Filter capacitor values

[telephonebills]

Hi,

I'm hoping for some clarification on the subject of filter caps. I opened up a Silvertone 1484 head that I bought 2 years ago and saw that while it had been recapped fairly recently, with nice looking F&T caps, the values of the new filter caps are way off from the originals. The schematic shows four 100uf 150v caps, and what I have is four 47uf 500v caps. I am just learning about tube amps and definitely not clear on how important cap values are in this stage. I assume that the tech who put them in knew what they were doing, but I would appreciate it if someone could explain why such a difference in values is okay.

Thanks in advance.

F

[52 Bill]

If the 100uF at 150v caps were wired in series, they would become the equivalent of a 50uF at 300v cap.

Post the correct schematic for your amp and we can see how they were wired.

[telephonebills]

Thanks for the reply. Here is a schematic.

[52 Bill]

Okay, those 100uF at 150 volt cans are part of a voltage doubler circuit. In fact the amp has two stacked doublers to get the 480 vdc for the power amp. The 47uF value might be a bit low for that usage.

How well does the amp work? Any hum?

[soundmasterg]

The caps that are in there now will function, but not as well as something different. Every time you use a voltage doubler, the cap value used is essentially half of its rated value...so since there are two doublers stacked on top of each other in these amps, the stock value was basically the same as a 25 uF, and what is in there now is the same as a 11.75 uF. I would recommend to go with a cap that is at least the same as the originals for value and voltage, and going larger would be of some benefit because voltage doublers don't regulate as well as bridge or full wave rectifier setups, so they will hum more. If you went with 220 uF 150v (or higher on the voltage) caps, then you would have the equivilent of 55 uF caps for that first section, which will give less overall hum and give a tighter bottom end.

You want to make sure not to go overly large on the bias supply cap because that essentially slows down the bias supply too much to where it doesn't track changes in the amp as quick as it should. No larger than 100uF for the bias supply cap. For the other filter caps, if there is no tube rectifier, you can go quite large without any problems, but as you go very large (1000 uF for example) then the power transformer could heat up excessively because it has to supply a much larger cap than it was designed for, and the sound of the amp can suffer as it gets too clinical and too quick. For most tube amps that aren't hi-fi or bass amps, or don't have tube rectifiers, then 220uF in the first stage is probably as high as you would want to go. Since yours uses two doublers stacked on top of each other, you could then go with 880 uF caps if you wanted to reach that value, but I think the 220 uF would be fine for this application, and would also fit in the spot better than larger value caps would.

Greg

[telephonebills]

Thanks very much for your replies.

Surprisingly, hum has not been an issue with this amp. My reason for opening it up and checking things out is that I've been unhappy about the tone when the volume gets up to 40% or higher. The dynamic range gets squashed and I get really flabby, farty clipping, mostly in the low end. It sounds like these caps probably have something to do with that. The bypass caps are original so I'll change those too. All other caps seem to be brown chocolate drop looking things. I'm not sure if it's standard policy to swap those things out. Would be a lot of work given the space in that chassis.

Greg,
Regarding filter caps, would you recommend definitely going higher than 100uf? I have sourced out some Sprague 100uf 350v caps from a local seller. I haven't pulled the trigger on that order yet so I can still get 220uf F&T caps if that would be best for this amp. Remember, hum wasn't really an issue with the 47uf caps so I'm not that worried about noise. I'm really looking for tone improvement.

Which reminds me. There is also an original cap-can there with 20-10-5 450v. Could this be screwing with potential tone?

Thanks again for helping me with this!

F

[Bruce / Mission Amps]

If they are the red Planet caps.... get rid of them. One year warranty from the factory? ha ha
Just use standard, small and inexpensive radial (or axial) 100uF@ 150v-250v caps, but then put a larger value 47uF@500v across the end of the B+ rail before the center tap of the output tranny.
There is spot to glue one down right in front of the pots along side the vertical Masonite eyelet board.
Been doing this for years and it works fine.... Jack White doesn't seem to mind it.

[telephonebills]

If they are the red Planet caps.... get rid of them. One year warranty from the factory? ha ha
Just use standard, small and inexpensive radial (or axial) 100uF@ 150v-250v caps, but then put a larger value 47uF@500v across the end of the B+ rail before the center tap of the output tranny.
There is spot to glue one down right in front of the pots along side the vertical Masonite eyelet board.
Been doing this for years and it works fine.... Jack White doesn't seem to mind it.

Thank you for the info Bruce. Just to clarify, you mean installing four 100uf caps as per spec, and then adding a fifth cap at 47uf? I'm confused by the reference to the front of the pots. The pots are on the other side of the chassis. I'll try to post a pic of the guts and maybe you could point out the spot.

Thanks again.
F

[soundmasterg]

Thank you for the info Bruce. Just to clarify, you mean installing four 100uf caps as per spec, and then adding a fifth cap at 47uf? I'm confused by the reference to the front of the pots. The pots are on the other side of the chassis. I'll try to post a pic of the guts and maybe you could point out the spot.

