Capacitor rated values - what is 'current' thinking?

Linear and non linear

Capacitor rated values - what is 'current' thinking?

Postby chrismb » Sun Jan 29, 2012 6:53 pm

I seem to recall that the recommendations for electrolytic capacitors was to aim to run the voltage pretty high up to [but below] the rated voltage, because the oxide layers in the things build up thicker and 'adapt' to a high applied voltage over time [or conversely lose the ability to tolerate the rated voltage].

Not sure if this is still considered true for modern construction.

Anyhow, my immediate question is just to get the current view on where the optimum point of operating a capacitor is, and whether they can be operated up to the full rated value.

My issue is that I am putting some finishing touches to a high frequency power supply I started a long while back (I've mentioned this one to some of you before, off line). Back then, we were getting an excessive voltage here, over 253 V, which was bad for all sorts of reasons not least of which one of the power supply channels in this thing I'm building ran 25V electrolytics right up to the full 25V. The voltage is better improved these days, hitting a fairly well regulated 235V or so, giving 23V at the caps.

This is running through a regular bridge off a transformer. I added a load resistor to pull ~ 0.1A off the transformer at minimum (to stop the output voltage floating high, as some seem to do with no load at all).

So, I'm not sure I should be that concerned about running these 25V caps at 23V [should I be?] (they feed ~2 to 3A but get only very slightly warm to the touch, less than body heat) but I am concerned about what happens if the darned local voltage floats high.

What's the usual 'professional' recommendation these days on such matters?
chrismb
 
Posts: 620
Joined: Thu Aug 05, 2010 7:32 pm

Re: Capacitor rated values - what is 'current' thinking?

Postby johnf » Sun Jan 29, 2012 9:00 pm

Chris
All depnds on how long you want it all to hang together before something fails

for my own professional use I would want the voltage 2/3 to 3/4 of the rated maximum

but more important is the actual use and ratings @ that use ie ripple current --which leads to capacitor heating.

For what its worth larger voltage values are bigger for the same capacity therefore they can handle more ripple current.
Tall thin ones are better than short fat ones as there is more surface area to get rid of heat.

If you want to see how hot they are internally you can take the end off the electro (furthest from the leads) and poke a thermocouple down the centre. Warm on the outside maybe but I have measured some @ 150 degrees intenally (bad for longevity)
johnf
 
Posts: 375
Joined: Sun Aug 08, 2010 4:51 pm
Location: Wellington New Zealand

Re: Capacitor rated values - what is 'current' thinking?

Postby Doug Coulter » Mon Jan 30, 2012 8:11 pm

John's hammered it again here. It's temperature with modern electrolytics. Digikey lists special ones that will hack it hot - and as John says, it's usually ripple current that's heating them in the first place.
What happens to an electrolytic that sits too long with no or low volts on it is the oxide gets thin - capacity goes up(!). But then you get to re-form that by gentle application of low currents, else you heat and arc them and that breaks them. You can run most caps somewhat over the rated voltage fine - the capacity goes down a little bit, that's all, or they arc - and die. Too high a voltage will oxide right through the metal and then they'll fail open instead. If you look in the parts catalogs, or download the manuf data sheet, they are specced for a higher voltage for short times than the ratings which are more or less nominal. You can run a 450v cap at 500v forever usually - about that ratio applies all over.

Here's a link to Digikey - note that temperature is one of the selection criteria. http://search.digikey.com/us/en/cat/cap ... capacitors
You can also get to catalogue pages that spec ripple current vs life on the higher temp rated ones, most often.

The MLCC ceramics don't have this problem, and now you can get them in big values. I tend to slap one across the low volt electrolytics in a switcher, since they have almost zero ESR and don't heat up - but do protect the bigger cap from some of the higher speed ripple. You can now get a few hundred uF in a 1218 smd cap!

Back when I fixed consumer electronics for a living (1960s or so) - other than jacks busted by sheer force - and takeup clutches in tape drives - most of the failures were the electrolytics. Still true I bet.
Posting as just me, not as the forum owner. Everything I say is "in my opinion" and YMMV -- which should go for everyone without saying.
User avatar
Doug Coulter
 
Posts: 2934
Joined: Wed Jul 14, 2010 8:05 pm
Location: Floyd county, VA, USA

Re: Capacitor rated values - what is 'current' thinking?

Postby chrismb » Mon Jan 30, 2012 8:52 pm

The particular caps in question are Nichicon PR types. I don't have a datasheet for these, I think they have been superseded by the PS series - which distinguish themselves in having the highest ripple current capacity in Nichicon's electrolytic range... ;)

I've not been concerned about that, only the voltage question, whilst the mains power is of questionable voltage! I have found an application note from Nichicon who politely say 'Please do not exceed the rated voltage of our capacitors', but as you say, Doug, it is not like they are screaming never to do it, only that the knee of the leakage current starts skywards at that point.

Lifetime is unimportant - my only concern is several thousand uF depositing its acidic electrolyte all over my hard-grafted electronics! I still shudder at the memory of accidentally running a 40V cap to 60V (I forgot it was in the circuit of a breadboard test) and it went off like a fire-cracker in the room. I jumped so high I almost hit the ceiling - but not before an orange goo got there first! It took me a moment for the old brain-box to re-boot and realise it wasn't actually raining in the room, it was something else!
chrismb
 
Posts: 620
Joined: Thu Aug 05, 2010 7:32 pm

Re: Capacitor rated values - what is 'current' thinking?

Postby Starfire » Tue Jan 31, 2012 10:37 am

Chris, bear in mind that the voltage is at full peak on switch-on even with a resistor (no current flow initially and no voltage drop across the resistor) before dropping to zero as the current rises to maximum - voltage dropping is fast but it does reach maximum voltage at the start. Better to limit the start-up voltage with a variac on the source supply primary if you are nursing old e-caps back or caps that have not been used for some time and start-up slowly. Voltage breakdown is relevant for electrolytic's
Starfire
 
Posts: 143
Joined: Thu Aug 05, 2010 5:26 pm
Location: North Ireland

Re: Capacitor rated values - what is 'current' thinking?

Postby chrismb » Tue Jan 31, 2012 2:36 pm

Starfire wrote:Chris, bear in mind that the voltage is at full peak on switch-on even with a resistor ...

I think I'll 'scope it, just to check (and also to check whether my meter might be getting confused by the underlying 50Hz signal). It might be less .. or more than the value I was reading.

But in this particular circuit it is very unlikely to be high on switch-on. It is driving 5 DC cooling fans, with a combined draw of ~2A, so when they are cranking up as the voltage comes on, [I reckon] the volts are sure to be pulled low.
chrismb
 
Posts: 620
Joined: Thu Aug 05, 2010 7:32 pm

Re: Capacitor rated values - what is 'current' thinking?

Postby Starfire » Tue Jan 31, 2012 3:00 pm

You will need a very fast scope to see it - the first spike is only a few pico-sec but the rated voltage can punch the dielectric occasionally.
Starfire
 
Posts: 143
Joined: Thu Aug 05, 2010 5:26 pm
Location: North Ireland

Re: Capacitor rated values - what is 'current' thinking?

Postby chrismb » Tue Jan 31, 2012 4:08 pm

No spike here, John! 8-)

P1310948.gif


Actually, looks like I seriously overdid the capacitance. Can't see any ripple voltage at all! Maybe the inductance of the multiple motors it is driving also serves to filter the voltage content.
chrismb
 
Posts: 620
Joined: Thu Aug 05, 2010 7:32 pm


Return to Analog

Who is online

Users browsing this forum: No registered users and 1 guest