what's best to read ESR with a microchip ADC

dr pepper said:

I programmed with asm till recent when I switched to C, certainly saves time, I studied c, but at the time compilers were trash compared to now.

Click to expand...

I plan to do C...even got the mikroelectronika C kit...still in box after a year. I seem to be a minimalist...extracting max perf. out of the 16F/12F series. using ASM.
Gotta move on up soon though. Got a 24 bit DSP mikroelectronika dev kit as well.

I guess not being able to set my vref+ to 1.1 volt nocks first ideal out of the woods. I.m only getting like .4 v0lts which is good but I'll have to use a opamp to scale the value up..

some PIC chips have an analog OPAMP or 2 built in.

If you dont choose express install, then you can disable many shared experiences which Win10 uses to interpret your application and metadata to unify and customize your experience. Like whom you share data with most often are prioritized, favorite lists etc. just like any social media software.

They want to build on the Apple and Google model of revenue stream for to build up a large user base then figure out how to make revenue from this large user base with a free OS.

Either updated paid apps or free ads or or free cloud storage with pay as your go increases. This can lead to more sustainable income than the OS cost and the hidden Scheduled Tasks can ensure they keep pirates out.

The abundant secrecy of a dumber user interface like the iPad is what is so appealing to MS that they want to copy this model without patent infringements.

Unfortunately, for corporate uses and geeks like me, this obfuscation is a worry or a frustration. They add more features, but take away control.

Does anybody know where this ESR tester design came from? Schematic is shown as one of the illustrations on the left...

Edit: Click on "downloads". Manual and circuit is there.

Bob Parker designed the Blue ESR meter supplied by
Alltronics LLC/AnaTek Instruments
2761 Scott Blvd.
Santa Clara, CA 95050 USA

It uses selectable series Rs to drive a squarewave from 100R to 10k clamped by diodes and with AC peaks amplified by 7 with Q fb ratio and RC values to decay the peaks. Thus it measures the peak step rise as ESR.

Does Z86 have software SAR ADC? hmm no S&H , must be a comparator method?

The meter puts out regular bursts of 10us pulses at a 2KHz rate, at an open-circuit amplitude of about 600mV P-P. At a pinch you can use it as an audio signal source to check speakers, amplifiers etc. The pulses have a fast rise/fall time, so it would probably make a crude RF signal injector as well.

They say the blue is great meter. That mc used doesn't have adc.

But I have no idea how the BJT amp and software gets such accuracy with a pulse amplitude and comparator with fixed threshold

My guess at how it works is by differentiating the ESR voltage step back into a pulse and using then integrating each pulse into the emitter cap and adjusting the gain to achieve a calibrated ESR when the 680 Ohm crosses the internal comparator of 5 to 10 mV. THis also compensates a bit for Beta variations. Doesn't look very accurate.... non-linear , drift, etc. By autoranging the 100,1k,10k it can count the number of pulses every 50us and check the comparator for > 5mV threshold . THe staircase ramp is proportional to the step size of the Vesr * Rseries pot*C ratios {series cap/emitter cap.} but still sensitive to Beta even with feedback.

It is explained in this pdf-file:

LOl

It’s worth noting that ESR increases rapidly as the temperature drops. As a result, defective electrolytics are often indicated by faults which are worst in winter and when the equipment is first switched on, with the symptoms gradually diminishing as the temperature rises.

Click to expand...

I had a old computer a 286 IBm that would run fine every summer and never turn on in the winter I remember working in my basement it was hot down there no ac. It run fine there I moved new house had Ac in the basement the dang thing never ran again then one day i was selling stuff I didn't outside it was hot that day I hooked up that old computer it run the first press of the power button.
This is why I'm making a esr meter LOL I never could find the bad cap on the dang thing LOL

I would expect 20% accuracy or much worse with this design , even after calibration for one part. But who cares as long as it is relative. That probably explains the non linear ESR vs C vs V curves.

As long as it reads bad cap's without unsoldering lol I'd be happy but then I don't need to put cap in the freezer anymore LOL. to test it..

That design uses 2% of 5V pulse so Vavg always approaches 1V. Differentiating the ESR*I is not precise as there is no S&H and Vavg is affected by Rext*Cd.u.t. such that 100*1mF with 100R ESR gives an error in source current and longer time constant than the reference RC cap and thus more error comparing the two.