ESR Questions

[ADWSystems]

First querstion: How does the ESR of a standard capacitor compare to that of

1) a regular alkiline or NiCd battery

2) a supercap such as **broken link removed** (this one is actually listed at 30 ohms)

and 3) a "memory back-up" cap like https://www.allelectronics.com/make-a-store/item/CBC-18/4.7-FARAD-2.5-VOLT-MEMORY-DYNACAP/1.html

Secondly. Am I remembering correctly that the ESR reduces the voltage available depending on the current draw? If you have a "storage device" charged to 5V with an ESR of 10 ohms and pull 100mA you lose 1V due to the ESR and only receive 4V across the terminals?

 

[RCinFLA]

A capacitor Rs is much more straight forward compared to a battery. For a capacitor the Rs is relatively constant and dominated by the terminal and plate resistances. It also has a frequency dependence. Some dielectrics have charge transfer delay and in this, act with some of the effects of a battery.

Battery Rs is much more complicated as in not only involves the resistance of metal and foil but also includes a chemical effects called the electro-chemical kinetic effect. This effect is caused by the creation and movement of ions with chemical recombination and, in some batteries, actual movement of molecules within electrolyte. Heavy current load will cause a localized depletion of charge near the anode and cathode within cell dropping terminal voltage while the kinetic effect falls behind on electron demand requirements.

 

[duffy]

There's no such thing as a standard capacitor. ESR varies wildly, a polyester cap can be less than .0005 ohms, a multilayer cap can be over 50 ohms.

An AA alkaline starts about 150-200 milliohms internal resistance and goes up over an ohm long before it is "dead". An AA nicad is about the same but stays flat through discharge. An HB1030-2R5106-R supercap has about 60 milliohms ESR -
https://www.electro-tech-online.com/custompdfs/2012/01/Bus_Elx_DS_4375_HB_Series.pdf

Yes, the ESR reduces the available voltage, just the way you said.

 

[simonbramble]

just picture it as a resistance in series with a perfect voltage source. The more load you put on the output, the more the drop across the resistance, hence the lower the output voltage.

 

[Boncuk]

just picture it as a resistance in series with a perfect voltage source. The more load you put on the output, the more the drop across the resistance, hence the lower the output voltage.

That bites!

Since when is a capacitor a perfect voltage source?

 

[ADWSystems]

That bites!

Since when is a capacitor a perfect voltage source?

It isn't, hence the ESR.

I'm looking at a design that is to only be used intermittently and infrequently but may be activated at any hour and only runs for 5-10 seconds (think of a status panel that you push a momentary button to light the status's and when you release it shuts off). It should work with or without line power available. I'm thinking of proposing an off-grid solution and make it solar recharged. The next design point is battery or supercap. The supercaps will last longer (more discharge-recharge cycles), so I'm trying to understand all the required specs.

If I was to connect to caps in parallel, the capacity (F) doubles but working V halves. If I connect two series pair in parallel, then I should end up with the spec'ed working V at twice the capacity. Right? What happens to the ESR?

 

[ericgibbs]

It isn't, hence the ESR.

If I was to connect to caps in parallel, the capacity (F) doubles but working V halves. If I connect two series pair in parallel, then I should end up with the spec'ed working V at twice the capacity. Right? What happens to the ESR?

hi,
Simon didnt say a capacitor is a perfect voltage source, he said 'assume' it was, in series with a resistance equal to the ESR.


When you connect two identical value capacitors in parallel, you double the capacitance and the working voltage remains the same. In series its half the capacitance and double the voltage [ balancing resistors are recommend if the applied voltage is greater than the rated voltage]

Also the ESR of two identical capacitors is halved when connected in parallel, you will often see in a SMPS two or more caps in parallel in order to reduce the overall ESR value.

 

[ADWSystems]

I'm looking at needing to double the working voltage but also maintaining the capacity rating. So I'm looking at a bank of 4 caps, 2 in series and then the pairs in parallel. Can you refresh me on the balancing resistors when connected in series?

 

[ericgibbs]

I'm looking at needing to double the working voltage but also maintaining the capacity rating. So I'm looking at a bank of 4 caps, 2 in series and then the pairs in parallel. Can you refresh me on the balancing resistors when connected in series?

hi,
Look here. https://www.coilgun.info/theorycapacitors/capacitors2.htm

 

[ronv]

Don't forget to calculate the time constant of the capacitor and the load as well as the drop from the esr. It may be a bigger problem if you draw much current.

 

[duffy]

Can you refresh me on the balancing resistors when connected in series?

You need to make them low enough so the V/R is considerably greater than the cap leakage current. For the 10F I mentioned, about 100k would be equivalent to the leakage and 10K might be a good balance choice. Unfortunately, this will cut the reserve time down from around 11 days to 1 day.

If that's an issue, you might try using 2.7V zeners in parallel with the caps to balance them instead of resistors. Use ones with low leakage, like a MM3Z2V7C. They leak about the same as that cap (around 20 uA) so your reserve time is maybe 5 days.

 

[ronv]

How much current does your circuit need?

 

[ADWSystems]

Still working that part through. Luckily I know it is not for a very long time (though in the 5-10 seconds in electronics is equivalent to 5-10 human years). I'm also thinking of a capacitor bank backed up by a couple of alkaline batteries.

 

[ronv]

If I was in doubt I think I would go with 3 NICD cells and just use the solar to trickle charge them at .05C (110ma). So even if your button used 100ma you could put it all back in an 10 seconds of sunlight. That way even without sun you could push the button fo 22 hours as oposed to 20 or 30 seconds.

 

[diy didi]

Why are capacitors always the culprit. The clever guys haven't come up with a better designed Condenser yet??

 

[duffy]

There is no "culprit", he is doing design, not repair.