Brian Ahern's replications of the Parkhomov experiment are interesting, because they avoid very nearly all the potential errors in other calorimetry methods.
Brian puts a reactor in an oven, heated to operating temperature, and compares the temperature inside the reactor body with that outside. Any excess heat will be seen as a temperature difference.
Providing the thermocouple's used inside and out remain accurate - and cross-calibration before and after test is possible - this is an exquisitely sensitive test of excess heat from the reactor.
It is inherently free of the following errors known to exist in other methods, even when controlled:
- TC contamination by current from heating coil (see MFMP's latest sterling work on this issue)
- Movement of TC's or heating coils
- Input power errors
- Change in total thermal resistance to ambient (many possible mechanisms here)
Should Brian's tests show significant beyond chemical excess heat they should be taken very seriously. Equally, should Brian's tests show no excess heat that should be taken seriously.
Those who have followed the science of the second Lugano test known that there is no evidence of excess heat in that test. Even given the possible errors the fact that the two different active temperature tests had identical COP shows no temperature-dependent exothermic reaction.
The Lugano reactor was given to the testers by Rossi as his "best, latest" evidence of a working device, and Rossi presumably depended on a positive result because he used the test data to support his patent application.
Therefore, based on this one test, we can say definitively that either Rossi is so incompetent he thinks non-working devices are working, or he was deliberately giving a known non-working device to the testers.
I'm not willing to speculate on reasons or motives, nor to say which of the above explanations is true. Either way, it is overly complicated to imagine the Lugano reactor was "half working" - with some LENR but not enough to make measurable excess heat. Superficially the isotopic evidence supports this. Actually such complete conversion to 62Ni is incompatible with no excess heat. To support a "nearly working but at very low COP" hypothesis you need a number of complex assumptions:
- The Ni testing showing 98% 62Ni is "lucky" with the tested material very atypical of the sample as a whole
- The "lucky" sample isotopic change is due to extraordinary nuclear reaction rather than unusual but ordinary fractionation
- In spite of having 100 reactors or more working in a long-term semi-commercial test Rossi gives his testers a non-working (or unreliably working) device
- Rossi did not realise that his device was not working when tested (he had done tests himself, read the report, was instructing the testers how to test, has stated that he considers the test positive, surely could have corrected any obvious errors)
I think an unbiassed observer - regardless of their a priori views on LENR - would find it difficult to swallow these assumptions when given the logical alternative of Rossi giving the testers a truly non-working device.
So what is the strong interest in replication? Parkhomov made one initial experiment which appeared to confirm the Lugano results. We know now that it did not, because Lugano showed no evidence of excess heat.
Subsequent Parkhomov experiments have never validated the initial claims. He could have replicated his own experiment looking for the same results again. Instead he moved to a less reliable method of calorimetry. We still have no clear data from him.
Lack of clear data is understandable. It is hard when first doing an experiment to make it bullet-proof. But when you have a relatively simple experiment that gives results which will win you a Nobel Prize you go on replicating it. What has happened once can happen again and the original calorimetry method was pretty good.
Scientists, knowing this, reckon a single claimed result is not evidence. A repeat under completely different and questionable methodology is also not evidence. And should there be some real effect here replicating it with identical methodology must be possible.
Looking at the large number of replication attempts scientists would check what are the uncertainties in each case. Where there is no real effect they would expect the positive results to correlate with the flakier methodologies with larger positives from more flaky experiments. They would expect the best experiments to show lowest effect.
That is exactly what has happened. And not surprising, given that the apparently strong evidence from the Lugano test of excess heat turns out to be based on a basic calculation error.
I think that many people following LENR just don't understand how easy it is to get experimental errors in all of these non-direct calorimetry methods with heaters. A proper device with no electrical input after ignition and variable cooling to stablise temperature would not have this problem - but might be difficult to engineer. Brian Ahern's test does not have this problem.
Dead cat's don't bounce.
