The following is in regard to the Rossi 7 Oct E-cat experiment as
reported by NyTeknic here:
http://www.nyteknik.se/nyheter/energi_miljo/energi/article3284823.ece
http://www.nyteknik.se/incoming/article3284962.ece/BINARY/Test+of+E-
cat+October+6+%28pdf%29
A spread sheet of the NyTecnik data is provided here:
http://www.mtaonline.net/~hheffner/Rossi6Oct2011.pdf
Note that an extra 0.8°C was added to the delta T value so as to
avoid negative output powers at the beginning of the run. This
compensates to some degree for bad thermometer calibration and
location, buy results in a net energy of 22.56 kWh vs 16.62 kWh for
the test, and a COP of 3.229 vs 2.643.
The 22.56 kWh excess energy amounts to 81.2 MJ excess above the 36.4
MJ input. If real this is extraordinary scientifically speaking.
However, the lack of calibration and placement of the thermocouples
makes the data unreliable. The experiment was closer than ever before
to being credible. Just a few things might have made all the difference.
First, a pre-experiment run could have been made to iron out
calorimetry problems. A lower flow rate and thus larger delta T would
have improved reliability of the power out values.
Second, the lack of hand measurements of the cooling water
temperatures Tin and Tout periodically was unfortunate, especially
when large values of delta T was present. The thermometers should be
relocated down the rubber hose a short distance and insulated.
Third, a kWh meter could have been fairly cheaply purchased or
obtained and read at the same time the other electric meters were used.
Fourth, a filter to smooth any pulsed current demand from the E-cat
power supply could have been used, or an oscilloscope used to ensure
no such pulses were imposed on the input current.
Fifth, the flow meter volumes could have been manually recorded at
the same times temperature readings were recorded.
GENERAL COMMENTS
A control calibration run was not made, as evidenced by a 0.8°C
minimum error in the delta T for Tin and Tout.
No kWh meter was used to measure the total input energy. It is far
better to record E(t) frequently and then drive power P(t) by
P(t) = d E(t)/dt
than to occasionally and sporadically take power measurements and
integrate to obtain E(t).
Flow meters were used but apparently no one thought to record the
time stamped volume data. It is much more accurate, depending on
flow variations, to calculate flow f(t) from volume v(t) as:
f(t) = d V(t)/dt
than to integrate:
V(t) = integral f(t) dt
(or a similar integration to obtain energy) using occasional sporadic
short interval flow measurements. This is the value of using volume
meters. This appears to actually be a small point in this case,
however, because fortunately overall flow volume was measured, and
total volume vs sum of periodic flows does not appear to be an issue,
at least compared to the other issues.
The flow rate chosen was too large, resulting in a max delta T of
about 8°C and thus unreliable accuracy in the heat measurements.
The measurements might have been more reliable if the thermocouples
had not been placed on insulated metal parts, i.e. connected
directly, metal to metal, to the heat exchanger itself. They should
have been separated from the heat exchanger by low conductivity
material, such as a short length of rubber hose, to avoid thermal
wicking problems through the metal. The same applies to the output
temperature measurement for the E-cat. This is the same problem as
before, when the thermometer was buried in the earlier E-cats, but
compounded. This makes the temperature data highly unreliable.
From the report:
"Room temperature was between 28.7 °C and 30.3 °C."
"18:53 Tin = 24.3 °C Tout = 29.0 °C T3 = 24.8 °C T2 = 116.4 °C"
"18:57 Measured outflow of primary circuit in heat exchanger,
supposedly condensed steam, to be 328 g in 360 seconds, giving a flow
of 0.91 g/s. Temperature 23.8 °C."
"19:22 Tin = 24.2 °C Tout = 32.4 °C T3 = 25.8 °C T2 = 114.5 °C"
"Measured outflow of primary circuit in heat exchanger, supposedly
condensed steam, to be 345 g in 180 seconds, giving a flow of 1.92 g/
s. Temperature 23.2 °C."
These values indicate a significant problem with temperature
measurement. The most serious problem is the output temperature
recorded for the "condensed steam". Perhaps that was a repeated
recoding error. The "condensed steam" is measured leaving the heat
exchanger at a temperature lower than room temperature by at least 5°
C, and lower than the Tin of the exchanger by 1°C.
It is notable that when the power is turned off, for example at time
14:20, and 14:51, and 15:56, the power Pout actually rises. This may
be a confirmation that the Tout thermocouple is under the influence
of the temperature of the incoming water/steam in the primary
circuit. Water carries a larger specific heat. Cutting the power
may introduce water into output stream, as before. If the
thermocouple within the E-cat is subject to thermal wicking, the
water temperature may actually be 100°C, as before. This sudden flow
of 100°C water could then account for increased temperature from the
Tout thermocouple, which is located close to the hot water/steam
input. In any case, it is nonsensical that when power is cut that
output power quickly momentarily rises. This kind of mystery can be,
should be, unravelled using a dummy or inactive E-cat during
calorimeter calibration sessions.
If the heat exchanger were 70% efficient as estimated by some
individuals, then the "condensed steam" water temperature should have
been above Tin. Given a delta T of the cooling water of 32.4°C -
24.2°C = 8.2°C, we might expect a "condensed steam" temperature more
like 34.8°C, not 23.2°C if the coupling of the two circuits were
imperfect. The insulated condenser itself and the insulated flow
lines do not appear to be a significant source of loss of energy, and
thus low measurement efficiency. Further, the low temperature of the
"condensed steam" water upon output from the primary circuit
indicates no loss of energy in the heat exchange process due to
dumped heat in the form of "condensed steam" going down the drain.
Based on all the above, the temperature measurements lack the degree
of credibility required to make any reliable assessment of commercial
value.
Noted in report: "15:53 Power to the resistance was set to zero. A
device “producing frequencies” was switched on. Overall current 432
mA. Voltage 230 V."
The power measurement during this period may be highly flawed,
depending on the circuits involved and where the measurement was
taken. Filtering between the power measurement and E-cat is
essential, unless a fast response meter, like the Clarke-Hess is used.
Even if it is real, a COP of 3 is marginal for commercial
application. It is much more difficult to achieve self powering with
a cop of 3 vs 6. Unfortunately the temperature data is unreliable,
and the COP does not look to be anywhere near the advertised 6 or
even 3. Further, the temperature tailed off after less than 4 hours
of no power input. The device should not have been shut down there,
but re-energized. To be shown to have any commercial value the device
should be shown producing net energy for an extended period, like the
24 hours originally touted for the test. The claim was the E-cat can
run for 6 months without refueling. This test was not useful as
demonstration of commercial value.
As in the numerous prior demonstrations of the E-cats, we are left
tantalized by the indication of possible excess energy, and
disappointed that with a little extra effort the evidence might have
finally been at hand.
Best regards,
Horace Heffner
http://www.mtaonline.net/~hheffner/
