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

Re: [TCML] NST Measurements



Hi All,

I made some extreme low voltage measurements:

Feed      Meas.      Integer
Vsec      Vpri       Ratio
-----     -----      -----
120.2     684mV      176
110.5     625mV      177
100.7     565mV      178
090.1     502mV      180
080.5     445mV      181
070.5     386mV      183
060.1     326mV      184
050.4     271mV      186
040.6     216mV      188
030.1     158mV      190
020.1     104mV      193
010.2      52.5mV    195 **
008.15     42.1mV    194 **
006.06     31.1mV    195 **
004.02     20.6mV    195 **
002.106    10.8mV    195 **
001.069     5.5mV    194 **

This data presents some obvious questions.

When I managed a low enough level (10V down to 1V), the voltage ratio showed about 195 consistently (what was affecting the voltage ratio seems to have stopped). Did I manage to get below shunt inductance influence? Is it possible the turns ratio is 200, yet the voltage ratio is entirely something different thanks to the shunts at greater voltages (and the manufacturers account for this in their Voc rating)?

My measured output at 120.2V input was 16390 which is a voltage ratio of 136 (not 125 as spec'd). Voltage ratios and assumed turn ratios are very different. There is now no doubt about that, but still, how to determine the real turns ratio is a mystery (and maybe not possible?).

Bart

bartb wrote:
Hi Dave,

Yes, I agree. The magnetic shunts are causing the non-linearity. As I mentioned a moment ago in my reply to Phil, the voltage ratio is non-linear because of this. Were trying to use the voltage ratio to determine turns ratio. This won't work "IF" the shunts have influence because it affects the voltage ratio. I think everyone measuring is seeing this affect.

The problem I have is the inductance factor is based on the turns ratio. So, how to remove shunt influence? The only way I can think of is to use "extreme" low voltages. Maybe instead of inserting 120Vac to the secondary, we need to bring down the voltage to something between 1 and 10V input to the secondary and measure the millivolts at the primary? There must be a point at which the shunts have little affect and the voltage ratio becomes linear enough for a decent approximation.

Take care,
Bart

sparktron01@xxxxxxxxxxx wrote:
Bart

Is it possible that the inductance varying is hosing the apple cart?

Instanteous voltage for an inductor (complete) is:
V = iR + L (di/dt) + i (dL/dt)

The last term in "linear" circuits tends to zero and (usually) has
no bearing on circuit.  But with a magnetically shunted circuit
the third term starts influencing the second, and vice versa
(i.e the whole circuit becomes VERY non-linear...)

Regards
Dave Sharpe, TCBOR/HEAS
Chesterfield, VA. US

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