turns ratio

I was demonstrating that with much more reasonable turns counts (than

1000000 tuns) it was possible to have ppm differences.

condsider a 1001 turn autotransformer tapped one turn fron each end. with that you can get the two ratios above, with just over 1000 turns. that's the sort of thing that could be wound by hand, given time to concentrate.

continued fractions. will often get uou two numbers with a product that's less than 1000000 and a ratio less than one ppm different from any target. some rations can't be got this way.

what do you mean? most of the interesting words in there seem somewhat ambiguous.

you don't need to.

--
?? 100% natural

--- Posted via news://freenews.netfront.net/ - Complaints to news@netfront.net

that

measure

with

And the signal is less too. The phase shift error, and the copper loss as a fraction of the excitation voltage, is the same if u is constant. It's a linear system. The disadvantage of using a low voltage is that low AC voltages are harder to measure accurately.

If permeability decreases at low drive, the *fraction* of the voltage lost to IR drop will increase. Fortunately, it's a quadrature error so it's usually small.

--

John Larkin                  Highland Technology Inc
www.highlandtechnology.com   jlarkin at highlandtechnology dot com   

Precision electronic instrumentation
Picosecond-resolution Digital Delay and Pulse generators
Custom timing and laser controllers
Photonics and fiberoptic TTL data links
VME  analog, thermocouple, LVDT, synchro, tachometer
Multichannel arbitrary waveform generators

BRAVO!!! I do appreciate creativity. Let me paraphrase. For some ratios, (integer ratios) it's easy to get within a PPM. If you don't need it to be exact, it's easier to approximate. In the general case, you can use fractional turns, at least in theory. But if you want to scale it up to integral turns, you need about a million turns. "1.001.001" x 1000 = 1001001.

Points for creativity that appears to be outside the realm of the original question and ends you up right back where we started.

I do loves me a good pissing contest, but I'm gonna need more beer.

Wind your Frankenformer any way you want. Be sure to tell the vendor about that .001 turn when you get the quote.

Oh crap! Just had another question. round numbers... Say you have an E core with 100 laminations. What happens when you put a spacer between 1 and 2 and skip that first lamination on the first turn. Is that .99 turns? Or does that just add a bunch of core loss.

be

great

that

measure

with

Certainly, any procedure can be carried too far.

When I recommended a low excitation level to avoid the IR drop in the copper, I suppose a reader might have reasonably asked, "how low is low"?

The main thing to avoid is saturation of the core. Real transformers usually operate with the core somewhat into saturation for economic reasons. Saturation in iron cores is not sudden (except for "square loop" cores, not usually used in

60 Hz power transformers or audio transformers). So for an applied sinusoid of voltage at the rated level for a winding, the exciting current typically shows considerable effect from the non-linearity of the core.

A person trying to determine the turns ratio by measuring voltage ratio might start by applying full rated voltage to the excited winding, especially if they wanted to get ppm accuracy. This would probably give an exciting current with a very peaky waveshape. For a typical 60 Hz power transformer, the exciting current can be monitored with a scope and the applied voltage reduced until the exciting current looks sinusoidal. Doing this with a small power transformer and exciting the 120 volt winding, the current doesn't look sinusoidal to my eye until I reduce the applied voltage to about 15 volts. That's without using the FFT function of the scope to determine harmonic levels.

So, when I say to use a "low" exciting voltage, I mean to use a voltage sufficiently low that the current looks sinusoidal on a scope.

Power transformers are designed for and used at a single frequency, but for audio transformers the situation is more complicated. At the low frequency end of their usable frequency range, core saturation causes the same problem. The exciting voltage should be low enough to avoid non-sinusoidal exciting current. One might think to use a higher frequency, but then capacitive currents increase in the winding, and those currents don't contribute fully to the flux in the core, and this causes an error in the ratio measurement. Judgement will have to be exercised to find a reasonable compromise.

To get better accuracy, one might measure the DC resistance of the excited winding, and the exciting current, and then use the calculated IR drop to compensate the measurement.

The classic text, "Magnetic Circuits and Transformers" by the MIT staff, after pages of approximate calculations, has the following significant sentence:

"There are, however, no electrical measurements from which the turns ratio can be determined exactly."

