and you have to work a bit to get them to give it up.
So to me I use as little filtering as possible as long as the amp is working right for what I am trying to do with it. More is less in my book.
Re: Power supply component values
Great post, BD!
What about the difference between using a choke and a large resistor?
I just built a Spitfire clone that calls for a choke, but being the cheap one I am...I used a carbon comp 2 watt 220K resistor.
I suspect it might be the culprit behind a problem of frying egg sound when I turn the standby to "Off", but the playing tone is very nice.
What about the difference between using a choke and a large resistor?
I just built a Spitfire clone that calls for a choke, but being the cheap one I am...I used a carbon comp 2 watt 220K resistor.
I suspect it might be the culprit behind a problem of frying egg sound when I turn the standby to "Off", but the playing tone is very nice.
Re: Power supply component values
I don't know. I actually like the choke better in the amps I build which right now are only going to be champ type SE low wattage ones.
Man, a choke is a pretty cheap little part. I'm sure if you asked on this board a few guys will have one laying around to sell you for the cost of shipping.
I guess you mean the 220K resistor to ground. Did you put it before or after the switch?
It's funny by just asking and reading I have learned so much about power supplies this past few weeks. But I hate math so I do it by experimenting.
Man, a choke is a pretty cheap little part. I'm sure if you asked on this board a few guys will have one laying around to sell you for the cost of shipping.
I guess you mean the 220K resistor to ground. Did you put it before or after the switch?
It's funny by just asking and reading I have learned so much about power supplies this past few weeks. But I hate math so I do it by experimenting.
Re: Power supply component values
Yes, 220K between B+ (1) and ground, after the switch.
I just read Randal Aiken's long treatise on chokes, and resistors.
As long as it sounds ok, I'll stay with the resistor.
I just read Randal Aiken's long treatise on chokes, and resistors.
As long as it sounds ok, I'll stay with the resistor.
Re: Power supply component values
Do you have a filter cap before the switch?
Even a .1 Orange drop.....just an idea from what other schematics have.
Thinking about it makes me wonder. Since it's coming through the speakers which side is the problem, before or after the switch, if that is the problem. Since you are basically shutting off the voltage supply I would think the problem is on the other side of the switch that goes to the filter bank. Try removing the 220K????????
Could it actually be the sound of the energy being drained?
I have no idea, I'm thinkin' out loud and I have very limited knowledge. I did have AC DC theory classes. But that was 1980.
Even a .1 Orange drop.....just an idea from what other schematics have.
Thinking about it makes me wonder. Since it's coming through the speakers which side is the problem, before or after the switch, if that is the problem. Since you are basically shutting off the voltage supply I would think the problem is on the other side of the switch that goes to the filter bank. Try removing the 220K????????
Could it actually be the sound of the energy being drained?
I have no idea, I'm thinkin' out loud and I have very limited knowledge. I did have AC DC theory classes. But that was 1980.