I stated here several times that in my opinion the flow meter used in the 1
year test of Rossi's 1-MW reactor was probably in error. It probably
over-estimated the flow. I said that in the data I have seen, it measured
36,000 kg of water every day, exactly. I found this number impossible to
believe. I was not free to explain the problems with the flow meter. In the
response filed by I.H. in response to the suit, there is a letter from an
I.H. expert, Murray, to Penon. This is Exhibit 5, Case 1:16-cv-21199-CMA
Document 29-5.


Read this exhibit carefully and you will know everything I learned about
the flow meter from Rossi's data, and more. Let me quote part of it:

"The turbine flow meter used for your measurements was manufactured by
Apator PoWoGaz. The model number is MWN130-80-NC.

The Apator PoWoGaz’s device label clearly states that the unit has a
minimum operational flow rate of 1.6 m3 /hour. That is a minimum of 38.4 m3
/day. Using 977.8 kg/m3 as the density of water at 70ºC, the minimum
operational mass flow rate measurable with this sensor is 37,548 kg/day.
With few exceptions, your daily valuation reports reflect a flow rate
clearly below this level. How can the measurements of the flow meter be
valid when they are consistently below the minimum operating value?

The flow meter requires that the entire pipe volume be full of liquid to
function properly, as described in the Apator PoWoGaz Operating
Instructions [section 6.6 in document I-EN-2- 003/2013, Operating
Instructions, Flange water meters DN40 - 500]. The visible iron stain
waterline marks on the static vanes indicate that the pipe was not
continuously full of liquid, as required by the manufacturer’s
specifications, but rather had a substantial portion free of liquid. See
Exhibit A. How can the measurements of the flow meter be valid when the
pipe volume was far less than full?"

Some time ago, I said that the meter data shows exactly 36,000 kg per day.
In the interview with Lewan, Rossi hinted that this is a round number, and
the actual amount might be 35,792 kg (for example). From the specifications
of this meter, you see that this is not a round number. This flow meter
measures in increments of 1,000 kg. That's the minimum reading. There are
several problems with this:

Rossi and Penon claim that it clicked over 36 times every day, or once
every 40 minutes. EXACTLY once every 40 minutes; exactly 36 times a day.
Not 35 some days, 37 others. That's ridiculous. Even if the flow was
remarkably stable, surely there would be days when the instrument clicked
over at 5 minutes to midnight (37 times that day), so the next day it would
click only 35 times.

As you see in this exhibit, Penon recorded this flow rate of 36,000 kg for
every day in the ERV, including days when the reactor was not operating.

More to the point, a flow meter operated well below its minimum recommend
range will give the wrong answer. In my experience, when instruments
measure below the minimum threshold, they tend to measure too high, tending
toward the threshold.

Also as described in the exhibit, a meter that is supposed to work with a
full pipe will give the wrong answer when the pipe is half empty.

Those are the two main problems with the meter. I believe there may be
others. Both of those problems are described in the Omega guide I
referenced earlier:


"The accuracy requirements should be separately stated at minimum, normal,
and maximum flowrates. Unless you know these requirements, your flow
meter's performance may not be acceptable over its full range. . . .

Expected minimum and maximum pressure and temperature values should be
given in addition to the normal operating values when selecting flow
meters. Whether flow can reverse, whether it does not always fill the pipe,
whether slug flow can develop (air-solids-liquid) . . .

In other words, these are known problems, described in the literature and
in the manuals for the flow meters affected by these problems.

I believe these problems caused the flow meter to measure the flow too high
by a considerable margin. I believe the COP was 1, not 50. This flow meter
problem alone is probably not enough to account for the apparent COP of 50,
but this combined with other errors inflated the COP. I have estimated the
flow meter error, but I would like to carefully examine the data released
by I.H. before publishing my estimate.

- Jed

Reply via email to