Ask questions to the Working Group - ECAT long-term test

Good idea.

Specifically the input current and voltage curves over this period would allow the resistance of the heater at different temperatures to be calculated. While I know of no material that has such a dramatic fall 500->1250 and then a very slight (typical of metal) rise (1250-1400) if the resistance could be determined from the data that would be best.

While extrinsic (ie undoped) semiconductors exhibit dramatic decreasing resistance over a temp range as thermal energy knocks more carriers into the conduction band they do not fit the bill:
they have much higher resistivity than you'd need for this heater
they do not have this 1% increase until much higher temps than the region over which resistance decreases.
they have low conductivity at room temperature - so the heater could never get started.

here for resistivity of Si vs temperature. We expect a total resistance on order of 4 ohms (ball park figure, based on the comment about 40A + in the report).

Of course, if such a material existed (not Inconel) all that would be needed is for Rossi to state what he uses. I doubt that would happen - Rossi knows when absence of information serves his purposes better than giving information. So the answer has to be voltage and current, or power and current, values as the device warms up. They should have these since they have tracked power and current throughout the active tests.

Indeed. And the most likley secret he could be hiding is that it has lower COP than everyone hopes.

Very topical for this thread - ask the testers to validate their anomalous results by giving evidence of change in resistance by a factor of 3.3 in heating element during the active run.

They have the PCE-830 data stored to do this (current and power during active warm-up, from which they can easily determine resistance at temperatures 0 to 1250C).

It is necessary because the figures they publish don't make sense. If they had realised this themselves I'm sure they would have checked, and then revised their assumption that the heating element was Inconel. Maybe then they could have tested the dummy at higher temperatures!

Testing team - please answer the resistance question.

Tom

So -

Mat's Lewan's site has some additional (very clever) evidence from andrea.s that this test had wrong power measurements due to a reversed current clamp. The current waveforms from the PCE-830 are consistent only for the I1 and I3 line clamps being reversed polarity. I am not 100% convinced by this since I have not myself checked out how the PCE-830 can be triggered, but I have no reason to disbelieve andrea.s.

This would explain everything:

The higher COP (actually only 1)
The anomalous Joule power measurements (the reversed clamp condition would be true only for the active tests, giving a roughly X3 difference)
The waveforms from the PCE-830

In the face of such strong evidence the testers must surely feel it proper to resolve the matter, which they have the recorded data to do, and amend the report suitably. Does anyone know whether they actually read this stuff and therefore know about the issue? It does not take much time to check if they have the data? Just to summarise:

• The reversed clamp could be detected looking at the individual line powers from the PCE-830 during the active test.
• The wrong power measurement (from reversed clamp or maybe some other cause) could be verified by checking resistance of the heater during active test warm-up phase. If this stayed roughly constant (as expected from Inconel etc) then the power measurement is definitely inconsistent with the measured currents by a factor of roughly 3. The report would not pass competent peer review without resolving this matter, since the wires are stated to be Inconel and in any case such a strange resistance variation as is needed to make these measurements consistent is unusual and needs to be explicitly checked.
• (This is a very minor issue which does not affect the results much) The calculations for wire resistance are wrong because the ratio of C1/C2 currents is not 2 as stated, but something less than sqrt(3). The report would not pass competent peer review without correcting this error, even though it mostly cancels out and so does not much affect the results.

Tom

After this time, my question is:

When will we get answers to these questions???

Quote from David

Working group will answer questions!
The authors of the long term test-report of the hot-cat had offered us the possibility to ask them a bunch of questions. Our idea is that YOU can post your question to the authors of the report here in this thread.

At the end we will together select which questions (let's say up to 10) we will actually send to them. You can do this by simply upvote your favorite question as seen in the picture below

Update 2014-10-25
The Working Group is not able to give us the answers right away, but they promise to take in what we write.
I guess we'll see an updated report that takes insperation from this thread.

Think periods of negative temperature thermal resistance evident in furl element temperature traces in GS3 test troubling fuzzy logic controller. Could lead to burnout and cycling at higher temperatures. Responsible for 2 to 3 second glow variations seen in Alan's videos

Please, think through the issue and explain in detail.
Also, appreciate if the member can be pressent in this thread.

Display More

Hopefully the spelling "present" not "pressent" and "furl" whatever that means will be clearer as it develops. And the functionality of the links will lead to actual images or actual lists of questions. Anyway, thanks for the effort to establish this monolog / dialog with the E-cat working group. Very much appreciate it!