uneven filter caps

[HiGain]

Most of the time I see the B+ electrolytic filter caps as being equal in value, e.g., two 16uf, two 20uF, two 40uF, etc.

Is there a reason for this? And, is it OK to use uneven caps, such as a 20uF and a 30uF electro cap?

Thanks guys,

Jake

[ApexJr.]

If this helps I sell quite a few of these caps to amp builders and there
not prime values. I've been selling 22 Uf 450V and 33 Uf 450V
So it shouldn't make a difference.


Steve @ Apex Jr.

[stoo]

I bought a 50/50 uf 500v and a 30/30 uf 500v. I used one 50 and a 30 (80 total) for the B+. The other 50 for the screens and the last 30 for the preamps .. NO problems.
stew

[mlp-mx6]

OK, there is a BIG difference in how you pair the caps. If you stack them, they should be equal in value (excepting for tolerance). (The voltage equalizing resistors should be equal in value also.) If you parallel them, any combination is fine.

[gearhead]

OK, there is a BIG difference in how you pair the caps. If you stack them, they should be equal in value (excepting for tolerance). (The voltage equalizing resistors should be equal in value also.) If you parallel them, any combination is fine.

Why, what's wrong with stacking (I assume you mean in series) caps of different values? From a uf value standpoint, it's just (x*y)/(x+y), as in with parallel resistors. Is there some parasitic thingie problem?

[mlp-mx6]

This is due to the currents being delivered, and the purpose of the balancing resistors - indeed, the purpose for stacking caps in a power supply. Usually this is done because the supply voltage exceeds the rating for one cap. The purpose of the resistors, then, is to balance the voltage across the 2 caps so that they see approximately equal differentials. When one caps is larger it is harder to maintain that balance across the resistors, due to the current differentials to/from the caps. Perhaps if the resistors were scaled similarly (but inversely) it could work. I must admit I've never tried that.