How to design ripple meter for 19V lines?

[zaoka]

I had this idea for a long time.

What I want to do:

Measure ripple/noise level on AC adapters for laptops.

So instead of using a scope evey time + resistors for dummy load I was thinking to build a meter that will test these and provide Voltage, Current and Ripple measurements.

Most of the AC adapters for laptops are 18.5 - 19.5V and currents 3.42 - 7.5A. Operating frequency is from 10Khz to 50Khz from what I measured. Whats happening with these is that filtering capacitors on 19V line goes bad or change value and this causing different kind of problems for laptops. Sometimes it does not charge all the time, it turns off randomly etc.
I have also seen main filtering capacitor goes bad.

We all know that weak capacitor will recover when temperature goes up, so this mean if AC adapter is not used it may not work OK untill it warms up etc..

Im not designer but we should be able to detect 120Hz modulated ripples in case that main capacitor is not good and frequencies from 10Khz to 50Khz. All these should be measured with a small analog meter, under maximum dummy load, in pp.

That Daves dummy load meter could be used for dummy load and we just need ripple meter.

Does anybody have schematic for ripple meter or some tips how to make it so that it measure those frequencies?

[alm]

How about a high-pass filter (capacitor) to block DC and a DMM on AC with enough bandwidth? Or a DIY AC meter (I don't think true RMS is a big deal for this, you're mainly looking for order of magnitude, so averaging is fine) if you want a stand-alone instrument.

You would need some kind of load (ripple tends to increase with load), which can either be a laptop running some kind of stress test+charging battery or a dummy load.

[zaoka]

I was thinking to use variable dummy load with MOSFET and separate ampmeter, this way if AC adapter is rated at 4A maximum I can set that current and measure ripple

[Zad]

DC-blocking capacitor and a precision rectifier op-amp circuit should be all that is needed. Scale the feedback gain resistor as needed.

[Zero999]

Are you looking at the peak voltage or RMS?

Yes you need a DC blocking capacitor, precision rectifier and a peak detector.

[Eliminateur]

Zaoka, it sems we're in the same line of business hehe,

i completely overlooked this thread!, and it's actually useful for my larger project which i posted last night(https://www.eevblog.com/forum/index.php?topic=2295.0)

i'll try and implement ripple tester in the voltage tests portion for future revisions(i don't want to get the project so out of scope i'll never get to build it).

hmmm i've found some app notes and circuits, but first i'd like to define what's the accepted ripple?, 20mV?, 100mV?, 200mV?, if i where to do some sanity check, i'd say anything above 250mV ripple is BAD...
should we use a full wave or half-wave?, seeing as ripple *should* be somewhat symmetric i could simply use half wave, is it an acceptable assumption?

also, it seems most(full wave)circuits use split-supply opamps, and i'd need a single-supply opamp.

i've found an interesting article here: http://sound.westhost.com/appnotes/an001.htm

i like this circuit:

but has a lots of caveats like nonlinearity, trouble with input capacitor, etc.

Is it possible to build a half-wave single-supply precision rectifier?, there's no examples on that page..

I've also found http://www.analog.com/static/imported-files/application_notes/AN-573.pdf page 6 which uses an OP727 for a single supply full-bridge, but it says it's for <2kHz freqs and only 2vpp(which is OK since coupled), probably due to the opamp used, i wonder if it would be possible to extend said freq by changing opamp...
What would be the effect of the coupling capacitor in freq response there?(i should do some simulations...)

It seems a problem with single-supply is non-linear impedance for + and - part fo the signal which causes trouble with AC coupling