Inverter as a power source

[orac02]

Hi
Has anyone out there tried using a DC to AC inverter for the input to thier system?

I recently designed one at work for a customer, to supply a pure sine wave at 230VAC running from a battery. This was supposedly to improve on main bourne noise.

I took it home and ran my CD and amp from it expecting no difference, at least nothing positive.

Much to my supprise there was a distinct improvement, Sound stage, bass and overall clarity.

Has anyone else got any experiences or thoughts?

One thing i have established so far is it sounds bet(probably) with a large battrey. But still better than the mains with a 17 Ah one

Cheers

Steve

 

[Eva]

Have you tried it in a blind test where you have to guess which power source is being employed from what you listen and without previously knowing it?

p.s: It's sad to see engineers entering the subjectivist field.

 

[N-Channel]

Eva,

Perhaps Steve is convinced that it sounds better because of the absence of line noise often found on conventional AC mains.

However, I would be suprised if there wasn't any noise on the output from FRI generated from the DC-DC step-up section.

(The Other) Steve

 

[Eva]

I said that because I have been considering such a sine-output inverter design recently and there are a lot of chances to generate gentle RFI with such a circuit. Note that the most harmless part is the DC-DC converter, since the output PWM section is a high voltage class-D amplifier that is forced to perform hard switching of 200V to 400V over a *conducting* diode in each clock cycle (If two phases are used to obtain 230V AC, then hard switching happens twice in each cycle)

The output filter of such converters is also quite a can of worms, because you have to chose between either really bulky output inductors and capacitors, or living with a few volts and amperes of output ripple current and voltage at the switching frequency.

He must have certainly designed a really damn great converter if he thinks that its output contains less non-50Hz harmful stuff than any mains line.

 

[sagarverma]

Steve,

there is nothing to justify what u experienced except saying that it is your perception or like that''what u wanna hear''.

please subject yourself to a blind test.

SAGAR

 

[N-Channel]

OOPS! I meant RFI, not FRI. It's getting late here in Michigan. Gonna go to sleep now......

 

[orac02]

Hi
Not that I consider it to be an obvious advantage fron the point of view of unwanted noise, but the inverter switches the 12v to produce the sine wave and is then stepped up through a (relatively) standard toroidal transformer to produce the 230v .Not as Eva suggested it might be done.

Secondly, you are absolutely right , at the moment there is a measurable ripple volltage at the switching frequency present on the ac output. (improvable)

So I expected it to be dreadful as a source for an audio system. Thats why we hadnt bothered putting it in a case (then).

I was still working on the principal of "I dont think this will work but if someone wants to buy it........."

The difference to me seemed large enough to accept the "wot no blind test" to see if anyone else had any ACTUAL experiences.

Off to set up a blind test..!

Cheers
Steve

 

[orac02]

Hi
How about this idea.

Although there is some residual switching frequency ripple on the output this is at a fairly low frequency, the unit is switcing at about 15Khz. At this sort of frequency the input electrolytic capacitors of the amplifier (for instace) will filter it out, as they do the 50Hz ripple.

But

The impedance between the power source (battery) and load
compared with
the impedance between the power source(power station) and the load.
must be less, especially with increasing frequency.

This surely cant be a bad thing!

Thoughts?

Steve

 

[lagoausente]

I was still working on the principal of "I dont think this will work but if someone wants to buy it........."

Do you sell it?

 

[KISS]

It doesn't have to be an inverter.

I noticed a major improvement in bass if I use a sine wave voltage regulator. I got a used Sorenson that retails for about $800 for $100.

 

[jamesrnz]

i wonder if you could divulge some details,

i was wodering what kind of efficiency you are getting?

i wonder if you are experiencing some power factor correction at the bass power draw?

 

[KISS]

In my case, I can say that the AC mains voltage does change in response to BASS, thus an AC regulator kinda gives you a regulated supply because the AC input to the amp is vary constant. Never tried a regulated tracking supply for an amp.

