ESR of electrolytic capacitor ar desired frequency

Status
Not open for further replies.
N

NOT2MUCH

Newbie level 6
Joined
Jul 26, 2012
Messages
11
Helped
0
Reputation
0
Reaction score
0
Trophy points
1,281
Visit site
Activity points
1,348
Hi, I used an aluminum electrolytic capacitor, which is rated at 220uf, 160V, as output filter in boost converter. The datasheet of capacitor is shown below. My converter is switching at 100KHz, how can I calculate ESR at this frequency?


He, Siyu
 

As you might know, I need to caluculate the ESR of capacitor at 100KHz, the question is how to get it? But I don't instrument to measure it.

Siyu
 

I looked this site :- **broken link removed** to see what their caps do, one version 's ESR is .4 X that of its 120 HZ value at 10KHZ, but the graph flattens out so it won't change much for 100KHZ. The particular device was a "low inductance type", with an inductance of 13 NH, compared to 40 NH of a non low inductance type. Try to google to find more data.
Frank
 

The capacitor I used is a EB series from Panasonic, part number EEU-EB2C221. I tried to find data, but I did not get any other data except the datasheet I inserted in the first post.

Siyu
 

On testing that I have done with different electrolytic capacitors, the ESR value remains relatively constant for frequencies between 0 and 150Khz so you are safe in using the 120Hz value. Above 150K to around 1-2Mhz the ESR gradually drops with frequency. Above 1-2Mhz inductive effects kick in and the ESR varies wildly with frequency due to resonance.
 

If examines the max ripple current 1280mA spec is constant for various size that are approximately the same volume as are other values.
I believe the reason for this is to limit the internal temperature rise. Supposed we say the dielectric is a thermal insulator and the conductor foil too thin to be a good thermal conductor. If we estimate a thermal resistance of 40deg/W and say 20'C max internal temp rise is significant I would estimate the max desired internal power dissipation to be 0.5W = I^2 * ESR implies an ESR of 0.3 Ohms. This may be a big low, if my assumption for Pcap is too generous @ 0.5W.

My experience is if you have a controlled ripple current for AC on DC voltage, then use these caps with rated ripple current on DC voltage

But if you want low ESR for lower voltage ratings, then choose a cap rated for low ESR. They tend to be much lower in value and thus lower power dissipation and much higher higher ripple current.
However, ESR increases with rated voltage and similar parts appear to be somewhere under 1 Ohm and none better than the one you selected.

**broken link removed**
 
Last edited:

Based on my caluculation, the esr of electrolytic capacitor I used is 0.15*1/(2*pi*120*220e-6)=0.9043 ohm, which is very large. Do I calculate it right? I am still wondering what kind of capacitor should be used as output filter in DC-DC convert?
 

NOT2MUCH said:
Based on my caluculation, the esr of electrolytic capacitor I used is 0.15*1/(2*pi*120*220e-6)=0.9043 ohm, which is very large. Do I calculate it right? I am still wondering what kind of capacitor should be used as output filter in DC-DC convert?
Click to expand...

That's correct, and not very surprising. Many higher voltage electrolytic caps have large ESR, and therefore aren't suitable for some applications. Generally ESR will decrease a bit at higher frequency though.
 

It may be better to use a polypropylene cap, even if that means using a much lower value to keep the cost reasonable. e.g. for about $10USD, you can get polypropylene caps of 10uF to 20uF with ESR < 10mΩ.

If the design really requires a high value, you could use e.g. 220uF aluminum bypassed with 10uF polypropylene.
 

How would the bypass capacitor work? I read a capacitor application note, it says the paralled capacitor should be ''identical'', same rated voltage,ESR and current ripple. The other question is if I parallel a 220uF aluminum with 10uF polypropylene, how do I calculate ESR? Just parallel two ESR?

- - - Updated - - -

Thank you, I just tried to find a low esr cap, it is like what you said.
 

NOT2MUCH said:
if I parallel a 220uF aluminum with 10uF polypropylene, how do I calculate ESR? Just parallel two ESR?
Click to expand...
Assuming the 220uF cap has ESR = 1Ω, then the impedance of the 10uF poly cap will be lower than that of the 220uF above 16KHz. At lower frequencies, the impedance is set by the 220Uf cap.
 

I need to use esr to calculate feedback loop parameters, how would I do with alum cap parallel with poly cap?
 

Zc=1/(2pi fC)+R and two in parallel are the 1/Ztot = 1/Z1 + 1/Z2 = (2pi f*C1*R1+1)/(2pi fC1) + (2pi f*C2*R2+1)/(2pi fC2)= You can conjugate the rest.
Without knowing your freq range, I might be able to assume your Polycap is ideal and 1/10th the value of C..
 

As the ESR and capacitance value of electrolytic capacitors have a big tolerance and are subject to ageing, it would be best to do your calculations with a, .9 Ohms, 2,Ohms, and with .05 Ohms. Then once you under stand what effect ESR has on your circuit, make due allowances for it getting bigger and the capacity getting smaller, so the circuit becomes tolerant of such effects.
Frank
 
NOT2MUCH said:
I need to use esr to calculate feedback loop parameters, how would I do with alum cap parallel with poly cap?
Click to expand...
If this is control loop analysis, then the ESR at the switching frequency probably isn't very relevant. Rather, you should be interested in ESR within the loop bandwidth of the controller.

In the general sense, you just take whatever your system is (whether is has one or two or whatever output caps) and just calculate its loop response, and use that as a basis for designing the controller. Depending on your specific needs, adding another capacitor in order to get lower ESR may or may not be worthwhile. In general, capacitor selection should be guided by ripple requirements, not just loop response.
 
Actually I don't have such a polypropylene cap fit for my circuit, I used a ceramic capcitors to parallel with alum cap, I know esr of ceramic cap is very low, but how low it is? I can find such data on datasheet.
 

What does a low esr cermaic cap looks like? There are a lot of ceramic cap in lab, but most of them are bought long time ago and can't find them by pn on digikey.
 


Status
Not open for further replies.

Similar threads

Menu
Log in
Register
Cookies are required to use this site. You must accept them to continue using the site. Learn more…