DRRI Power Supply Capacitors

[printer2]

Anybody know why the DRRI has two caps in series for the power supply, a 220uF 100V and a 47uF 500V?

[img:4609:3080]http://griffineffects.webng.com/images/ ... page_1.png[/img]

[Firestorm]

Gotta assume it's cheaper.

[printer2]

Gotta assume it's cheaper.

Doubt it, in that case just leave out the 100V capacitor as the 47uF is rated for 500V.

[Mescalero]

The resistors serve to accomodate each capacitor's voltage rating.
Edited this post.

Mea culpa - see post below.

[Stevem]

I guess it's not truly a reissue when as they show on the schematic that they use that 220uf and that 47 uf cap to end up with some 123uf.
Not that that's bad, as I would rather have that much filtering there if it was my amp anyway!

[martin manning]

Those two caps in series are going to make 39uF at 600V.

[David Root]

Maybe they had the 220uF/100V already, or got them at a fire sale price, and they would probably have the 47uF/500V in stock anyway.

Even though 600V caps are now relatively reasonable $$, especially in Fender quantities! justradios.com has 40uF/600V at $8 each, maybe $4 to Fender!

Still too much probably, even with the extra labor and resistors cost of two balancing resistors.

[printer2]

I have not heard the amps running over 500V so the one cap should work just fine. I was thinking it might effect the sag characteristic if you had a larger cap and a smaller one in series, but in that case I would think the larger value cap should carry the majority of the voltage. I am having trouble seeing the reasoning behind it.

[printer2]

Asked at another site and tha answer makes perfectly good sense. The original DR had a filter capacitance of 32uF in that spot. Guess what value you get when you put a 68uF and a 100uF in series? Bet you its cheaper than sourcing a 32UF cap.

[Mescalero]

My mistake! I should know better. Caps in series do not add together. Instead, it is a "reciprocal of the sum of the reciprocals" thing. The voltage withstanding is increased. The total capacitance is within a 5AR4's capabilities. Boy, do I feel silly.

[Stevem]

Next time a screw up that cap in series formula somebody kick me in the head please!
Thanks!

[martin manning]

I have not heard the amps running over 500V so the one cap should work just fine. I was thinking it might effect the sag characteristic if you had a larger cap and a smaller one in series, but in that case I would think the larger value cap should carry the majority of the voltage. I am having trouble seeing the reasoning behind it.

For caps in series, the voltage on each cap is inversely proportional to the capacitance. Look at the values of the "balance" resistors and you will see evidence of that. 33uF is a common value, so this was likely the cheapest way to get to a 600V rating. With only 400V on the plates that might seem a little excessive, but with the standby switch next to the choke there will be some large amplitude swings during transients.

[didit]

I have not heard the amps running over 500V so the one cap should work just fine. I was thinking it might effect the sag characteristic if you had a larger cap and a smaller one in series, but in that case I would think the larger value cap should carry the majority of the voltage. I am having trouble seeing the reasoning behind it.

For caps in series, the voltage on each cap is inversely proportional to the capacitance. Look at the values of the "balance" resistors and you will see evidence of that. 33uF is a common value, so this was likely the cheapest way to get to a 600V rating. With only 400V on the plates that might seem a little excessive, but with the standby switch next to the choke there will be some large amplitude swings during transients.

Was given pause (a long one it turned out) when first considering this unique arrangement of stacking power filter caps. Struck me intuitively that it improved reliability and provided better filtering. Last weekend I finally found/took some time to pencil it out. My back-of-envelope modelling is based on more than simple capacitance spec, and in particular involves looking at ESR & ESL. In both DC & AC steady state the circuit is equal to a single tall cap or two in-series of equal size. During transients is where possible benefit kicks in. If one considers where the voltage and coulombs will be in steady state and then does a bit of math on in-rush and large load transients the bigger cap at the top takes/gives more and mixing in ESR/ESL from datasheet on common commercial caps (I looked at F&T) has better specs to do that well. That plus what you'd already noted with max of 600V the circuit has 1.5x transient voltage spike tolerance leads me to think someone at Fender was thinking positively about reliability etc.

Best .. Ian

[Hepcat]

It seems to me that not long ago (before internet sourcing of parts and resurgence of tube circuits) it was difficult and expensive to get 450 and 500 volt electrolytic capacitors. If you looked in the electronic magazine ads for parts, you found tons of "photoflash caps" for sale on the surplus market. The James Electronics ads had them and now they are Jameco. Photoflash was the storage caps in the xenon strobe flashes for photographers. They were much cheaper than the usual 450 and 500 volt electrolytic caps that we needed to repair and build amps with. To use them, you just put two in series. the halving of the capacitance made it possible to use them in tube amps with a tube rectifier. Time frame is the late 70s through the 80s. Or at least that is what I remember.