Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Electrolytic capacitor polarising voltage. (LG please look here)

Status
Not open for further replies.

JimB

Super Moderator
Most Helpful Member
LG, in another thread you stated:
I cant see a valid reason you would want to charge the cap up????????
And I responded:
How would you test an electrolytic capacitor?
Does the capacitance vary with polarising voltage?

What I wanted you to realise was that for an electrolytic capacitor to work correctly it needs a DC polarising voltage to make the dielectric form the thin insulating layer between the plates.
Measuring an electrolytic capacitor which has been previously used but now discharged, you will get a reasonable measure of its capacitance.
Also, the thickness of the dielectric layer can vary with the amount of polarising voltage (ie the DC voltage across the capacitor), and as the dielectric varies, so will the capacitance.

So, going around in circles, if you need to know accuratly the capacitance of an electrolytic capacitor, you must also consider the DC polarising voltage.

If you look at the specifications of electrolytic capacitors, the tolerance is very wide, usually -50% to +100%. So wide you could drive a Barnsley bus through it!

JimB
 
As an extension to what JimB has said, it is also the case that the dielectric layer can change over time. A capacitor with say a 16 volt rating has been formed at just over that voltage; BUT if the capacitor is used at a lower voltage, over time the capacitance can increase and the voltage rating can fall, and vice versa. I had experience of the capacitor in a photo flash unit being unable to be charged because of leakage that was too high for the small battery power supply to overcome. After setting it on an external power supply and reforming the dielectric gradually up to the required working voltage, the capacitor became OK again.
JimB's remark about the capacitance tolerance is based on the fact that users of electrolytics don't care if the capacitance is higher than stated. So tolerances of +50% are not uncommon. But if a capacitor has such an excess, it is likely that modern manufacturing methods would dictate, from the material usage point of view, and use less material to conserve costs.
 
ok thanks, I over simplified things, and my answer was lazy. I looked at it purely from taking a cap and testing with a LCR meter that I have.
On a slightly different path,
Electrolytic caps are wide in tolerance, most the time this dosnt matter too much.
When tolerance matters in a design, is this the point you switch over to say tantalum?
Alot of test equipment I have uses Tantalum caps, they are a pain when they fail, why are they chosen over electrolytics in some test equipment?
The other thing I have trouble understanding, normally when you change a electrolytic with a tant or some other cap, normally the recommendation is for a different value, I am mostly thinking of mother boards in a pc, my head says 1uf is 1 uf no matter the cap, so why recommend a different value when replacing caps of different types?
 
Tantalum caps, they are a pain when they fail, why are they chosen over electrolytics in some test equipment?
I think that tants have a lower series resistance and inductance than the usual aluminium foil type electrolytics.
So you get more effective decoupling in a smaller physical package.

my head says 1uf is 1 uf no matter the cap, so why recommend a different value when replacing caps of different types?
Probably to compensate for the less effective decoupling due to the parasitic R and L.

Often it is more effective to put two smaller value capacitors rather than one big one.

Many years ago at a place where I worked, we had strange problems with some flow measurement computers*.
The problem was in the power supply board.
The maintenance philosphy was to just replace the board with a spare and send the faulty one back to the manufacturer for repair.
The boards would come back to us "No Fault Found", and would still cause problems when installed in one of the computers.
Eventually we ran out of working spares, so some real faultfinding was needed.
I found the problem to be excessive ripple on one of the supply lines, due to a dried up electrolytic.
So, put in the closest replacement we had, the ripple was much better but could be improved.
So I tried a second capacitor in parallel, much better, so try a capacitor which was twice the value.
That was OK but still not as good as the two in parallel.

I cannot remember the final solution to the problem, whether we got some better capacitors or just sent the whole load of iffy boards back to the manufacturers and said "Look here dumbo, the problem is this cap. Fix it!". I think that was about the time when I was moving on to other things.

* Not a computer with a processor, but something with lots of analogue and logic which calculated gas flow rates from differential pressure and density. There were a couple of umpteen bit frequency dividers which had to be set using a whole load of link plugs.

JimB
 
I agree with JimB on all that. I have the Philips databook for resistors and capacitors, and this book is full of data such as ESR(equivalent series resistance), max current(AC) and in the case of the small polypropylene and other small capacitors, the book has data on resonant frequency. Electrolytics do have a high ESR and to be effective bypass capacitors, it is often the case that a small (low inductance/high resonant frequency) capacitor is placed directly in parallel with the electrolytic. Tantalums are known to have better high frequency characteristics than electrolytics and are used in lieu. Particularly with high speed systems and logic, it is good design practice, to bypass each IC with a HF bypass capacitor. The problem with PCB design is the parasitic coupling and inductance of PCB tracks which occurs, and to bring some certainty to the design, good practice likes to see heaps of bypass capacitors "just in case".
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top