[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [TCML] NST Measurements



Hi Gary,

This average ratio I was looking at is not working. I used a second 15/30 to power the test NST measuring HV up to 17kV output open circuit. I started at 1250V in and the average ratio then was high (200 falling down to 160 as voltage was increased). The numbers I received were simply coincidental at the low 124V feed I guess (funny how that seems to happen more times than it should).

Back to the drawing board.

Even with primary feeds and secondary measurements, only at 10V input did the turns ratio equal the rated Vout/Vin. Below and above this value, it varied. However, in the data below which is all measured, I show the ratio's and although it goes above the rated value of the transformer (in open circuit), it is not overly high when actually measured.

Using methods like feeding the secondary is not working out (yet). As you can see, I was looking at low voltages, but I never went below 10V. When I did, the ratio just kept falling, so that upsets my apple cart. Everything else is strait forward, but the turns ratio measurement is not (quite the opposite of what I took for granted). Neat problem to solve. I think these shunted transformers are a bit more complex than anyone assumes. If we know the transformer ratings, were ok, but if not, then large error is probable.

Feed     Meas.
Pri      Sec      Ratio
-----    -----    --------
10.0     1250     125.00
20.1     2680     133.33
30.0     4130     137.67
40.2     5590     139.05
50.1     7030     140.32
60.0     8380     139.67
70.1     9770     139.37
80.0     10970    137.13
90.0     12350    137.22
100.2    13550    135.23
110.3    14850    134.63
120.2    16390    136.36
130.0    17120    131.69

Take care,
Bart
PS. Anyone know any Neon Sign Transformer engineers?


bartb wrote:
Hi Gary,

I have done this, but not checked it. So, I did just now. The 15/30 measured 175X at 124.2Vin. So this presents a possible large error depending on where the input is in the stream.

I believe we should be looking at both sides equivalently. If you feed 120V to sec and measure 0.784V at the primary, you should then feed 0.784V (or as near as you can get) to the primary and measure the secondary. Take the both ratio's and average. So I checked this.

Measurements of my 15/30:

Feed sec = 124.2V, pri = 0.708V = ratio of 175.4
Feed pri = 0.76V, sec = 55.0V = ratio of 72.4
Average (72.4 + 175.4) / 2 = 124 turns ratio (125 expected)

Decent considering meter error (and using a single meter). I'll try to go through this at various voltages to see how it plays out.

Gary, can you confirm on your NST's? If this works out decent on your NST's, then it's an easy procedure to ensure turns ratio accuracy as well as inductance ratio on transformers with no data (or confusing data).

Thanks for looking at that.
Bart

Lau, Gary wrote:
Hi Bart,

As often as I have dispensed the same advice to measure and feed the mains voltage into the NST secondary and measure the primary to determine the sec/pri ratio, I don't recall ever having actually done that to mine. So, I just pulled out a 15/30 and a 15/60 NST. The 15/30 measured a ratio of 153X, so assuming a 120V primary, predicts a 18.3kV secondary.

The 15/60 measured a ratio of 145X, so assuming a 120V primary, predicts a 17.4kV secondary.

The meter that I used is a simple Fluke DMM, not RMS. But given that the NST is going backwards, I wouldn't think that it's getting near saturation and the associated waveform weirdness. Unfortunately I don't own a HV probe that I would trust to measure it in the normal operating sense. Have you done this and gotten closer to the faceplate voltage rating?

Regards, Gary Lau
MA, USA
_______________________________________________
Tesla mailing list
Tesla@xxxxxxxxxxxxxx
http://www.pupman.com/mailman/listinfo/tesla

_______________________________________________
Tesla mailing list
Tesla@xxxxxxxxxxxxxx
http://www.pupman.com/mailman/listinfo/tesla