Are Swedish filter caps more sexy?

[overtone]

Even with a lower uF value, I have found that the Rifa PEG124 filter caps perform much better than modern no-name radial filter caps. So I consider using these for old tube rectified amps where you have a limitation on your first filter cap uF value after the rectifier.

The Panasonic EE series have an even higher ripple current rating and are smaller than the Rifa PEG types, but not so sexy looking.

Is the ripple current rating more important for filtering than the uF value? Or is there a kind of double bind that I am unaware of? Is there a point where it is just overkill?

[jmaf]

Even with a lower uF value, I have found that the Rifa PEG124 filter caps perform much better than modern no-name radial filter caps. So I consider using these for old tube rectified amps where you have a limitation on your first filter cap uF value after the rectifier.

The Panasonic EE series have an even higher ripple current rating and are smaller than the Rifa PEG types, but not so sexy looking.

Is the ripple current rating more important for filtering than the uF value? Or is there a kind of double bind that I am unaware of? Is there a point where it is just overkill?

The ripple current rating is probably more about reliability than anything else. Low ESR caps will heat less, will allow more AC current without generating as much heat. There are tons of little details in designing a tube amp power supply, so perhaps you should have a look at Merlin's latest book

Designing Power Supplies for Valve Amplifiers by Merlin Blencowe in Engineering

Ps. I've seen sexier things than capacitors coming out of Sweden

[overtone]

Thanks jmaf for the pointer, I have the book, took it up on the roof and page 20 helped. So far I have taken on board:

1. the first cap is a reservoir cap
2. the size of my reservoir has a capacity, measured in uF - the capacitance
3. the reservoir gets topped up in pulses from the rectifier as necessary
4. those pulses have to charge (re-fill) the reservoir quicker than it is being emptied
5. if I make my reservoir too big the rectifier cannot keep up, but I think this must also be dependent on the load drawn more than reservoir size? There is some allusion to this on page 44
6. those pulses are called the ripple current and they can be 14x the load current
7. somewhere in this process heat is generated inside the cap, so ideally, for a long happy life, the capacitor has to be able to handle that ripple current. Higher ripple current capacity = less heat inside = longer life

Does this make sense so far?

[diagrammatiks]

can electrolytic caps be sexy?

[jmaf]

Thanks jmaf for the pointer, I have the book, took it up on the roof and page 20 helped. So far I have taken on board:

1. the first cap is a reservoir cap
2. the size of my reservoir has a capacity, measured in uF - the capacitance
3. the reservoir gets topped up in pulses from the rectifier as necessary
4. those pulses have to charge (re-fill) the reservoir quicker than it is being emptied
5. if I make my reservoir too big the rectifier cannot keep up, but I think this must also be dependent on the load drawn more than reservoir size? There is some allusion to this on page 44
6. those pulses are called the ripple current and they can be 14x the load current
7. somewhere in this process heat is generated inside the cap, so ideally, for a long happy life, the capacitor has to be able to handle that ripple current. Higher ripple current capacity = less heat inside = longer life

Does this make sense so far?

1) Correct. The first cap stores charge for the entire amp, especially the power tubes.
2) Yes.
3) Yes.
4) Ideally, yes.
5) Tube rectifiers are limited to a certain amount of current and most need protection. Solid state rectifiers are made in countless sizes, you can find any level of voltage x current you may need. Correctly sized SSR's can handle any size reservoir cap.
6) As you load the capacitor, a current I is requested. The ripple voltage is V = I / 2fC where f is the frequency of the pulses, C is the size of the cap in Farads. I is in Amps. As you can see from that formula, the ripple voltage depends on the current being drawn. If there is zero load, the ripple does not exist as long as the capacitor is good.
7) Heat only exists if there is ohmic resistance. Reactance does not create heat. If the capacitor leads and internal construction offer resistance to the passage of the pulses, it'll produce heat. In practice, if a capacitor is anywhere near warm, it's already gone. Caps don't heat up in any sensible form in normal conditions, they work at cabinet temperature.

[overtone]

Got it. Thank you.

Merlin's worked through examples from page 54 onwards will help me to select reservoir values and check the ripple current.
Would a ripple voltage of 10% of the supply voltage be a good ball park figure to aim for?
Perhaps that should be even lower for an SE amp?

Have a good Easter all.

[jmaf]

Got it. Thank you.

Merlin's worked through examples from page 54 onwards will help me to select reservoir values and check the ripple current.
Would a ripple voltage of 10% of the supply voltage be a good ball park figure to aim for?
Perhaps that should be even lower for an SE amp?

Have a good Easter all.

For an SE amp you should aim for less than 7%. At least that's what I've had on top of my mind for a long time now, don't know if less is required these days. Good luck with the project, and a happy Easter for you as well.

[mooreamps]

On SE amps, I believe we can tolerate a few volts of ripple on the plate. It's the few volts of ripple on the screens that would be more noticeable...Look at some of the old vintage Fender amps and see how they were set-up...

-g

[olddawg]

can electrolytic caps be sexy?

Only if they are made with fish oil .

[jmaf]

Only if they are made with fish oil .

It don't get any sexier than that. But only caps made in Sweden with Swedish fish oil.

[Steve Conner]

Oil is the wrong stuff for electrolytic caps. They use lutefisk

Rifa make excellent capacitors. I've used the bigger screw terminal ones for solid-state amps and other power electronic type projects, including some really nasty stuff with maybe 50A RMS of ripple. The extremely low ESR means they will vaporise your circuit instantly in the event of a short, I have seen pieces of transistor flung over 10 feet. IMO, that frisson of danger makes them extra sexy.

As for the small axial ones, I've only used one or two. Still working after 10+ years, but so are other cheap no-brand caps in the same equipment. I would prefer them to Sprague Atoms and the like any day.