using inverters to power computer gear

when using 12V or 24VDC inverters (or indeed inexpensive inverter petrol generators) to power standard 240VAC computer equipment, what are the implications of the noise inherent in such inverters?

people talk about electrical noise from inverters, and "dirty power", is this really a thing?

 

Yes it is. The ac waveform from cheap inverters tends not to be a nice smooth sine wave, more like a sawtooth. The sudden voltage changes at the tip of the tooth can cause issues with older kit and inductive loads, in some cases will cause airborne or wireborne RFI/EMI, though that is uncommon.

 

As has been said, that can be at least a theoretical issue with cheap/poor inverters. However, as for the concept (with decent kit), bear in mind that all the millions of pieces of computer equipment run via UPSs are effectively being run from inverters.

Kind Regards, John

 

The majority of computer equipment uses a switch mode power supply which turns it all to DC anyway, then back to high frequency AC so even with a simulated sine wave I have never had a problem, you would be very unlucky to have a problem but clearly possible.

In the main inverter generators are not simulated sine wave, although few inverters are true sine waves they just have more steps, there is a problem with the larger inverters in that the fet's are protected by many fuses which also load share between all the fet's with simulated sine wave, but this is not a problem with small inverters.

 

Even the cheapest inverter-generator I can find, the SIP TI800 at £170, claims "• 100% sine wave stable power for running sensitive electrical equipment". It would be interesting to put one on a scope and/or a spectrum analyser to see how clean it is.

 

It's a problem for power supplies if an active PFC is present and non-sinewave inverters are used. Avoid wherever possible and make sure it's a sinewave output like a quality APC unit

 

quite a high volume of flannel there.

aptsys, can you expand on the PFC issues please?
JohnW2, regarding the UPS's; indeed when the UPS is called on to provide backup power its inverter runs.... but I don't think the inverter runs when line power is OK (do they?) these UPS units are clearly designed and accepted worldwide for powering at least IT equipment.

as far as I understand, the problem with so-called "modified sine wave" power supplies is the high-frequency elements which result from FET switching of polarity. (?)
(nearly) all IT gear doesn't actually use AC power, rather it uses 12V / 5V /3.3V DC which appear to be developed from more or less low quality inverter power supplies.
any inverter power supplies immediately imply FET chopping which leads to high frequency "dirt" in the power supply (if filtering is anything less than perfect)
therefore, even if we use pure sine wave mains power, the first thing that happens is SMPSU's which introduce "dirt", which the equipment clearly gets on with.

The only difference I can see is that in the case of a 240VAC inverter genny, we operating the FET at around 100Hz, to produce the 50Hz power (?) and in the case of a DC switchmode, we are operating the FET at 3kHz or therabouts.

I have never worked with vibrator packs seriously but presumably they generated terrible noise and "dirty" power

am I comparing apples and oranges here (by comparing SMPSU's with inverter generators)?

 

There are at least three types of UPS - "online", "offline" and "line-interactive". What you say is true of the latter two, but the "online" ('double conversion') ones, which are pretty common (particularly higher power ones) have the inverter running continuously (and powering the connected equipment), even when line power is available. This has the advantage of there being no switching (or transformer tap-switching) required on loss of power (or low or high supply voltages), so the transition from line power to battery power is completely 'transparent' - the connected equipment is always supplied by the inverter, whether the inverter itself is being powered from the mains supply or the battery.

In the earliest parts of my youth, they (and rotary converters) were really all we had. The vibrator packs were incredibly 'noisy', in all senses, sparking all over the place and affecting all sorts of nearby things (i.e. 'EMC') as well as the connected equipment Fortunately, semiconductor inverters appeared soon thereafter!

Kind Regards, John

 

Ah yes. rotary converters. Played with them a bit at Uni.

One application I remember from school was the HT supply for the transmitter section of a WS No19 (HF radio system used in tanks):



The rotary converter is on the left. Ran off 24V accumulators. You pressed the transmit key and waited for the thing to spin up before you spoke!

 

That brings back memories

Didn't the HT for the receiver work from a vibrator?

The rotary converters were popular as a means to run a 240v shaver from a car battery.

 

Ah - nostalgia Back in the 60s, for years I used an R107 receiver, which had an inbuilt vibrator PSU for use off batteries. The 'muting & R/T" plug on the left, did the switching by connecting an HT supply (via a relay) to 'earth'. If, as often was the case, the plug was not present, one could get a nasty shock by touching the exposed pins (and if enough current flowed through one's body, it would do the R/T switching!) ...



Kind Regards, John

 

Yes - it was in a tin can with 4 pins and plugged into a socket inside the unit IIRC.

We had 2 R107's, I spent hours listening to radio amateurs and chatting on the CCF radio network (5330Khz generally, otherwise 4972.5KHz - funny the things that stick in your mind) using the 19 set (and a linear amp to boot the power) or a 62 set. I had my own callsign - 27C I think. A friend over at Kingston GS was 27D

 

I thought so. I remember buying one of the 19 sets, from a military surplus dealer and carrying it the four miles back home on my back
There used to be lots of such dealers around back then, dealing with war surplus.

I remember one ex-military set I had, had a roller coaster antenna tuner coil. A coil of bare wire, wound on a large former, the tapping contact being made by a small travelling roller. I was rather disappointed to find when I got that home, that the wire had been removed - Guess it must have a precious metal.

 

Seem to remember Pye Cambridge may have been Westminster it was a long time ago some had rotary converters and some the vibrators, you didn't know which until you took off the case.

 

Yes, we had one of those ATUs for the 19. Silver wire I believe.