Quick ESR answer needed

Ceramic, I say in the 10m to 50m range.

--
Kevin Aylward
ka@kevinaylward.co.uk

What _kind_ of ceramic?? Kemet's data sheet for ceramic SMT caps suggests that the ESR of X5R caps in the 0.1uF to 1.0uF range is under

0.01 ohms. That applies to 0603 through 1206 sizes. But the ESR varies with dielectric in their data sheet, and I'm sure it varies with manufacturer. Kemet, for example, suggests an ESR in the 0.1 ohm region for Z5U and Y5V parts. "YMMV" What I say, bottom line, is start with manufacturers' data sheets for the particular parts in question, and if you have reason to suspect they aren't accurate, MEASURE it.

Cheers, Tom

The low-uF ceramic chips with X7R dielectric often have an ESR of 0.02 ohms or so in the 50kHz to 500kHz range.

Some 100uF 6.3V Y5V ceramic chips spec as low as as 0.002 ohms over the same frequency range.

A couple of the LDO regulator app notes warn that the devices may not be stable with low-ESR caps on their output and in fact show resistors in series with such caps!

Tim.

Much lower IME, more in the range Kevin gave but it depends on the frequency. I guess your client would want a somewhat more "official" story about it, so here goes (the MLC versus Tantalum shoot-out):

Ahem, you wouldn't think about counting on the ESR for a loop design, would ya? No, can't be ....

--
Regards, Joerg

http://www.analogconsultants.com

Sounds about right for a Z5U dielectric low volt cap. These things are the pits. Terrible temp and voltage curves with lots of microphonics and absorbtion. All they are 'good' for is power filtering at the vcc terminals of ICs.

Bet he's designing a chip with built-in LDO!

I like the National AnyCap idea, an internal dominant pole that's merely increased by external capacitance, not cascaded with it. Basically, it's Miller capacitance off the output pin.

John

Sounds like as much fun as repairing a dry-rot problem.

I think that's Analog Devices. Anyhow, yes, you can buy good LDOs but for the privilege of having one that doesn't throw the occasional tantrum you have to pay extra. "Oh, you want a car that doesn't have a wobble at 65mph? Ok, but it'll cost ya."

As far as I am concerned I avoid LDOs like the plague.

--
Regards, Joerg

http://www.analogconsultants.com

Just measured one to hand.

100nF, Y5V, 25kHz, 5ohm. 1kHz 55ohm. (for comparison, a 100nF poly' gave 8ohm at 1kHz)

Looks like I always end up at the cheap tat end of the market :)

Those numbers seem extreme, high by 2 or 3 orders of magnitude. How did you measure them?

John

Nope. Dumping said capacitor into a 1.5 Ohm load using a PMOS/NMOS totem pole that uses up 1/3 of the chip :-(

...Jim Thompson

--
|  James E.Thompson, P.E.                           |    mens     |
|  Analog Innovations, Inc.                         |     et      |
|  Analog/Mixed-Signal ASIC\'s and Discrete Systems  |    manus    |
|  Phoenix, Arizona            Voice:(480)460-2350  |             |
|  E-mail Address at Website     Fax:(480)460-2142  |  Brass Rat  |
|       http://www.analog-innovations.com           |    1962     |
             
         America: Land of the Free, Because of the Brave

Hopefully this one can run on a cheap process then.

--
Regards, Joerg

http://www.analogconsultants.com

About a hundredth of that.

2 milliohm is reasonable provided the device is rated at 10V. If it's rated at 6V, then figure about 10 milliohm.

Cheers

PeteS

pits.

With ESRs like that, they aren't even good for that.

Cheers

PeteS

"John Larkin" "john jardine"

** With one of these maybe ?

Told us he owned one back on July 17.

God knows how he got those mad results.

....... Phil

Measured on a bit of kit I bought last week.

Cheap but very handy. It measures the component impedance and VI phase displacement. All other secondary components such as LCR are then derived mathematically. Just for comparison I checked a part I can trace. It's a 470n open frame Polyester, EPCOS part #B32560J1474K. The data sheet states a loss factor of

0.008 at 1kHz. The meter (test at 1kHz) shows D=0.0046, L=475.6nF, Series R=1.55ohms. The values tie up nicely.

(Had assumed most people would be equipped with fancy HP analyzers and would have posted a measurement or two but haven't see anything come in yet.)

All those numbers sound very weird to me. And you can't calculate ESR from a single-frequency vector impedance measurement. The polyester loss is most likely dielectric absorption, which does not translate to esr.

I just measured an axial-lead 0.1 uF ceramic cap directly, by applying a current pulse and observing the voltage waveform. It's showing about

50 milliohms, but that's close to my resolution limit.

John

Well, I don't use anything worse than X7R right now which is where I get my numbers from. The cheaper stuff will always have worse numbers, of course ;)

Cheers

PeteS

And it isn't necessarily linear either. When I was a kid I thought I had it all figured out with RF power stuff. Measured the ESR of a largish cap at 21MHz, the frequency where I needed it. Assumed it would remain constant regardless of current level. Yeah, right. Calculated the dissipation this would cause, lots of margin, nothing would get hot here. Or so I thought. Soldered it in place and about five minutes into using it ... KABLOUIE! Molten stuff all over the place.

BTW, just curious, does your TDS2024 also lack delayed trigger? My Instek 2204 does, and so do lots of others. Shook my head when I found out.

--
Regards, Joerg

http://www.analogconsultants.com

"john jardine"

** The published data for that same cap shows it has an impedance minimum at just over 3 MHz of 33 milliohms.

........ Phil