NetNews Usenet Archive 1992 #16/NN_1992_16.iso/spool/sci/physics/fusion/1776

Internet Message Format | 1992-07-22 | 8.1 KB

Path: sparky!uunet!pmafire!news.dell.com!swrinde!mips!pacbell.com!tandem!zorch!fusion
From: [e-mail] (Jed Rothwell)
Newsgroups: sci.physics.fusion
Subject: Miscellaneous responses
Message-ID: <[e-mail]>
Date: 22 Jul 92 22:07:12 GMT
Sender: [e-mail] (Scott Hazen Mueller)
Reply-To: Jed Rothwell <[e-mail]>
Organization: Sci.physics.fusion/Mail Gateway
Lines: 139

To: >INTERNET:[e-mail]

James White points out that some energy can be stored in heavy water.
Hmm, well, okay. This is a good reason to make a small cell, with a
limited amount of heavy water in it. If the water can hold some energy,
then perhaps it is best to compare the CF cell to a wet cell battery. The
question then becomes: has this cell continuously generated 100, or 1000
times more heat energy than a battery of the same weight could generate
in the form of electricity? If the answer is yes, then we do not need to
worry whether part of that energy came from short term chemical storage.
The output has to be *continuous*, however.

"If the calorimeter is calibrated by adding a known amount of heat to a
working cell, the increase in temperature could increase the rate of D2O2
formation. This would absorb energy and thus make it appear that a given
amount of heat causes a smaller temperature rise than is actually the
case. Thus, a balanced cell would appear to be giving excess heat."

Maybe, but I don't think so. It seems to me it would have a puzzling,
opposite effect. If you have a resister, you always calibrate before the
experiment, before electrolysis, when there is no free oxygen. So, the
pre-run calibration would show, say 2 watts per deg C. Then, during the
run, you would do a recalibration on the fly, by adding heat to a working
cell with the resister. This is an excellent idea, by the way, but it is
not particularly accurate, because the electrolysis background is noisy.
This time, however, you would add 2 watts and the temperature would only
go up, say, 0.98 C. That would not look like excess heat. It would
confuse the issue, but not in such a way as to look like excess heat.
After a while, there would be a surge of heat, which you might recognize
as being related to the recalibration, especially if you did it twice a
week for four months. I personally would write it off as part of the
background noise problem. It sounds terrible, but I am not interested in
fractions of watts. I am looking for a 10 or 20 watt excess which lasts
for weeks, which is far, far out the range of effects from things like
D2O2. I am hoping, someday, to see the kind of reaction that P&F,
McKubre, Takahashi or Mizuno saw, which totally swamps any possible
chemical effect.


Not to beat a dead horse on my end either, but Tom Droege remarks:

"That is exactly why I used the initial slope.  On the (gross) assumption
that it is a first order system, the initial slope (and final value) tell
me everything."

This assumption is incorrect. Your method of deriving the TCal from the
slope does not work, it does not come close. Mark Hugo left town
yesterday, but he would be happy to supply the correct equation if
anybody is interested. It is based upon Newton's law of cooling, with a
factor thrown in for the added heat from electrolysis and the unknown
heat from CF. It is very complicated; as I said, all of the
electrochemists I have talked to strongly recommend you work with a
system which has come into equilibrium. They say you should stay away
from transitions, and base your computations on a pre-calibrated points,
after an hour or more of "settling time."

Not to beat another dead horse, but:

"My only point, I repeat, is that the Takahashi cell is not a very good
calorimeter.  One just can't say anything about what is going on.  I know
there are calibrations at various points.  I just don't know how to be
sure that they mean anything."

My only point is that I have built a Takahashi cell and operated it for
four months, and I have data which proves that it is a very good
calorimeter. I *can* say what is going on, with assurance. My cell
measures heat reliably, week after week; it shows a nice, fat, zero
excess -- except when we dump water on the computer, or roach the flow
meter, cut a wire accidentally, or monkey with the cooling water
temperature. It has definite, known limitations, for example, with my
particular device, the "box temperature" does not measure amounts less
than 2 watts very well. However, we have compared the box temperature,
which is what Takahashi uses, against the best instrument available,
which was supplied to us by a superb electrical engineer by the name of
Tom Droege, and we have found that the box calibration agrees with flow
calibration as measured by the Droege Device *almost* *exactly*. It is
uncanny. As I said, they separate only at very low power levels, and this
remains true three weeks into the current run.

You had various doubts about the performance of the cell, primarily
because you believed the cell mixes the water at different rates during
high and low phases, (if I understand your argument correctly.) My data
shows conclusively that this is not the case. If you have any other
doubts about the design, let me know, and I will check the data to
resolve your questions. You might also consult with Takahashi, or with
Mike McKubre, who just came back from visiting Takahashi.

That is my last word on the matter, unless someone would like to see an
actual sample of data, or has any additional questions.


Todd Green asks:

"I'm not sure I understand under what conditions Takahashi measures the
excess heat. Is the 20 minute current sawtooth just used to charge the
electrode, or is he doing the XS heat measurements during these ramps? I
would have thought it very hard to do such a measurement because the
calorimeter never reaches the steady state - i.e. temp always lagging
behind the power somewhat."

He measures excess during the 6 hour "high-low" runs. With the last
experiment this was 100 watts in for 6 hours, followed by 1 watt in for 6
hours. He got some boiling during the high phase, so for this run, he
lowered this to 50 watts for 6 hours, followed by 1 watt for 6 hours.
Takahashi says it is too complicated to measure excess heat during the
sawtooth run, and in any case, he only runs the sawtooth for a week, and
the cell does not build up significant excess until a few weeks after
that.

You are quite right, "it very hard to do such a measurement because the
calorimeter never reaches the steady state..." The electrochemists I have
talked to have suggested I avoid such computations; it is much better to
wait until the cell reaches equilibrium (or gets close to equilibrium;
there are always slight fluctuations.) This is the suggestion I passed on
to Tom Droege, above.

However, I have had some success getting a zero by simply adding up all
of the "instantaneous" power in and power out measurements for a day,
with the usual fudge factors for heat leak and electrolysis. Since I have
not seen excess heat, I don't know whether this really works. I think I
should try adding some heat from our auxiliary resistance heater during a
sawtooth run. There are so many things I would like to try! It would be
nice if the DoE would try one or two of these things... These experiments
are conceptually simple; they are far easier than the computers and
telephones I tangled with for 14 years; and easier than the biology I
dabbled in 20 years ago in Japan; but alas, the experiments are time
consuming and expensive. It is just not possible to do a decent job in
your spare time.

Metallurgists I have talked to say that the sawtooth "charges up" the
cell by pushing the deuterons in, and then relaxing, and letting them out
again. Ikegami and others who have been measuring loading have recently
confirmed this. The saw tooth method has been know for quite a long time,
long before CF came on the scene. Mark Hugo cited a study of hydrogen
embrittlment of steel (I cannot find the title in my notes, alas), in
which it was found that a 20 minute period was several orders of
magnitude more effective than day long periods, or very short periods.

- Jed