I am currently playing with tuners... Specifically link coupled... And
I have been making my own condensers... Part of the exercise includes
efficiency 'that' word again of power transfer, more specifically
tank coil losses... I had been using my fingers as my calibrated
temperature differential meter but I have gotten the tuners engineered
to the point where there are not enough differences to tell by feel...
So I purchased an Infrared measurement gun...

fascinating little instrument - did you know that on a clear day with
the air temperature +12.1 F and zero wind, that a 5 foot diameter
black rubber tractor tire facing the clear sky to the North can have a
surface temp of -2.6 to -3.0 F? I didn't but that is what I found...
OK, I digress again

So, I have been running power soak measurements of the tuner components
with the power source remaining energized during measurements - the
nice part of a non contact instrument... The measurements turned up
surprising results to me...
The continuous power level was 680 watts input to the tuner box at 3508
kc ( two Bird meters, uncalibrated and the power averaged / 670 and 690
W. respectively)... The power soak time was ten minutes total... The
outside air temperature was +12.1 F throughout the measurement period -
mid day... The tuner components were +12.8 at the start (- + 0.2 F)...
The tuner parts are in a plastic waste bin with the top on, except
during measurements, to control air currents... The box sits about 4
feet away from the wall of a large building and shaded from direct
sunlight weak sun at the time of measurement...
The coil is ~4" diameter and consists of 20.5 feet (22 uH) of #10
enameled wire wound on a paper mailing tube, and glued down with
epoxy.. The link coil is the same wire (2.2 uH) wound on PVC tubing
and slid inside of the tank coil tube - it was not available for
temperature measurement, a finger slid inside of the PVC tube did not
detect any temperature rise power off!... The tank condensers are
aluminum and are ~24 square inches in area per plate (a pair of caps in
a series tank configuration)...
At 5 and 10 minutes the top was popped and the surface temperature of
the tank coil and condensers were measured - taking about 30 seconds -
and the top replaced... The results for both times were within a few
tenths of a degree...
The coil was found to be +12.9 to +13.1 F and the condenser plates were
+31.5 to +32.3 F... A result which sent me away mumbling to myself...
Not what I expected...
This is consistent with an earlier test in which the tuner was powered
for 10 minutes at 1800 watts with the exciter sending dahs at 20 wpm (
typical contesting environment)... The components were finger measured
power off and the coil felt to be barely above the starting
temperature and the plates to be just a bit warmer than the starting
temperature, which prompted the purchase of the infrared gun because it
couldn't possibly be right - we all know that condensers are 100%
efficient and coils are lossy - right?...

denny / k8do

couldn't possibly be right - we all know that condensers are 100%
efficient and coils are lossy - right?...


Good work. Congratulations on doing some real experimenting and thanks for
reporting.
What do you attribute the losses in the cap to? Could it be contact
resistance between the plates and the rotor? Would it be possible to
substitute some commercial cap temporarily to compare?

Rick K2XT

Rick wrote:
couldn't possibly be right - we all know that condensers are 100%
efficient and coils are lossy - right?...


Good work. Congratulations on doing some real experimenting and thanks for
reporting.
What do you attribute the losses in the cap to? Could it be contact
resistance between the plates and the rotor? Would it be possible to
substitute some commercial cap temporarily to compare?

Rick K2XT



Show me a 100% efficient cap first, then I will show you where to patent
it and sell it. You will then have the money to purchase my bridge in
the desert!

Regards,
JS

John Smith wrote:
Rick wrote:
couldn't possibly be right - we all know that condensers are 100%
efficient and coils are lossy - right?...


Good work. Congratulations on doing some real experimenting and
thanks for reporting.
What do you attribute the losses in the cap to? Could it be contact
resistance between the plates and the rotor? Would it be possible to
substitute some commercial cap temporarily to compare?

Rick K2XT


Show me a 100% efficient cap first, then I will show you where to patent
it and sell it. You will then have the money to purchase my bridge in
the desert!

Regards,
JS

Of course, there must be lossless caps somewhere, huh? Because a cap
with no dielectric would suffer no losses (ignoring the resistance of
the plates to the inrush, exhaust of electrons), huh? So then, a cap
consisting of plates in a vacuum would have no dielectric and no loss.

Strange, I seem to still notice a loss in such a device when examined
mathematically. Perhaps the ether is serving as a dielectric?

Chuckling,
JS

Rick wrote:
What do you attribute the losses in the cap to?

Besides I^2*R losses in the leads, there are dielectric
losses in the dielectric. Ever use a disc-ceramic to try
to pass one amp of RF? My experience is that it will
light up the night sky. :-)
--
73, Cecil http://www.w5dxp.com

John Smith wrote:
Of course, there must be lossless caps somewhere, huh?