You still need the 999:1 outer-leg-balancing windings to enforce the correct ratio, otherwise that one lam carries leakage flux.

If you get the ratio of core areas equal to the unbalance ratio, you'll keep core utilization high, which means all legs run at the same flux density. This probably isn't a big deal in signal transformers, where operating levels are far enough below Bmax that a factor of 2 won't *really* do a whole lot to linearity or accuracy.

Read my posts again in more detail, preferably later. TGIF, I think I shall get a beer or two myself :-)

Tim

--
Deep Friar: a very philosophical monk.
Website: http://webpages.charter.net/dawill/tmoranwms

On Tuesday, May 1, 2012 6:30:22 PM UTC-7, Phil Allison wrote: [concerning turns ratio with measured voltage ratio of transformer]

If you excite one winding, and measure both, the excited winding has Ve= Ie*Re + Le * d(Ie)/dt and the driven winding Vd = (Nd/Ne)* (flux_coupling) * Le * d(Ie)/dt

So, the flux_coupling term is where you can get a couple percent error.

The error due to resistive losses is an out-of-phase signal, so if your excitation is sinewave and you do phase-sensitive voltage measurements, that part vanishes. If you have true RMS measurement, it's a very small correction at frequencies where excitation doesn't make much quiescent current. IF, however, you excite with square waves, there's lots of frequency range in the signal, and that can get complicated by losses in the harmonics.

Tell me, when the primary is supplied from an ideal voltage source, what's the (unloaded) secondary voltage of a 1:1 transformer with: k = 1 k = 0.999 k = 0.9 k = 0.5 k = 0 ?

If you provide only vitriolic reply I will accept that as your forfeiture of this argument.

Have a nice day,

Tim

--
Deep Friar: a very philosophical monk.
Website: http://webpages.charter.net/dawill/tmoranwms

"Tim Williams" Phil Allison

** k simply does not change in the scenario under discussion.

So your pedantic, math only assumption is irrelevant.

And so are pricks like you.

.... Phil

Sure it does. Real transformers are maybe k = 0.9999 tops. Since at k = 0, no secondary voltage is generated, there is necessarily less voltage also at k = 0.9999 vs. k = 1.

If you state k has no effect, then how does one inductor on the other side of the galaxy not happen to couple into one here, since k ~= 0 in that case?

Argument forfeited, thank you :)

Tim

--
Deep Friar: a very philosophical monk.
Website: http://webpages.charter.net/dawill/tmoranwms

"Tim Williams

** Bullshit.

**Clearly, that is not what I posted.

False arguments are the ONLY things you have and you are mindlessly besotted with them all.

Just like all the other know nothing, autistic jerkoffs on usenet.

... Phil

You can get magnet wire with different colored enamel. Made for bifilar winding.

--
"For a successful technology, reality must take precedence 
over public relations, for nature cannot be fooled."
                                       (Richard Feynman)

Think of a rocket whose mass decreases as fuel is burnt.

--
"For a successful technology, reality must take precedence 
over public relations, for nature cannot be fooled."
                                       (Richard Feynman)

Good morning Phil, all the way down under!

Jamie

Let me put my 2 cents in !

Objects acting as Eddy currents in the Aether !

There, I said it! Who knows, I may even be correct ;)

Jamie

Jerk is an important parameter in the design of roads and so on. If you make a road curve from straight segments joined with circular arcs, the lateral acceleration has to go from zero to maximum, instantly, if you're going to stay on course. Same with trains and elevators. You want to build up to maximum acceleration gently so that people have time to compensate. A motorman who went from zero to full power instantly would cause injuries if anyone was standing.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC
Optics, Electro-optics, Photonics, Analog Electronics

160 North State Road #203
Briarcliff Manor NY 10510
845-480-2058

hobbs at electrooptical dot net
http://electrooptical.net

I saw it. And saw him miss it. Now, we get to see him step deeper into his "manness" as he makes excuses for why he blatantly missed it the entire time.

Talk about being wrong. Bet he never asked for directions either, two decades ago before all this mobile gadgetry came about

He's "one of those".