Measurement Errors

Measurement errors with the Kawai motor?

Last time, I said that from data given in US Patent 5,436,518 for the Kawai motor, an efficiency of 318% was apparently obtained. There is a very slight possibility that a simple measurement error could have been made, which I'll discuss below. First, let's take a more general look at measurement errors, and then at electrical power in particular.

Deliberate deception

“I’ve got five ways of making a P&G meter read any engine displacement you want. You name a number and I’ll have it come up on the gauge. Four of the ways everybody in the business knows about; the other one is my secret.”

— Trans-Am race car builder Bud Moore, quoted in Donald Farr, Mustang Boss 302, Mustang Publications Inc, Florida, 1983, p13.

In other words, it's easy enough for someone who knows what they're doing to make meters read anything at all.

Self-delusion

An inventor sets up instrumentation which he hopes will show that his machine will deliver Free Energy. But something is obviously wrong on the first trial. He goes looking for errors, finds one, corrects it, and tries again. Perhaps a few more errors will be found and corrected. And then — the inventor’s great hope is realised — the meters are now showing what he expects to see: Free Energy! So he stops looking for any more errors, when there may well be more to find.

I believe that this premature end to error-checking can even trap very experienced orthodox scientists, when they achieve an "expected" result (although they would probably deny it). As evidence I would cite the nature of the inconsistencies, as they have occurred over time, in the measurement of some of the fundamental constants, such as the speed of light. See for example the measurements at http://www.sigma-engineering.co.uk/light/lightindex.shtml. As is noted there, "The chart shows a gradual decrease in the measured values and substantial inconsistency between studies, even when measurement uncertainty is considered."

Measurement of electrical power

As amateur experimenters sometimes find out "the hard way", care is needed when measuring the true power being consumed by an electrical device. Otherwise it can appear to consume less power than is really the case. This problem can become especially acute for devices employing rapid switching, as the Kawai motor does.

However, this potential problem, and its solution, i.e. using the correct instrumentation to avoid it — e.g. a "power analyzer" appropriate for the device being tested — are well known to any competent electrical technician. (The instrumentation must correctly aggregate the instantaneous product of voltage and current, over time. In practice, "instantaneous" means "a time interval short enough to cause negligible error".)

A bad example

This issue of incorrect electrical power measurement is brought up in Mats Lewan's recent book An Impossible Invention, about Italian inventor Andrea Rossi's "E-Cat". The E-Cat seems currently to be the cold fusion device nearest to commercial availability. (Yes, I know that "LENR" — low energy nuclear reactions — is now preferred to "cold fusion").  On p238 of his book, Lewan discusses the E-Cat test done on September 6, 2012 for a group of Swedish investors, in which Rossi's simple instruments measured a power consumption of only one-half to one-third of the consumption measured by a hired researcher from the Technical Research Institute of Sweden, who used correct instrumentation. This test, like several others, ended badly:— "...when Rossi insisted that the test was successful, the investors more or less accused him of fraud and left."

The Kawai motor example

Here is the relevant section from Kawai's US Patent 5,436,518:—

"Electric power of 19.55 watts was applied to the electromagnets at 17 volts and 1.15 amperes. Under the above condition, a rotational number of 100 rpm, a torque of 60.52 kg-cm and an output of 62.16 watt were obtained."

The mechanical power output would certainly have been measured by a simple, orthodox brake test. We can easily check that 60.52 kg-cm (= 5.935 Newton-meters) multiplied by 100 rpm (= 10.472 radians/sec) gives 62.15 watts output, near enough to the stated 62.16 watts.

As for the electrical power input, I have a slight problem with the phrase "at 17 volts and 1.15 amperes".  17V × 1.15A does equal 19.55 watts, and the way this is written leaves open the possibility that simple, overall voltage and current measurements were taken, and just multiplied together, which is precisely the kind of error already discussed. 

On the other hand, all this would have been utterly obvious to Professors Otsuki and Mukaidono, if not to Kawai himself. It appears that a power analyzer was used in the tests shown in the video, and not just a simple voltmeter/ammeter combination. I find it very hard to believe that Kawai, and those who checked his work, could all have made the same measurement error.

The only way to be certain about this would be to have full details of what instrumentation was provided, and how it was connected and operated. In the past I have tried to find out more details about the Kawai motor, and possible developments of it, by writing a polite, low-key letter to one of the professors named above. But I only got what I now know is the standard "response", i.e. no reply at all.

A deliberate impression of incompetence?

However, sometimes inventors want to seem less competent than they really are. As one example, Mats Lewan seems to agree with another researcher, Jed Rothwell, that Andrea Rossi is often deliberately trying to appear fraudulent and/or incompetent. By doing so, he could expect to be "written off" by potential competitors (and the general public), and consequently could expect far fewer problems from these sources.

Perhaps Kawai was doing something similar, with the wording of his patent?

On a personal note, I can certainly see the merit in this approach, but only for possible future use. For now, I'm happy to keep as low a profile as possible, while still placing what I think should be placed on record.