Opinions on capacitor type ...

On Tue, 5 Aug 2008 17:51:39 +0100, "Arfa Daily" put finger to keyboard and composed:

Some time ago I investigated the accuracy of Bob Parker's ESR meter for low valued caps, including X2 film types. Perhaps the following posts may provoke some thoughts:

If your meter uses a different testing methodology, then the above observations may not be relevant.

I'd say don't risk it. According to this datasheet, these caps "are not intended for continuous pulse applications", nor "for series impedance applications":

APPLICATION NOTES

? For X2 electromagnetic interference suppression in across the line applications (50/60 Hz) with a maximum mains voltage of 275 V (AC).

? These capacitors are not intended for continuous pulse applications. For these situations, capacitors of the AC and pulse programs must be used.

? These capacitors are not intended for series impedance application. For these situations in case safety approvals are requested, please refer to our special capacitors of 1772 series with internal series connection.

- Franc Zabkar

--
Please remove one \'i\' from my address when replying by email.

Blue tends to suggest polypropylene caps IME (good for pulses) which figures in such an application although you don't state where they are in the circuit.

How about some more info on what they do ?

Graham

Agreed.

Graham

On Wed, 06 Aug 2008 06:09:45 +1000, Franc Zabkar put finger to keyboard and composed:

This datasheet doesn't have the above disclaimers:

However, page 6 has a voltage versus frequency graph which shows that the maximum RMS AC voltage of 300V starts dropping from about 300Hz. At just above 10kHz the max voltage rating is only 10V, at 1kHz it is about 100V.

- Franc Zabkar

--
Please remove one \'i\' from my address when replying by email.

On Tue, 5 Aug 2008 17:51:39 +0100, "Arfa Daily" wrote:

:A particular board that I'm currently involved in repairing, makes use of a :couple of 15nf 630v dc caps in the primary side of a smps transformer. They :go open or low value, and are easily found initially by running an ESR meter :over them. Good ones go about 38 ohms, bad ones over 80 ohms or beyond the :range of the meter. Once you've got them out, a value check will confirm :that they are faulty. The originals are those 'boxy' blue metalized film :ones with 15mm pitch on the leads. Some are fitted with the shiny brown :epoxy dipped metalized film types, which interestingly, don't seem to fail. : :My problem is that I am having difficulty finding any suitable replacements. :I can find blue box metalised film caps with a 630v dc rating, but 15nf :seems to be a rare value. Also, they are smaller than the originals, and :have only a 10mm lead pitch, and legs that are too short to be able to open :out to 15mm pitch. I can't find any 'dipped' ones at all in that value / :voltage rating. : :What I have found is an X2 rated metalized film cap in that value which :quotes an ac rating of 300v, and a dc rating of 630v. Its from BC, and it's :their 2222 338-4 Series. It is the right physical size, and seems to work ok :in the application. Thing is, when you run an ESR meter over these caps, :they go about 80 or 90 ohms. Now I know that you can't trust the reading of :an ESR meter that's designed to evaluate large value electrolytics, on a :small value film capacitor, and the fact that I was using it at all, was :just a matter of convenience in that it quickly gave a useful indication of :the state of the caps, without having to dig them out of the board. So, am I :seeing this much different figure on the ESR meter, because of the way the :meter works, and the fact that these caps are designed to be low inductance, :whereas the originals are just bog standardly constructed ? Can anyone see :any problem in using an X2 capacitor in a pulse application like this, given :that they supposedly have a superior pulse characteristic, and claim to be :low loss in pulse applications ? : :Opinions for discussion, appreciated. : :Arfa :

X2 is not recommended for the particular application.

I have always tried to use Evox-Rifa (in the 60's they were simply Rifa) because they were very reliable. Of course, with all the mergers/takeovers since those days I don't know how their reliability has changed, if at all.

Have a look at the range of PHE426 pulse caps here

This series does not have 15nF/630V in 15mm lead spacing but if you want more than a few you may be able to order the 10mm spacing with lead length of 30mm instead of the standard 6mm.

On Wed, 06 Aug 2008 03:05:19 GMT, Ross Herbert wrote: :X2 is not recommended for the particular application. : :I have always tried to use Evox-Rifa (in the 60's they were simply Rifa) because :they were very reliable. Of course, with all the mergers/takeovers since those :days I don't know how their reliability has changed, if at all. : :Have a look at the range of PHE426 pulse caps here :

: :This series does not have 15nF/630V in 15mm lead spacing but if you want more :than a few you may be able to order the 10mm spacing with lead length of 30mm :instead of the standard 6mm.

The X2 you are using is probably this one

Since X2 caps are generally physically larger for a given value than a standard MKP or other film type the space available appears to allow a component measuring at least 5 x 11 x 17.5. You may be able to find ahigher voltage standard pulse cap of similar size.

Sticking with the Vishay (BC) brand, the 2222 378 64153 in this range would be ideal

Thanks all for your views and good input. Plenty of food for thought and further research. I have looked at Vishay's list of re-sellers and will have a look at their websites to see if I can find any that keep the 378 series in the value I need. The 'usual suspects' like RS and Farnell are there, but these caps don't seem to be in their catalogues here in the UK, although others - the X2s that I have tried for instance - are. I will have a look at Arrow etc when I have time.

TA

Arfa

OK. Here's some additional info (I did post a reply to what we've come up with so far, but it doesn't seem to have appeared - strange ...)

Looking again at one of the original caps, the trade name is "PILKOR" which seems to be a Korean component manufacturing arm of Vishay. So far so good. The numbers "378" do appear, so this would suggest that it is indeed one of Vishay's 378 series MKP film caps. Now, Farnell keep some of the 378 range, but not this one. They do, however, keep that value in the "385" series. As far as I can see, the main difference here seems to be that the caps are rated at 2000v dc rather than 630v dc, and are physically taller, whilst still having the same footprint and lead pitch as the 378s. Taller would not be a problem. Higher voltage may be an advantage ...??

Any comments or opinions on this range being more suited, appreciated.

The caps are located on the primary side of a resonant mode smps. It is a sort of half bridge circuit with two FETs. The drain of the upper one is connected straight to the 400v rail, and the upper connection of the transformer primary, is returned to the same point via one of these caps. The upper transformer connection is also connected to the second capacitor, the remaining end of which is returned to deck via a very low value R, which is used to sense primary current. The lower FET has its drain connected to the upper one's source, and it is this 'mid point' which hooks to the lower connection of the primary winding. The source of the lower FET is returned to the same deck as the low value R detailed above. So I guess that the caps are for coupling and resonating the primary. Looks like they would be working hard ...

Arfa

On Wed, 6 Aug 2008 12:44:05 +0100, "Arfa Daily" put finger to keyboard and composed:

That explains a lot. For example, you measured an ESR of only 38 ohms instead of 80, which is what you would expect if the two caps were connected in parallel. In fact your circuit does effectively parallel the two caps via the low ESRs of the current sense resistor and 400V bulk capacitor.

If one cap is OC, then you will see an ESR of 80 ohms, which is the ESR of the other good cap. If both are good, then you will measure 40 ohms.

- Franc Zabkar

--
Please remove one \'i\' from my address when replying by email.

Ha! Well spotted Franc ! Good answer, and obvious now I look at it again ...

Arfa

Metalized film has many drawbacks, you really have to read the fine print on the spec sheets. The metalization is soo thin it can't take much peak current.

So instead of looking for exact replacements, which will just repeat the cycles of failure, I'd do the exact opposite-- look for something sturdier.

The caps will be slightly larger but if you can squeeze them in they circuit might have a chance of lasting quite a bit longer.

because

I agree with your comment about Evox-Rifa.

Epcos (formerly Siemens) also do several ranges of polypropylene caps with various pitches which may help. The data on them is also very extensive, indicating max pulse handling etc.

Graham

X2s AIUI are not designed for *repetitive* pulse work. Polyprops are. You also have a choice of film/foil or film/foil/film construction (at least from Epcos) which affects peak and average current handling.

Graham

That's why Epcos make film/foil/film ! Not just metallised film.

Graham