 

[IVX]

I think that 15Khz is too low switching for audio purpose. It's well audible, and too much big LC filter (rather two big LC filters) required. If your invertor will run at 100Khz, those filters also will much simpler and compact. However, if you want to achieve fairly better precision of sine vs usual wall outlet (2-7%), you must keep output impedance lower as possible, so any stepup trafo or feedback before LC filters is back off. For instance, PS Audio sell P300-P1000 regenerators, the euro version based on the class AB amp and stepup trafo (AFAIK 100:230VAC), question is -how much THD will give such regenerator, if USA version (without stepup trafo) at the real loads give 2%? BTW, they promise .05% in the specs, but if you'll ask about the kind of load, you'll be surprised that .05% THD means for no load condition! Example #2, check this out http://www.audiophileaps.com/pdf/comparisons.pdf . Indeed, PWM approach do a miracle vs class AB+trafo, but if you'll keep try to find THD figures for this PWM regenerator (it will not so easy), you'll surprised again.. 3%!! So really good AC regenerator must be class D amp based, which use after two LC filters feedback for fairly low residual.
PS: Such approach gave me <4mOhm output impedance, 12mVAC RMS 100Khz residual (i can't measure how much it came from the 230: 38VAC trafo even, just a zero), THD@real_load just a zero again vs those with 2-3% (500-1000 times lower), 140x140mm PCB (PFC+ATmega8+amp), 15A current limit, actually, continual output power is well scalable by bigger heatsink or cooler, efficiency >95%(TBD).

 

[lagoausente]

I think that 15Khz is too low switching for audio purpose. It's well audible, and too much big LC filter (rather two big LC filters) required. If your invertor will run at 100Khz, those filters also will much simpler and compact. However, if you want to achieve fairly better precision of sine vs usual wall outlet (2-7%), you must keep output impedance lower as possible, so any stepup trafo or feedback before LC filters is back off. For instance, PS Audio sell P300-P1000 regenerators, the euro version based on the class AB amp and stepup trafo (AFAIK 100:230VAC), question is -how much THD will give such regenerator, if USA version (without stepup trafo) at the real loads give 2%? BTW, they promise .05% in the specs, but if you'll ask about the kind of load, you'll be surprised that .05% THD means for no load condition! Example #2, check this out http://www.audiophileaps.com/pdf/comparisons.pdf . Indeed, PWM approach do a miracle vs class AB+trafo, but if you'll keep try to find THD figures for this PWM regenerator (it will not so easy), you'll surprised again.. 3%!! So really good AC regenerator must be class D amp based, which use after two LC filters feedback for fairly low residual.
PS: Such approach gave me <4mOhm output impedance, 12mVAC RMS 100Khz residual (i can't measure how much it came from the 230: 38VAC trafo even, just a zero), THD@real_load just a zero again vs those with 2-3% (500-1000 times lower), 140x140mm PCB (PFC+ATmega8+amp), 15A current limit, actually, continual output power is well scalable by bigger heatsink or cooler, efficiency >95%(TBD).

Is any advantage if you want to feed a preamp with a battery, use an inverter, and then a linear psu? or use directly a DC-DC converter?

 

[IVX]

Is any advantage if you want to feed a preamp with a battery, use an inverter, and then a linear psu? or use directly a DC-DC converter?

Sorry, i'm not sure, if i understand you correctly.

 

[Tweeker]

Is this a mobile preamp psu? If not it would be best to series batteries and feed the psu directly, perhaps with a capacitor and/or other passive components, or a linear regulator.

 

[lagoausente]

yes, is for a mobile preamp. I´ll try both, the DC-DC converer ,and the batteries to check the posible noise diference.
Thanks for your replies.

 

[Tweeker]

Id probably use the DC-DC converter followed by a little RC filtering. More efficient and cheaper than the other approach, said watts can be blown in a linear regulator if need be instead.

Most inverters contain what amounts to a DC-DC converter followed by a not so great amplifier.

 

[lagoausente]

said watts can be blown in a linear regulator if need be instead

What do you mean?

 

[rcavictim]

I recently tried running my MingDa MC34B tube amp, 28 watts/ch, P-P 6CA7`s in full stereo (requires 130 watts from 120 volt, sine wave mains) from a small cigarette lighter 160 watt Xantrex modified sine wave inverter from an 18 AH, 12 volt gell cell. The amp sounded fine and was black silent, but I measure the B+ at 25% or so lower than design (about 320 volts instead of 420 volts). As I understand it this is because there are no peaks in the square wave from the inverter to charge the amplifier B+ capacitors fully. This being the case I wonder if the reasson our poster heard an improvement in the sound, inotherwords sound different enough to notice, was because the amp was running in a totally different power regeime (different operating points on the tubes)? Just happenstance luck could have made this different regieme sound better.

I can buy that it actually does, if my hypotheseis happens to be the case. Power output will of course be less.