Al electro's vs polymer?

Oh, 15V supply GH

Al poly has way lower ESR for the same capacitance, and ought to last longer at highish temperature because there's nothing to dry out.

Usefully, Alpos come in all sorts of grades of ESR, so you can find one that will work with your regulator. They do cost more for a given value and voltage, but the ESR improvement may let you use a smaller one.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
ElectroOptical Innovations LLC / Hobbs ElectroOptics 
Optics, Electro-optics, Photonics, Analog Electronics 
Briarcliff Manor NY 10510 

http://electrooptical.net 
http://hobbs-eo.com

Tantalums are dangerous across power supply rails.

Polymers have very low ESR. That may or may not matter in your filter. Wet aluminums can be had with bigger capacitance values.

You could get voltage overshoot from the LC combination, so make sure the polymer caps are rated for at at least 2x the supply voltage. I've tested some that die hard at 1.5x rated voltage. Panasonic?

The United Chem-Com parts are great. At way over rated voltage, and a bunch of reverse voltage too.

Thanks Phil, I guess I'll have to order some of both. I might need a bit of ESR. The filter looks to have a Q of about 3... so ~ 1 ohm of resistance... I think shared between L and C. A lower Q would be nice. (maybe a bigger C.)

George H.

They seem to be getting worse with time. I was looking at a circuit we made ~20+ years ago and there are 16V tant's on all the

+/-15V rails... Now I've changed to all 50V tants on 15V rails.

George H.

Tantalums are just erratic. Some batches are bulletproof, some explode, with no apparent pattern.

3x voltage derating seems safe for tants. I like them on the outputs of LM317 and LM1117 regulators, which don't like all-ceramic loads.

(Unless you do my trick.)

Both are available in a wide range of ESR. The range for AlPo is generally lower, but there is plenty of overlap. Just shop for the appropriate part.

Tim

-- Seven Transistor Labs, LLC Electrical Engineering Consultation and Design Website:

Thanks Tim, I ended up just searching for all the Al electro's we have in stock. A bigger 100uF 50V gave a nice 'output'*, and fit! ESR was less than the tant, but still less ringing at the lower freq. Better and better, no one but me will really care :^)

George H.

*I'm filtering this commercial SMPS +/-15V at ~1A.. though I don't need all that current. The positive supply is worse than the negative, in that it has a ~60Hz ripple and (I assume) the negative supply is derived from the positive and clean there.

Oh, so here's a thing. Does anyone else look at the spectrum of some signal by triggering their DSO (normal trig.) up near the maximum voltage of the signal, and averaging?** This gives outputs that are 'almost' the auto-correlation function. And if you press your DSO FFT button, you get displays that are 'almost' the FFT... (the 'almost' is the FFT falls off faster at high freq. so the 3db point moves to 6db... that's for a low pass.)

GH

**(if you trigger in the middle of the signal, and average, you get a much smaller thing, where V(t=0) = 0)

For a Band Pass, higher Q filter it means the ringing happens 'sooner' in the time domain and I don't know what the scale factor is for the slopes, (probably sqrt(2) in power. 2dB?)

Why not just stick with aluminum electrolytic? The regular old not-so-low-ESR caps are fine for lots of purposes, and if you don't like 'em failing with age, just use two or three. A filter inductor seems to indicate you don't want to support high surges of current anyhow, so adding a current limiter (even if it's a fuse) seems more productive than looking for exotic components.

------------------------

** Huh ?

Doing that makes them last longer ??

On which planet?

.... Phil

I checked a Panasonic 56u 25V polymer. It's fine at +40 and -10 volts.

On this one. Ripple current is spread across the 3 caps, resulting in lower temperature and longer life.

** Well, that begs the question ....

** It's too small to matter.

** Nope.

FYI: EVERY case of early electro failure I have seen, the cause was high ambient temp, not ripple current.

Dopey PCB designers placing them too close to hot power semis and resistors.

.... Phil

If you need 500 uF, just use three 330 uF units. The usual failure mode being electrolyte loss/drying out, would leave the failed unit an open circuit, so the others cover for it.

I've done some troubleshooting on old machines (Mac IIci, if anyone recalls those) where seal failure could take out about a third of the distributed filter caps without anyone noticing (unless the electrolyte goo ate through a trace).

Interesting,

I had a circuit with a link to enable a -ve 5V supply, due to an oversight it got pulled up to +2.5V when the link was missing.

The Kemet 100uF 10V caps on the -ve supply (taht had seen the reverse voltage) all failed between a few minutes and a few hours after the application of the correct polarity supply. The really worrying thing is that some of them looked OK for an hour or two.

MK

--
This email has been checked for viruses by AVG. 
https://www.avg.com

Here are some 25V Panasonics run at -10 volts for a month or so. Total leakage was 46 uA after things settled down (DVM shows drop across a

1K resistor.) There is a bit of "forming", or maybe massive DA, like a regular lytic, so initial leakage was higher.

Declining leakage is a good sign!

I once had a 16V tantalum bead reverse connected to a 15V power supply. It lasted a few weeks before it shorted the power supply (capable of a few hundred mA maximum. There were no pyrotechnics fortunately.

John

I once had a 16V tantalum bead reverse connected to a 15V power supply. It lasted a few weeks before it shorted the power supply (capable of a few hundred mA maximum. There were no pyrotechnics fortunately. John