Help with low ESR numbers, please.

[gwozdz]

I've read that you have to consult the data sheets for low ESR. When I get there, what do I look for if it doesn't explicitly say low ESR. Do I look for an impeadance number or what ? How does one determine below what number is considered low ESR? And if it differs for each #uF, is there a list for 10uf, 22uf, 100uf, 220uF, 470uf, 1000uf, 1500uf, 2200uf, and 3300uf or is it the same cut-off number for all? Thanks

[sofTest]

Re: Help with low ESR numbers, please.

There is not a "standard" for what is considered low ESR, because there are different grades of low ESR, and that again is not consistent between the manufacturers. In datasheets, if ESR is not given, often the capacitor model is not low ESR. ESR is not the same as impedance, but if ESR is not given, impedance will give you in indication of ESR. For some capacitors, the two values are even the same, and listed in the same column. There are people in this forum who can explain the difference.

If you are going to replace capacitors, the safe thing to do is to use a list of manufacturers and models of known good capacitors, and select from this list a capacitor with same or better ESR than the one you are going to replace. In this thread, you can find a list of ranked capacitors from one of the forum members.

For a given manufacturer and capacitor model, the ESR value will be the same for the same physical size, independent of volt and uF. In some datasheets you will find that ESR is only given for the physical sizes of the capacitors, not for for individual values.

Hope this answers your questions.

[Toasty]

Re: Help with low ESR numbers, please.

<<<======== Patiently awaits arrival of cymbal playing penguin...

...just wait, he'll be here soon.

[PCBONEZ]

Re: Help with low ESR numbers, please.

<<<======== Patiently awaits arrival of cymbal playing penguin...

...just wait, he'll be here soon.

Why don't you just help the guy???
I'm here a lot lately but there's times when I'm gone for weeks and I would never see this.

~~~~~

We use "Impedance" out of the data tables and call just it ESR.

That's not 100% correct it's just that most of us are too lazy to do the math and just using that number is like pressing the "Easy Button".
That and the fact that after doing the math once or twice one realizes the Impedance number in the table is so close to the actual ESR that doing the math is a waste of time.

So now I'm going to do the math. Because I'm feeling masochistic.

Well, no. I just want to explain.

ESR + Xc make up most of the "Impedance" shown in the table.
-
The 'standard' condition to used in the data tables for Impedance is 100kHz and 20°C.
-
At 100kHz Xc is so small as to be ignored.
-

Xc = 1 / ( 2 pi f C )

So at 100kHz (100,000 Hz)

... with a 1000uF (.0010 F) cap:
Xc = 1/ [2(x)3.14(x)100,000Hz(x).0010F] = 0.0016 ohms

Not big enough to worry about and it gets even smaller as uF goes up.

... with a 3300uF (.0033 F) cap:
Xc = 1/ [2(x)3.14(x)100,000Hz(x).0033F] = 0.00048 ohms
-

~~~~
And that's why we can just use the Impedance shown in the tables for ESR.
~~~~

Units in the tables are usually:
mΩ at 20°C and 100kHz -or- Ω at 20°C and 100kHz
.

[Toasty]

Re: Help with low ESR numbers, please.

>>Why don't you just help the guy???<<
'cuz I ain't got yer Silver Tongue when it comes to this bit...

YET

[PCBONEZ]

Re: Help with low ESR numbers, please.

O I C

[gwozdz]

Re: Help with low ESR numbers, please.

Thanks All, John