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

Re: More Coupling...



Original poster: "Barton B. Anderson by way of Terry Fritz <twftesla-at-uswest-dot-net>" <tesla123-at-pacbell-dot-net>

Hi Paul, All, 

I want to clarify "modeling dimensions for measured inductances". It's
sometimes difficult to explain this situation because it is very easy to make a
model work vs. simply adjusting dimensions that accurately reflect the real
world. 

Example 1: 
Coiler measures length and diameter of coil. Typically, the turns are calc'd
based on wire size, etc.. There are insulation variations, winding tension
(technique), etc.. that will cause some error. In this case, the spacing
between windings must be a seperate input. In my programs, I calc wire size,
insulation thickness, and the space between each winding. Even closewound coils
have an overall sum of tolerance buildup and effect the result. It is therefore
important to "measure accurately" the "Turns Per Inch" value. I then adjust the
"spacing" to arrive at the correct TPI. Once this is accomplished, the
inductance and "turns" will be very close to measured because our secondary
calc's are nearly rock solid if dimensional input is accurate. Turns are
important in acmi and must be derived correctly. 

Example 2: 
Coiler says "my primary wire size is X and the pitch is Z". Especially with
primary's, the pitch between turns has a tendancy to vary (even on some of the
best primary's). Also, the primary may not be completely round and will have
slight variations. Also, a primary winding is "never a circle" and where you
take the measurement from affects input greatly. "Z" can be better determined
by measuring the OD then inputing a spacing value until the calculated OD
represents the actual. This identifies the average "pitch" and takes into
account all the little variations. Inductances then are better represented.
This is fine tuning the calculated inductances to more accurately reflect the
measured. If there is a large discrepancy between measured and calc'd, there is
probably an incorrect measurement or a primary that is slightly oval. Then, I
think the best that can be performed is to find the average OD. 

Obviously, without modeling in this or similar manner, acmi can be fed
misleading information and therefore will not accurately reflect the coil. The
standard enineering formula's become accurate when the actual input is correct
- and the same with acmi. I've found acmi calc's pretty close to standard
formula's when fed good food. 

Of the coil data thus far, Marco's is mostly a guess as I don't have all the
information. I need to contact Marco with specific questions and run his data
again. Marco's data was a learning experience with acmi regarding inductances.
With his coil, I did an actual inductance approximation (which no, we don't
want to do) but I was missing a lot of information needed. In so doing, his
measured matched acmi very well. This was a big surprise as wire size, number
of turns, etc.. were modeled. It will be interesting to find out how well the
"real" numbers match up. 

(more below) 

Tesla list wrote: 
>
> Original poster: "by way of Terry Fritz <twftesla-at-uswest-dot-net>"
> <paul-at-abelian.demon.co.uk> 
>
> Having said that, you seem to be demonstrating that if the error in 
> the primary inductance is 'put right' by means fair or foul, then the 
> calculated k converges towards the measured k - would you say that's 
> the case?

Absolutely Correct! 
>
> If so then it may suggest our attention should concentrate 
> on sorting out the primary self inductance.

Yes, the primary has the largest potential for error due to having few turns
and concentricity when measuring. I think we need to refine how to properly
measure the primary so that correct information is then input into acmi. I
think only then can you identify if acmi is really high or not on the
inductance (for TC primary's). 

Suggestions: 
* Measure ID and OD center to center of winding. 
* ID: Measure 2 points and average. 
        * Point 1: across start of winding. 
        * Point 2: move 90 deg from start and measure. 
        * Calc: (1 + 2)/2 
* OD: Same method as ID measurement, except: 
        * Point 1: Measure from tap point. 
        * Point 2: move 90 deg in from tap point and measure. 
        * Calc: (1 + 2)/2 

This is just "first thoughts". Maybe there are better ideas? 

Take care, 
Bart