0.22f 50v Capacitors

[kkkittehz]

Hi everyone,

As part of a circuit im wanting to make I need to use 6 0.22f 50v electrolytic Capacitors to achieve 1.2F of total electrolytic capacitor power (so to speak).

The problem is that i cant find these capacitors ANYWHERE. any thoughts?

The Circuit is

and his completed project with these mammoth capacitors is

for the complete 'guide' its Jean Hiraga's Super Class-A Amplifier

ANY help would be appreciated

 

[Nigel Goodwin]

I would imagine it's VERY difficult (and EXTREMELY expensive) to find capacitors that large.

I would also suggest it's totally pointless, the best possible quality amps don't have that much capacitance in the PSU.

If you insist, instead of 220,000uF, you could parallel 10x22,000uF - but even those won't be that easy to find.

 

[JimB]

Here is a 0.27F 63v capacitor at Farnell

UNITED CHEMI-CON|E36D630HPN274MEM9M|Aluminum Electrolytic Capacitor | Farnell United Kingdom

If you must use a sledge hammer to crack a nut.

It would make more sense on many levels to use a stabilised psu.

JimB

 

[kkkittehz]

I think ive found the capacitors i need, 220mf 50v is the same as 220000uf and is the 0.22 farad i was looking for, the only thing that worries me is that the site it is from is selling six for $1.50?!

**broken link removed**

if the capacitors are what they say they are then ill get started right away

 

[smanches]

No, those are 22uF. They are mis-labeled. Have to be really careful when looking at caps, as people often use mF to mean uF. It's an old style of rating.

In fact, if you were to take it literally, those would be 22 Mega-Farad capacitors.

 

[kkkittehz]

such a shame, why cant people stick to standard units of measurement? The Hunt Continues!

 

[smanches]

Just so you have a better understanding of what you are looking for, a 220000uF cap will be the size of a soda can or bigger. This is starting to get near the super-cap size of things. Although I agree with Nigel that this seems like pretty massive overkill.

 

[Nigel Goodwin]

Just so you have a better understanding of what you are looking for, a 220000uF cap will be the size of a soda can or bigger.

I would imagine bigger?, particularly as it's going to require a very big ripple current rating.

 

[Hero999]

The capacitor in Farnell is the size of a soda can - look at the dimensions!

It's a 130W amplifier (assuming a 4R load), so at total capacitance of 0.66F is overkill, perhaps the designer got the decimal place wrong?

A single 22,000µF capacitor per rail should be more than enough - I'd go for 10,000µF.

 

[Nigel Goodwin]

The capacitor in Farnell is the size of a soda can - look at the dimensions!

It's a 130W amplifier (assuming a 4R load), so at total capacitance of 0.66F is overkill, perhaps the designer got the decimal place wrong?

A single 22,000µF capacitor per rail should be more than enough - I'd go for 10,000µF.

Did you hit a key by accident?, it's a 30W ampl;ifier, not a 130W one.

 

[Hero999]

No, I used Ohm's law.

I assumed a peak output of 32V and with a 4R load that will be 128W which I rounded to 130W, if it's an 8R load then that can be halved.

Maybe I got the peak outout voltage wrong, perhaps it'll be more like 30V? Even so it'll be 110W into 4R or 55W into 8R

A 30W amplifier driving an 8R load only needs a peak output voltage of 22V.

 

[mneary]

This assumes that the output stage remains listenable all the way to clipping. The peak voltage into 4Ω would be significantly lower due to the 0.5Ω emitter resistors, IR losses in the mains transformer and diodes.

Anyway, interesting priorities, 128W standby for 55W output. (7.1 channel system would be >1kW standby.)

 

[Nigel Goodwin]

No, I used Ohm's law.

I assumed a peak output of 32V and with a 4R load that will be 128W which I rounded to 130W, if it's an 8R load then that can be halved.

Maybe I got the peak outout voltage wrong, perhaps it'll be more like 30V? Even so it'll be 110W into 4R or 55W into 8R

A 30W amplifier driving an 8R load only needs a peak output voltage of 22V.

You failed to notice it's a class A amplifier, you can't treat it as a class AB one. In any case, 70V HT rail gives you around 100W in to 4 ohms, not 130W - overdriving the amplifier to produce 130W would be grossly distorted, and the amp wouldn't last long.

 

[Hero999]

You failed to notice it's a class A amplifier, you can't treat it as a class AB one. In any case,

Yes it says class A at the bottom.

It seems silly to build such a large class A amplifier

70V HT rail gives you around 100W in to 4 ohms, not 130W - overdriving the amplifier to produce 130W would be grossly distorted, and the amp wouldn't last long.

You mean +/- 35V rails?

It depends on how much voltage loss the output transistors have and how low the distortion is, for 100W into 4R you need 28.3V peak and at 35V a 6.7V loss sounds reasonable for low distortion.

 

[Nigel Goodwin]

Yes it says class A at the bottom.

It seems silly to build such a large class A amplifier

I agree, class A is VERY inefficient, it's a lot of work for only 30W - and a LOT of wasted heat.

You mean +/- 35V rails?

If using spilt rails, or a single rail of 70V.

It depends on how much voltage loss the output transistors have and how low the distortion is, for 100W into 4R you need 28.3V peak and at 35V a 6.7V loss sounds reasonable for low distortion.

You're ignoring the drop in the HT rails under load - I've repaired, designed, and built many amplifiers over the decades - 65/70V gives you a reasonably decent quality 100W.

 

[Hero999]

What voltage drop would you typically allow for loading?

Surely it depends on filter capacitors and the impedance of the transformer?

I imagine it'll be more of a problem for bass frequencies.

 

[Nigel Goodwin]

What voltage drop would you typically allow for loading?

Surely it depends on filter capacitors and the impedance of the transformer?

And on the voltage drop across the rectifiers (which increases with current). But mostly on limitations of the amplifier itself, to get as close to the rails as possible you require extra higher voltage rails to feed the drivers - and this is quite often done.

I imagine it'll be more of a problem for bass frequencies.

VERY, VERY much so. One reason power output is commonly specified at 1KHz - but a decent amp will specify RMS power from 20Hz to 20KHz (or so) as well.