1 to 3 phase converter question

Built a 5hp idler 3 phase converter with start and tuning capacitors.

I did not have proper start capacitor so I used an old method of connecting two 500 microfarad 450 dc working volt capacitors back to back to provide(in effect) one unpolarized capacitor of approx 500 microfarad 450 volt rating. Back to back means connecting the positive terminals together and using the negative terminals as the unpolarized capacitor. Works fine but for how long?

Does anybody out there have the knowlege or experience to predict the outcome of my experiment?

Reply to
clifton bell
Loading thread data ...

Firstly you'll get a 250uF capacitor not a 500uF (caps in series are always calculated 1/C =1/C1 + 1/C2). Secondly, it won't last long; if the capacitors are equal, each will have

1/2 the AC volts across it and will be reversed biased which is certain death for a polarised electrolytic. The caps WILL overheat and WILL ALMOST CERTAINLY outgass, possibly explosively; at best you will produce a cloud of boiling electrolyte at worst a shower of aluminium shrapnel. Disconnect it NOW. Martin.
Reply to
Martin Whybrow

Capacitors in series are calculated like resistors in parallel, so you have

250 uF, not 500 uF.

Not long at all.

Boom!

There is a way you can use those two capacitors non-polar, but you will need a couple of diodes.

Wire them like this (view with fixed pitch font)

x---->|-----x x----|

Reply to
Gary Coffman

On Tue, 01 Jun 2004 02:33:00 -0400, Gary Coffman wrote: snip

Series connected polar capacitors with back to back shunt diodes work fine as start capacitors - I have one set that has done this for me for over 15 years. They're not normally suitable for continuous operation as run capacitors - too lossy.

The effective capacitance value is a trap that I also fell into a long time ago. At first sight, since the diodes appear to short out alternate capacitors on alternate half cycles, it would seem logical that the effective capacitance would not be halved and would remain at 500uf. This can only be true if each diode carries the full load current on the alternate half cycles.

This is not the case. After the first initial switch on transient the diodes charge each capacitor up to the peak value of the supply voltage. Because there is no DC discharge path, although the junction point is waving up and down at half supply voltage, the DC level of the junction point remains and is sufficient to reverse bias both diodes throughout almost the whole of the supply cycle. The diodes only conduct for a tiny period at the positive and negative peaks - just sufficient to make up the DC leakage current of the electrolytics - a few milliamps. This plays no part in modifying the effective value of the two series connected capacitors which remains at 250 uF.

Jim

Reply to
pentagrid

How well do series connected aluminum elecrolytics work without the shunt diodes? I am under the impression that the AC elecrolytic caps sold for motor starting are just two caps connected in series. If that is not the case, how are motor starting caps made.

Dan

Reply to
Dan Caster

How well do series connected aluminum elecrolytics work without the shunt diodes?

Not appropriate.

To place in series, just do so.

The capacitance rating is 1/2 of the individual cap; the voltage rating is doubled.

I am under the impression that the AC elecrolytic caps sold for motor starting are just two caps connected in series.

True for so-called non-polarized caps as used in speaker crossovers.

If that is not the case, how are motor starting caps made.

Specially designed for high-ripple, ac service; have a definite maximum connect time before they heat up and explode; have a definite number of "starts" per hour; have a definite number of "starts" over their lifetime.

Reply to
Peter H.

Back to back series connected electrolytics work without shunt diodes - the diodes are really belt and braces protection. The use of diodes has the advantage that the capacitors are working very close to their normal design operating conditions - the only difference is that their ripple current rating is being exceeded and this is OK for intermittent use.

Without the diodes the capacitors are subject to large peak reverse current flow during the switch on transient. While they appear to accept this without complaining it is way beyond any published rating so I prefer to fit diodes.

Non polar tantalum capacitors are commonly a couple of standard polar capacitors mounted end to end inside a common outer tube. Diodes are not fitted.

Non polar aluminium electolytics are usually a single component. They are wound as a normal polar capacitor but with both foil electrodes carrying an anodic film.

Jim

Reply to
pentagrid

Thanks. I did do a little googling but did not find the answer. I am under the impression that polar aluminium electrolytics are made and anodized after they are assembled.

Dan

Reply to
Dan Caster

I think you are correct. My wording was ambiguous. I only wanted to make the point that there was no "negative" electrode, both electrodes carried the anodic film. I've no direct knowledge of this but I'm pretty sure that the anodic films are formed in situ after final assembly of the capacitor.

Jim

Reply to
pentagrid

PolyTech Forum website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.