cap questions. / General Science and Electronics / Forums

Forums

4hv.org :: Forums :: General Science and Electronics

« Previous topic | Next topic »

cap questions.

Move Thread

LAN_403

Marko

Sat Feb 19 2011, 04:45AM

Registered Member #89 Joined: Thu Feb 09 2006, 02:40PM
Location: Zadar, Croatia
Posts: 3145

Hi guys,

I need some thoughts on the following:

Regarding the DC blocking capacitors such as commonly used on high power half and full bridges (SSTC's, SMPS's): Most people seem to throw their their finest polypropylene film/foil, low loss caps for the purpose. Now I happen to be out of suitable polypropylene caps, and need to replace some blown DC blockers. My question is, is there an actual justification why not to use plain cheap metallized mylar caps for this application? My dogma is that thisshould be fineand atleast as good as MKP caps actually, since capacitors in this application have hardly any voltage drop on them and hence very little Var's. They'll still suffer from ohmic heating but dielectric loss should be negligible - how right am I regarding this?

One other subquestion is - for low voltage dc blocking (like for gate drive) we often need larger capacitances not easily attainable with dry caps, especially for DRSSTC's and like. But because voltage change on the cap is so low, is it actually acceptable to use polarized tantalums or even electrolytics for this purpose?
DRSSTC is also a special case where short bursts and initial dc component in the signal actually seem to assure that voltage will never actually turn negative. I seem to vaguely remember someone using tantalums in a drsstc this way, but I'm wondering if this can be applied to CW as well?
What could I call a "safe" limit of negative voltage which polarized caps can withstand indefinitely without damage?

Marko

Sulaiman

Sat Feb 19 2011, 07:37AM

Registered Member #162 Joined: Mon Feb 13 2006, 10:25AM
Location: United Kingdom
Posts: 3140

For an sstc half- or full-bridge it is the rms/ripple current rating that is important and I've noticed that German coilers often use MKP or MMKP (single- or double-sided metalisation, usually Wima brand) with success.

For low voltage blocking;
- I have serviced quite a few inverters (ultrasonic) that use bi-polar electrolytics, usually by the time we repair them the bi-polars need replacing due to increased esr.
- If you are sure that there will be no voltage reversals then any polarised capacitor with adequate voltage and current ratings will do.
- 'normal' electrolytics are rated for 1000 to 3000 hours, low esr electrolytics are around 5000 hours.
- all electrolytics will withstand some negative bias for some time, how much for how long varies.
i.e. 'crappy' capacitors last longer than good ones under reverse bias
(where I work there are about 40 of us repairing industrial electronics, we always replace all small electrolytics, the large Bus caps need to be checked but are usually ok. approximately once a week someone will put an electrolytic in backwards, the results vary from warranty returns (up to 12 months), smoke, pop and BANG. We have an informal scoring system for 'spectacularity' !)

-solid aluminium caps can withstand 20% to 50% reverse bias repeatedly, last almost forever, have very low esr but cost a lot
-tantalum beads soon die with more than a few volts reverse polarity
-smd tantalums may last for years reverse-biased but become high esr

So, if there are no voltage reversals use normal electrolytics,
if there are voltage reversals then use bi-polar electrolytics or preferably MKP

Wolfram

Sat Feb 19 2011, 04:23PM

Registered Member #33 Joined: Sat Feb 04 2006, 01:31PM
Location: Norway
Posts: 971

Marko wrote ...

One other subquestion is - for low voltage dc blocking (like for gate drive) we often need larger capacitances not easily attainable with dry caps, especially for DRSSTC's and like. But because voltage change on the cap is so low, is it actually acceptable to use polarized tantalums or even electrolytics for this purpose?
DRSSTC is also a special case where short bursts and initial dc component in the signal actually seem to assure that voltage will never actually turn negative. I seem to vaguely remember someone using tantalums in a drsstc this way, but I'm wondering if this can be applied to CW as well?
What could I call a "safe" limit of negative voltage which polarized caps can withstand indefinitely without damage?

Marko

I use multilayer ceramics for DC blocking in series with GDTs. Multilayer ceramics have gotten much better over the last few years, and now they are available in capacitances up to 100uF, and they are compact and cheap. They also have very low internal resistance, much lower than tantalums. I also use them for decoupling gate drive ICs, no need to use a 10uF tantalum and a 100nF ceramic when you can use a 10uF ceramic that takes up much less room and costs less. Here

is an example of one.

Marko

Mon Feb 21 2011, 11:25AM

Registered Member #89 Joined: Thu Feb 09 2006, 02:40PM
Location: Zadar, Croatia
Posts: 3145

Hi guys,

I just ordered a batch of 500 1uF monolythic ceramic caps... I can never get enough of these anyway.

Nobody has answered my first question, to clarify it, it's basically polyester (MKS or whatever) vs. MKP - I went on to try polyester caps and they were indeed for some reason so terrible that I couldn't get any oscillation at all.

I didn't have time to investigate this as I had to leave home, but I'm somewhat puzzled with this result. I could also swear I've seen people using polyester caps for their voltage dividers in half bridges, in induction heaters and other power stuff as well. Both MKS and MKP should have about the same ESR since they both have single-metallized electrodes, and dielectric losses should be negligible due to low voltage drop on the cap.

A bank of rugged film/foil caps would ofcourse be a sure shot, but of those I only have too-valuable tesla coil caps (150nF cde caps) which I already used in parallel for DC blocking duty and I'd like to try and spare them from that use.

Marko

Steve Conner

Mon Feb 21 2011, 11:38AM

Registered Member #30 Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706

Hi Marko

I think mylar caps often have worse ESR, just because they're made out of cheaper materials. Thinner metal, worse end terminations and so on. They're usually used in cost sensitive applications.

James

Mon Feb 21 2011, 08:40PM

Registered Member #3610 Joined: Thu Jan 13 2011, 03:29AM
Location: Seattle, WA
Posts: 506

Try one and find out? That's the most sure way to tell if something will work.

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob