Upper limit on power dissipation in NTC thermistors makes it to Rev Sci Instrum

How come you weren't rude to them? They are clearly clueless.

ue

You've got to be rude to me before you get earn a rude reaction. And they are phsyicists and - by definition - clueless about electronics. They also seem to be a bit clueless about interpreting English sentences

Their reply incldes the line "Sloman also points out that extremely low power dissipation in the thermistor is required for the best possible temperature stability, and stabilities of dT < 100 =CE=BCK have be demonstrated in the literature".

What I actually pointed out was that two authors have claimed that relatively low power dissipations - 3.2uW and 16uW respectively, in different thermistors - were necessary for long term stability in their different thermistors, which isn't quite the same thing. I'm not sure that I believe them, and my comment was phrased to convey that scepticism.

-- Bill Sloman, Nijmegen

Not so. You insult people all the time, without cause.

The first time I encountered you, in January of 2000, I was discussing avalanche transistors in an unemotional and, I think, helpful way. You replied with what you declared openly to be a deliberate insult. You were, as usual, wrong about the substance of the matter because your need to insult overwhelms your ability to think.

Here you demonstrate that combination of contempt and factual error:

You are, of course, older now, grumpier and further out of touch with technology.

And

Thinfilm platinum RTDs are better longterm. At 3900 PPM/K, they have plenty of signal for this sort of performance.

John

Your perceptions are distorted by your insecure vanity. Telling someone that they are wrong isn't per se insulting, but telling you that you are wrong diplomatically enough that you wouldn't feel insulted would mean making the implication so marginal that you wouldn't notice that you had been corrected, and would go on and make the same mistake time and time again.

In fact, when your errors are corrected, you see only the insult and never the correction so you still go on and make the same mistake time and time again.

I must have been jet-lagged and grumpy. Your post *was* misleading and your reaction was so out of proportion to the "insult" - such as it was - that I wasn't minded to respond as diplomatically as I might have. I can't quite see the factual error - Zetex was British even if they did get some of the their transistors from Russia from time to time

I'm certainly grumpier, but probably no more out of touch with technology - I wasn't out of touch then, and you've yet to establish that I'm out of touch now. Wishful thinkig isn't exactly proof.

That is roughly a factor of ten less sensitivity than you get out of thermistors. As Libbrecht and Librecht's response makes cyrstal clear, if you want to get the same kind of senstivity with platinum RTDs you have to AC excite them to get rid of the DC offsets, thermocouple voltages and 1/f noise - a point I made in my 1996 paper. I've liked plationum resistors for the job for a long time now - see

Sloman, A.W. "On microdegree thermostats", Journal of Physics E: Scientific Instruments, 11, 967-968 (1978).

You do know stuff, but not as well as you'd like to think. Your revelations about using dither in A/D converters ought to have been based on an understanding of how the trick has been used in the industry for the past thirty years, but you presented them as if they were tricks that you might have invented for yourself.

As insults go, this is pretty mild, but you'll probably spend the next couple of days beating your chest and telling us that you actually do know everything, and have done so for many years.

-- Bill Sloman, Nijmegen

Hey! We have an address for Slowman. Now to hire a hit man ;-) ...Jim Thompson

I certainly did not claim to have invented either trick. I might have invented them, but I didn't.

Actually, I did independently "invent" dithered a/d conversion, for use in electric meters, but I did it in ignorance of prior art. I used a triangle wave, from a free-running r-c oscillator, and not noise. I figured that the flat probability distribution would smear ADC codes nicely.

John

He'll have to be quick - we are moving house in a few weeks.

-- Bill Sloman, Nijmegen

It isn't the best probability distribution you could have chosen. There's published literature on the subject, which John Watkinson references, amongst others.

I dug out a couple of the relevant papers for

Sloman A.W. "Comment on 'Noise averaging and measurement resolution" Review of Scientific Instruments, 70 4734 (1999).

The Gray and Stockman paper was fun, with a reference to one of The Who's more popular recordings.

-- Bill Sloman, Nijmegen

It worked fine. We sold or licensed thousands of 16-channel utility survey meters using this technique, something over $1e7 in sales.

John

I invented delta-sigma independently too, ignorantly, using a Tano Dragon (6809 uC) and a single-bit for voice i/o. Worked very well. Ignorance is bliss.

Same for dither. I first used it to measure an a/d noise problem in an instrument--when the noise got below 1 bit I fed the a/d a small ramp to nudge it into wobbling between codes, quantized and characterized the code probabilities through the transition, etc. I built on that a couple years later to increase the resolution of a phase measurement.

So, yeah, they're things you might've invented yourself for sure.

It's usually easier to figure things out than dig through the prior art. Not always, but usually.

-- Cheers, James Arthur

It's the fine details that take a while. Stockman set up Soundstream and one of his papers talks about a funny modulation on a long, slowly diminishing note in a digitally recorded performance from The Who - one of the first professional digital audio recordings.

IIRR it was dither that hadn't been done quite right ...

-- Bill Sloman, Nijmegen

It pobably could have worked better, making it more difficult for the competition to come up with something better.

-- Bill Sloman, Nijmegen

As far as I know, nobody ever has. End-use load studies were something of a fad, and after each study was over, after a year or two, the equipment was pulled and offered for resale. So eventually the whole market ground to a halt. There are still lots of C180's on the used market.

Well, it paid for my house.

John

I invented the dual-slope ADC when I was just a kid. I figured that I'd need to use a relay to do the switching, and that it would bounce and wouldn't be very accurate, so I gave up on the idea. I coulda been rich. But heck, who wants to be rich anyhow?

I also invented the successive-detection log video detector, ditto coulda been rich. In that case, Melvin Goldstein (of same TANO fame) told me it was a bad idea, so I gave up on that one, too.

John

Everybody has loads of good ideas. If you don't know what has already been invented, you will think that more of them are original than happens to be true.

This can lead you to re-invent the wheel, which is embarrassing, but more often leads you to wasting time checking out known traps for young players. A surprisngly large number of inventions are obvious to those skilled in the art - they get invented independently by different people in different places at much the same time. What is depressing is how few people are "skilled in the art".

-- Bill Sloman, Nijmegen

I think I invented both of those some time before they were patented by others.

John

hey

ve

sed

A dual-slope A/D converter using a relay as a switch is the kind of conceptual invention that wouldn't have got patented if someone else had invented it - it wouldn't have worked (because relays switch relatively slowly and not all that predictably) and would have prevented the inventor patenting the same idea when a better-behaved switch became available.

I haven't a clue what a successive detection log video detector might be. If your version didn't look as if it would work, you probably didn't invent the version that did work and proved to be worth patenting.

My father's patent on counter-current chemical pulping of wood (to make paper) wasn't for the invention of the idea of counter-current pulping, but for a scheme of doing it that worked.

The Norwegians (Kamyr) who built the continuous digester my father re- configured for counter-current cooking had been working on a counter- current process for some ten years when my father got his patent, and they were mightly peeved that he'd got there first, and with one of their digestors. Johan Richter eventually forgave him, but it took a while.

-- Bill Sloman, Nijmegen