Inverters - again

[meow2222]

ps fwiw motorgen sets were only around 60% efficient, so I dont see how including one could ever improve the efficiency overall. Even in the 1920s LT psus did better than that.

NT

[Dave Plowman (News)]

That's sort of how a car dynamo works.

[Andrew Gabriel]

Yes -- I wasn't very clear but I meant either a square wave or a higher frequency sine wave was an improvement. A 50Hz sine wave is a long way from ideal for a computer load.

Yes. Also whilst the power factor as you go into the rectifiers may be quite low, the motor/gen set inertia completely hides that from the utility supply, which sees near unity power factor.

[The Natural Philosopher]

No, it isn't.Its in fact as it happens designed to run off anything witha peak volatge of between about 300 and 400V,since all the PSU does to start with is rectify what it sees and smooth it - if that is mains, thats about 430VDC. It could just as easily BE 430VDC.

In fact I worked on a project once where the same power supply could be used on 48VDC, 120VAC, 230VAC with no reconfiguration required.

In short everything from 48-250V AC or DC was able to make it work.

It's treansformers that get hot under the collar with non sine mains. But even then not that much.

[The Natural Philosopher]

..generally found on the inverter anyway to try and keep RFI out of the picture.

[Arfa Daily]

That's rather higher than any switch mode power supplies that I work on, for the front end DC bus. Around 360v on the back side of the reccy is more typical. In fact the highest one that I can think of is about 390v after a front end resonant PFC stage.

This particular supply will work down to about 35v ac input, and still produces 390v across the main filter cap with that level of input voltage. However, it cannot deliver its rated output currents with that level of input, and I'm surprise that the one you cite could, without any reconfigurataion, as you say.

Many modern switchers work by the skin of their teeth and, whilst I agree with what you say about the fundamental principal of deriving the front end bus voltage *should* be independant of what you feed in, they are never-the-less designed with a sine wave input, with a known and predictable dv/dt, in mind, as that is what they would be expecting to see when on a mains input.

Waves with fast transients and rise times, can wreak havoc with wound filter components in the mains feed, and cause excess dissipation within filter network caps, and VDRs. Sudden polarity changes and voltage steps, can also bring the reccies close to or beyond their maximum ratings. Likewise, where the value of the main filter cap for the front end bus, is small and barely adequate for the job, as is often the case with small supplies, sudden changes in polarity and voltage level applied to the bridge, may well cause the cap's maximum ripple current rating to be exceeded.

Just a couple of weeks ago, I had a small switcher in for repair, which had been put on a 'modified sine wave' inverter in a motorhome. The front end was blown to pieces, having suffered a cascade failure. The owner assured me that prior to this, the item has been working fine in his house for a couple of years on that same power supply, and this was the first time that he had put it in the motorhome.

Arfa

[Andrew Gabriel]

You would normally see 340VDC from full wave rectifying 240VDC. One which has caught out some people are bathroom shaver sockets. They can deliver up to 270VAC at small power draw. Things like saver/toothbrush chargers are well below the isolating transfomer's full load rating, and can end up being fed with higher than expected mains voltage (that would be 381VDC after rectifying). The instructions for shaver sockets often advise using the 120V side for low power draw appliances, but the instructions are chucked out by the fitter and never seen by the householder;-)