RE: FW: Re: Tesla Coil Efficiency Test

["Tesla list" <tesla-at-pupman-dot-com>]

Original poster: "John H. Couture by way of Terry Fritz <twftesla-at-qwest-dot-net>" <couturejh-at-mgte-dot-com>


Dave -

Tesla coil efficiency is a tricky business. I found that out several years
ago when I was developing the JHCTES computer programs. At that time coilers
were mostly interested in how to make the calculations work. That is the
tuning equation  LpCp = LsCs  could not be made to work (to agree with
tests) so building a TC was a matter of trial and error construction.

Of course today we know why this tuning equation did not work and now
consider it elementary. However, for someone trying to make a computer
program that resonably represents the TC system the problem with this tuning
equation had to be solved. This tuning equation was fundamental to any TC
program and something like TC efficiency was considered to be of secondary
importance.

The JHCTES Ver 3.3 TC program represents the latest product in my struggle
to come up with a program that could easily be used by any coiler with a
high school education. This program not only solves the tuning equation
correctly but it also produces outputs that are always in tune with the
inputs. This was an improvement compared to other programs available at the
time. These programs could result in untuned coil designs.

 Over the years this program has been used by many coilers to design their
coils. It was natural that many questions regarding TC design would by
raised. One of the many questions that came up had to do with TC efficiency.
The JHCTES program does not show a parameter for efficiency. However,
coilers could estimate TC efficiency  by using the "watts/Ft of spark" in
the JHCTES Ver 3.3 program output. It came as a surprise to me to find out
that the program showed the TC efficiency not only as increasing with
increasing TC size but also decreasing with increasing TC size.

With this increasing/decreasing situation it was obvious that coilers would
have a hard time deciding how to interpret what the program was showing and
this appears to be the position we are in today. For example, the JHCTES Ver
3.3 program shows that if you increase only the number of turns on the
secondary the "watts/Ft of spark" will decrease indicating that the TC
efficiency increases as the TC is made larger. This can give a coiler the
false impression that larger TC's are more efficient. You can find other
examples if you experiment with the program.

It should be noted, however, that when the TC system is truly increased in
size by increasing the input wattage the "watts\Ft of spark" will always
increase indicating the TC efficiency decreases when the TC is made larger.
In other words with the TC system if you take only a PART of the system you
can sometimes show that the TC efficiency increases with an increase of one
of the parameters. This occurs because we are then talking about a different
coil. However, the total TC system efficiency will always decrease when the
TC system is increased in size by increasing the watts input.

Most coilers today prefer to design their coils using a TC computer program
so an understanding of how the program works is important. You can find many
other parameters with the program than the ones actually shown. The TC
efficiency is one of these parameters. There is much more to TC efficiency
and I will be posting more information that shows how the TC efficiency can
be determined when the correct tests are made.

You say the TC efficiency increases when the TC is made larger. All the data
I have seen indicates that TC efficiency decreases as the TC is made larger.
I would be interested in the data and calculations you have available that
shows the TC efficiency increases when the TC is made larger.

The black box efficiency is easily measured by measuring the watts input and
the watts output.
Energy = watts x seconds
The TC efficiency is   Eff = watts out/watts in
A problem arises when measuring the watts out when using sparks as outputs.
However, there is no problem if an incandescent lamp is used as an output.

John Couture

-------------------------------


-----Original Message-----
From: Tesla list [mailto:tesla-at-pupman-dot-com]
Sent: Friday, June 28, 2002 7:26 AM
To: tesla-at-pupman-dot-com
Subject: RE: FW: Re: Tesla Coil Efficiency Test


Original poster: "Dave Larkin by way of Terry Fritz <twftesla-at-qwest-dot-net>"
<teslaman15-at-hotmail-dot-com>

Hi John,

>It is obvious from your explanation below that we are talking about two
>different types of TC efficiency. You are referring to a transfer
>efficiency
>based on  E = 0.5CV^2 and I am referring to an overall black box
>efficiency.
>These two efficiencies will never be the same for any particular Tesla coil
>because the transfer efficiency leaves out many of the necessary TC system
>losses that the black box efficiency takes into consideration.
>
>Both methods, however, can be used  for finding the efficiency of a Tesla
>coil depending on what you are looking for. The transfer efficiency will
>always be greater than the black box efficiency because of losses that are

The transfer efficiency is of interest to coilers because it tells us how
well the spark gap is performing.  The 'Black Box' efficiency would also be
an interesting metric, how exactly does one go about measuring it?

>omitted. The efficiencies found by both methods will decrease as the TC
>becomes larger. This is contrary to most electrical apparatus where larger
>means greater efficiency. One reason is because larger with Tesla coils
>means larger voltages rather than larger currents. The higher voltages mean
>higher losses and they increase faster than the input wattage. This

I would have to disagree here, everything I have ever observed suggests
Tesla coils get vastly more efficient as they get bigger. Coils in the 500VA
class normally burn about 50% of their power in the spark gap, whereas a
50kVA coil *probably* burns 20% or less of its power input as gap losses.
The relationship between volume and area (and hence inductance and corona
loss) would suggest that the power loss/unit inductance will in fact
decrease as the scale of the coil increases.


means
>there is a secondary voltage limit for every size of Tesla coil. Above this
>voltage the insulation breaks down and the coil is damaged.

True, however given that even the largest coils ever built only just break
the 1MV barrier I don't think any decently designed system will ever operate
near the breakdown voltage of its resonator.

>The energies in the primary and secondary circuits represented by  0.5CV^2
>are easily determined if the primary and secondary voltages can be
>measured.
>However, finding the secondary voltage usually means making only an
>estimate
>of what the scope is indicating. The reason is because of the difficulty of
>calibrating the scope and secondary probe which usually gives poor
>accuracy.
>These problems are eliminated when using the black box method for finding
>the TC efficiency.

Well..... A calibrated spark gap, a field mill or one of terry's fibre optic
probes can all give you a secondary voltage to at least +/- 25%, if not
better.


>I agree that finding the output energy of a TC "sparking machine" will
>always be difficult and may be only a guesstimate. The output energy for
>the
>black box method, on the other hand, can be determined very accurately. The

But surely the only true 'black box' output measurement would be the power
coupled to the arc?

>question arises, however, as to how accurately the black box efficiency
>represents the TC sparking machine efficiency.
>
>  For coilers who are bored with output spark tests the efficiency tests
>are
>a good alternative. These tests would help to advance our knowledge of TC
>operation. It should be noted that for these tests you do not need a coil
>that produces the longest spark. You only need coils that are of good
>design
>and properly tuned. It should also be noted that TC efficiencies can only
>be
>obtained by tests and cannot be obtained by computer simulations. There are
>just too many unknowns in the TC system to do this job with computers.

Great advances in that area have been made in the last few years, and I
think we can now computer model coils at least as accurately as I've ever
been able to measure them, and probably better.

-Dave-

>
>John Couture