Thanks again.
F

What Bruce is talking about is if you look at the schematic, which I posted, on the right side you will see the power supply area where the four 100uF caps are, and you will see that they connect together and then go up to the center tap of the output transformer. The 47uF cap should be connected to that point before the center tap of the output transformer and to ground. As far as where he says to connect it physically, I'd have to look at my layout drawing or open up my 1484. Maybe he can explain a little further about where it physically is inside the amp. The 100UF values are fine if you go with Bruce's suggestion of adding the 47uF cap. The 220uF solution works too and gives a little better regulation....the bottom end will be a little tighter and quicker, though these amps have so many issues that one thing will not fix everything.

You can increase power about 5 watts RMS if you add another RC filter stage in and run the screen to the top point on those summed cap nodes, but if you do this, you really need to add a voltage doubler to the bias supply and make it adjustable as otherwise the tubes will run even hotter than they do now. You also run the risk over time of blowing the super small output transformer, though having said that, my brother's is still going strong after running this way for 6 years and being mismatched on the output from time to time. My brother's is 30 watts RMS and a stock one is 25 watts RMS. Running the screens at close to the plate makes the difference in power. If the output transformer was changed to a bigger one more power gains could be realized and the amp would sound better too, but some of the Silvertone character would be lost.

The 20/10/5 can cap should be replaced also, but there are no made to fit replacements since Vibroworld closed up 7 years ago or whatever it was. I bought the last two of those can caps that he was custom making and stuck one in my 1484 and another in my brother's. You could use a custom Weber can and even order it with separate grounds, but you would have to do some metal work to get it to fit to the chassis correctly. I don't like the custom Weber caps too much myself as their glue joint to their vulcanized fiber base is just super glued and isn't reinforced at all and they tend to break at that point. They can be re-glued easily enough but its a hassle. Sometimes you have space concerns and need to use them though. You could use a two section can cap like a F & T or JJ and then wire up the third section inside the chassis, or you could also just disconnect the stock cap and leave it in place for cosmetic reasons and use modern small caps inside the chassis too. If you decide to go a two section cap can route, then I would just get a 16/16 cap...its close enough....and then use a 20uF discrete cap for the screen section if you want to stay stock. You could also use a 4 section can cap and use the higher section for the screen then the small sections for the preamp and phase inverter. With either of those can cap options you have to do metal work to get the can cap to fit in place of the other one. I would use discrete radials myself if I was going to replace that can cap, but I would make sure they were secured well to withstand abuse. I wouldn't bother with Sprague either...they are way overpriced for what they are. The Weber axial caps are good and cheap. Nichicon are also good, Panasonic are consistently good and can be had from Mouser for not too much.

Hum isn't an issue with these amps in stock config because they are pretty low gain. You can resize cathode resistors and play with the voltage supplies (R58, R59) for the preamp stages, but as you increase gain and voltage the amp really starts to hum, even with tricks like elevating the filament center tap, because the layout of the amp is really poor. Putting the power transformer in the middle of the amp right behind the inputs to channel 2 is a stupid place for it. When I was messing around with my brother's years ago it was really humming a lot as I increased gain, and sticking some mu-metal in between the power transformer and the inputs very carefully made the amp very quiet, but it wasn't something that I could leave in there. It did help to identify the problem though. (do a search under my name for Silvertone 1484 posts and you will find a lot of info from like 5 years ago) Going towards typical Blackface Fender values in the preamp can get you more gain and a brighter tone, but again you lost some of what makes the Silvertone a Silvertone.

The resistors are all carbon comp stock, and they were 20% from new...so pretty lousy. My 1484 still needs those changed as it oscillates so something has drifted. I checked a couple and they were way off. The brown caps are derisively called "brown turds" by many people. As a rule they almost always function, but from new their ESR was high and they aren't particularly good caps. They were used in a lot of Fenders and Sunns too. Pretty much any modern cap will sound better, though I would stick with polyester caps rather than polypropylene unless you want more high end clarity. Orange Drop 225P or the PS series work fine, as do Mallory 150, the Antique Electronics Supply generic polyesters, Sozos, etc. Changing any of those resistors and caps is a pain not only due to the small space in the chassis, but also because the people who wired these amps up wrapped the leads around the terminals in a loop. Its a good and robust way to make an amp, but it takes a lot of effort, and is a total PITA when it is time to change the parts. I have no idea how Danelectro made money when they made these for Sears back in the day.

There is a lot you can do to these amps because there is a lot of room for improvement in them. Anything you do is a lot of work though due to the layout and the cheap construction, and the more mods you make to the circuit...increasing gain....getting more x, y, or z, the less it sounds like the Silvertone it is. My brother's was missing the reverb transformer and was otherwise hacked up before he got it so it was never going to be worth anything and was a good candidate for modding. The things I did to his I wouldn't do to every one of these amps, and in fact when I see these for repairs, I do the least amount to get them working well and sounding good. I much prefer to work on amps that are made better these days.

Greg