There's some in EZNEC. :-)
--
73, Cecil http://www.w5dxp.com

Well, there are dielectrics and there are . . . ummm well you get the
point...

In this case the main dielectric of my condensers is glass, probably
stannous float bath soda glass... 0.100" thick with an aluminum plate
on each side... After my fingers had found the disparate temperature
rise on an early test I did a literature search for the dielectric
constant and loss factor of glass and discovered that not all glass is
equal - or as Orwell put it, some of the animals are more equal than
the others...

Anyway, Soda glass has a loss tangent of 0.01 to 0.05 and a dielectric
constant of 6 - and Borosilicate glass (Pyrex) has a loss tangent of
0.001 to 0.002 and a dielectric constant of 4... So, it would appear
that Borosilicate glass is better as a low loss dielectric... The
trade off is that with 1/3 lower Dielectric Constant I would have to
increase my plate areas by 1/3 to maintain the same capacity... The
jury is out on this... 3.5 mc is relatively low frequency... I am not
sure how much of the heating is due to the loss factor of the glass and
how much is I2R heating from the current flowing across the plates...
I spoze I could order some custom made 8" X 10" X 0.100" Pyrex plates
and compare otherwise identical condensers... OTOH, I spoze some of
the more equal animals in my house would complain over sticks and
stones in their xmas stockings after I pay for the Pyrex...

denny / k8do

Cecil Moore wrote:
Rick wrote:
What do you attribute the losses in the cap to?

Besides I^2*R losses in the leads, there are dielectric
losses in the dielectric. Ever use a disc-ceramic to try
to pass one amp of RF? My experience is that it will
light up the night sky. :-)

Cecil:

What is the best fly wing scale you have?

I would think specs would call for one capable of
0.00000000000000000000000000000000000000000000001 gm resolution ...

Well, Santas coming, hang out a big sock!

Scratching head,
JS

On 4 Dec 2006 06:07:44 -0800, "Denny" wrote:

....
So I purchased an Infrared measurement gun...

fascinating little instrument - did you know that on a clear day with
the air temperature +12.1 F and zero wind, that a 5 foot diameter
black rubber tractor tire facing the clear sky to the North can have a
surface temp of -2.6 to -3.0 F? I didn't but that is what I found...
OK, I digress again

It is not such a digression.

You should do make measurements of the temperature of different
materials that you know are at the same temperature, and see what
results you get from your non-contact thermometer.

The emissivity of the surface is an important factor that limits the
absolute accuracy of these things.

It is an interesting experiment that you have described, though I am
not sure that surface temperature (if it is accurate) alone is a good
indicator of the power flow to the air. For example, would you expect
that the temperatures of natural coloured and black aluminium heatsink
to be the same if dissipating the same (non zero) power in the same
environment? Extending that to your experiment, is dark enamelled
copper wire (as may be used in a coil) a better black body radiator
than bright aluminium (as may be used in a capacitor).

Owen
--

Owen Duffy wrote:
On 4 Dec 2006 06:07:44 -0800, "Denny" wrote:

...
So I purchased an Infrared measurement gun...

fascinating little instrument - did you know that on a clear day with
the air temperature +12.1 F and zero wind, that a 5 foot diameter
black rubber tractor tire facing the clear sky to the North can have a
surface temp of -2.6 to -3.0 F? I didn't but that is what I found...
OK, I digress again

It is not such a digression.

You should do make measurements of the temperature of different
materials that you know are at the same temperature, and see what
results you get from your non-contact thermometer.

The emissivity of the surface is an important factor that limits the
absolute accuracy of these things.

It is an interesting experiment that you have described, though I am
not sure that surface temperature (if it is accurate) alone is a good
indicator of the power flow to the air. For example, would you expect
that the temperatures of natural coloured and black aluminium heatsink
to be the same if dissipating the same (non zero) power in the same
environment? Extending that to your experiment, is dark enamelled
copper wire (as may be used in a coil) a better black body radiator
than bright aluminium (as may be used in a capacitor).

Owen
--

And of course, the magnitude of the surface area over which the power is
being dissipated will also influence the temperature rise of the small
area being measured by the instrument.

Measured temperature differences of the same material and at the same
distance (under different experimental conditions) are likely to be more
accurate than absolute measurements with unknown emissivity.

Chuck

----== Posted via Newsfeeds.Com - Unlimited-Unrestricted-Secure Usenet News==----
http://www.newsfeeds.com The #1 Newsgroup Service in the World! 120,000+ Newsgroups
----= East and West-Coast Server Farms - Total Privacy via Encryption =----