[TCML] MOT Measurements

[bartb]

Hi Ted,

Some of my emails don't seem to be getting through. I'm not receiving 
the ones I sent out nor did I get your 200V measurement. Thankfully, you 
left the previous message in tact with your 800V update.

Your test as performed certainly shows 6.3H at L2, but now take it to 
the next level and calculate L1 and look how low it is. Then extrapolate 
currents and voltages which should be measured based on the inductances.

With the open circuit primary feed measurements and the resulting 
inductances, even the simulations get it right. For a MOT I measured, I 
showed 115.5mH at L1 and 39.5H at L2. Measured k at 120V was 0.967. With 
the secondary shorted, I measured 43A at L1. The simulation shows 42A 
with the secondary shorted (pretty close and would not be if the 
transformer model was wrong). This is just one aspect, but when the 
simulations match closely to the measurements, one has to assume the 
transformer model is correct.

I tried your method and applied 700V to L2 and measured 185mA. Thus, 
Z=3784, L2=10H, L1=32.8mH. I ran the same test again as above and the 
simulation then showed 132A primary current, which is very wrong. 
According the simulator, these inductances are not correct even with 
large coupling changes.

There's nothing wrong working L2, but you need to do it with the primary 
in open circuit and L2 should be at it's rated output voltage if we are 
talking about inductances at input value. The fact is, the inductance of 
the MOT changes with current as does the reactance of course. This is 
easily seen in a series of open and short circuit measurements in 
incremental voltages. For example, my MOT at 100V input is very 
different than 120V input. Short circuit primary current measured 29A at 
100V input. L2=87.7H and L1=270.7mH at 100V input. I change those values 
accordingly in the transformer model, adjust coupling to 0.983 (a 
measured value at 100V input), set the AC input into the model at 100V, 
and presto, 28.7A is what the simulation shows for primary current with 
the secondary shorted (just like in the real world).

I am not relying on simulation, but I am verifying with simulation that 
the currents and voltages in each part of the circuit match closely to 
measured values. In order for this to occur, the transformer model must 
be set correctly for the applied voltage. Thus, if one tries to simply 
change the source voltage in a model without changing the transformer 
model according to measured values, then the model is flawed. This is 
due to using a linear transformer model for a very non-linear 
transformer (which is why k and L's need to be adjusted accordingly).

I think your far better off using the open circuit secondary method and 
feeding the primary to find the correct inductances at the feed 
voltages. And many coilers have been using this method for quite a while 
with various transformers and it has proven correct over and over again 
by both measurement and simulation.

Regards,
Bart

tesla wrote:
> Update
>
> With 800v rms excitation 0.4mA flows so again Z=2000, same inductance as at
> 200v excitation
> Rgds
> Ted
> From: "tesla" <tesla at paradise.net.nz>
> To: "Tesla Coil Mailing List" <tesla at pupman.com>
> Sent: Monday, January 05, 2009 7:40 PM
> Subject: Re: [TCML] MOT Measurements
>
>
>   
>> Hi Again all
>>
>> Did measurements and advise.
>>
>> Single 600 Watt MOT, magnetic shunts in:
>>
>> With primary sided shorted (replicates low source Z mains) 200v rms of
>> secondary excitation causes 100mA to flow.
>>
>> The modulus of the impedance is therefore 2000 ohms. As the resistive
>> component is small compared to this the modulus is almost the same as the
>> Inductive reactance. So from 2pifL=Z L is 6.3Hy
>>
>> What are the measurement conditions used to give results as high as 40Hy
>>     
> ??
>   
>> I believe my technique is validly measuring the series output inductance
>>     
> of
>   
>> an operating MOT as an external voltage generator can replace the one
>> provided when the Xfrmr is operating. Have I missed something in my
>>     
> thinking
>   
>> ??. I will try the same measurement with an AC excitation of about 1kv to
>> bring the xfrmr closer to normal operating conditions but I expect this
>>     
> will
>   
>> lower further the Xl value as saturation impacts
>> Ta
>> Ted
>>
>>
>>     
>
> _______________________________________________
> Tesla mailing list
> Tesla at www.pupman.com
> http://www.pupman.com/mailman/listinfo/tesla
>
>