another Self looped Generator Claim

[TinselKoala]

"I did not come here to fight, but found here your post attacking me and my work which is not cool or a great way to get to know each other. Perhaps you might consider the subject matter presented rather than nitpicking "sig digs"."

Oh, yes, you did come here to fight.

I had to go back five pages to see what you were talking about, and it is off the topic of this thread. Did I attack YOU? I don't believe I did. I pointed out that the figures you cite are wrong, and I demonstrated WHY they are wrong, and I also told you what I think when someone tries to foist off WRONG data on me instead of dealing with factual issues.

You claim, apparently, that your results are accurate to all those figures. I say they aren't and they cannot be, and I've told you why, with support from REALITY that all competent experimenters and engineers respect. If you want to challenge that, the way to do it is FIRST to demonstrate that you understand sig digs, which so far you have not done, especially in your last, insulting, post, and SECOND... to demonstrate that your numbers are indeed that precise.... or change them so that they reflect the TRUTH in the degree of precision that you can manage.

If you like, I can attack YOU too, since you appear to be another classic Dunning-Kruger example and you can't deal with the actual issues I raise but would rather attack ME instead. Fine, you invited yourself, even if you can't make a part that is exactly 10.0005 inches long. Maybe you should put up your CNC machines and go back to basics for a while... at least until you can make a part that is accurate to, say, four sig digs. Can you at least do that? Make a part that is, for example, 10.05 inches long? Can you at least demonstrate that you understand the difference between MACHINING TOLERANCES, and SIGNIFICANT DIGITS ?

How many significant digits do you see in the measurement I am making here?  Do you want me to get out my micrometers, or perhaps the Mettler H-10? How about my Philips PM6676 counter which is a _calibrated_ instrument traceable to NIST standards and gives a frequency reading accurate to NINE sig digs?


No... I've got a better idea. How about YOU justify YOUR work and YOUR claims, in another thread that you start yourself, since this one is about Chuckie Pierce's failed demonstration and his continuing promises -- and failures -- to show something working.

[Temporal Visitor]

I apologize. - There you have it, I am Man enough to do so and it is not too hard to do so: when I am wrong.

Will that suffice?

FACT is we both may work "to" 0.0001" - however neither of us are able to honestly "hold" 0.0001" for that fact 1 degree of temperature change blows the best efforts.  FACT is all the rest of your insults are baseless. I will leave you as I found you: not worth the effort.

So how many digits of precision do you work with when you make a part that is 10.005 inches long? How about 10.0005? Can't you work to the ten-thousandth of an inch? Not a very good machinist then, are you. There are SIX SIGNIFICANT DIGITS in the dimension 10.0005, dear sir.

And if you don't believe that I can work to tolerances of greater than one-ten-thousandth of an inch.... you have never honed a cylinder to fit a piston.

I note that you do not understand the issue and you prefer to bloviate and insult instead of face the FACTS.

After you look up Significant Digits, and machining tolerances.... you may apologize. But I doubt if you will. You cannot refute me, though.

[TinselKoala]

I apologize. - There you have it, I am Man enough to do so and it is not too hard to do so: when I am wrong.

Will that suffice?

FACT is we both may work "to" 0.0001" - however neither of us are able to honestly "hold" 0.0001" for that fact 1 degree of temperature change blows the best efforts.  FACT is all the rest of your insults are baseless. I will leave you as I found you: not worth the effort.

Apology accepted, even if it is delivered from your left hand.

FACT is that I said you were wrong, and why, and I object to people presenting wrong data to me in an effort to dazzle me. That is not an insult, and it is the truth.
FACT is that you called me a liar when I said I could do machining work to six significant digits of precision. That is an insult, and it is not the truth.

Bye, now.

(Did you look at my LTD Stirling engine, which uses a CARBON cylinder (coefficient of thermal expansion about 0.5 x 10e-6/degree K) honed to fit an aluminum piston, with no seals, rings or added lubrication, and HELIUM as the working gas? Runs on a cup of warm water, or backwards on snow or ice? And the helium stays inside? I laugh at your "one degree of temperature change".)

[Pirate88179]

FACT is we both may work "to" 0.0001" - however neither of us are able to honestly "hold" 0.0001" for that fact 1 degree of temperature change blows the best efforts.

Totally not true at all.  Where the heck did you learn machining?  We used to do that all day long and shipped tens of thousands of parts to our customer who measured them in QC (Over 1,000 miles away) and agreed we met the tolerances required.  (Government research contract)

Your statement is dependent on the material being machined...more exactly, the material's coefficient of thermal expansion.  We were machining aluminum oxide (99.7%) which has a low expansion.  There are some materials which your statement would be true to but to make a general statement about all materials shows very little knowledge of materials which also must mean very little knowledge of machining in general.

I was in the precision ceramic machining business for over 20 years so yes, I do know what I am talking about.

Bill

PS  What TK is talking about with the digits is machine operator basic machining 101 stuff...something a machinist would have learned before becoming a machinist.

[Farmhand]

Hi TK, Just a clarity question or two. For my own benefit. I understand what you write about the "significant digits",
and it does make sense. However when I was calculating the MOI of my rotor I had a lot of input figures from previous calcs
that were quite a few digits to begin with even though I started with not very accurate measurements (in the big scheme of
things).

I'm sure you will remember as you had the same objection to my result. one of the input figures to the final calculations was
something like ( 0.000001893583- and many more digits ) For anything but the result I think it's best to keep all the digits if
possible to keep accuracy as good as can be and the "result" could be rounded back to as few digits as the person reading it
wants to. But if we talk a formula that gives us figures to work with that are very small for the scale we work in we can go
to a smaller scale. The way I see it if I said the result was 0.00000189 then I only went to three digits of accuracy in the
result because I dropped of the other 20 digits and rounded down.

Here is a question. How do we most accurately deal with many digit results during a calculation ? And if I were using, lets
say Watts is 0.00000189 really a very small amount if it's already rounded down from many more digits ? Isn't it just
another way to write 1.89 uWatts.

I don't understand how if a simple calculation gives a figure like 0.000001983583476297871  that that can be rounded back
to two decimal points and retain accuracy especially if the figure shows up in the middle of a set of calculations by continuing
with the large figures in the calcs sometimes they get small again. Rounding back the figure above to two or even four decimal places nets us a big Zero.

What is the most accurate result for this calculation if were to be carried through to further calculations.   0.0000152 kg x 0.000213 = 0.0000000032376.
remember that figure is then to continued in calculations.

Then if the "end" result ends up being something like 0.00000046328- and many more figures and I round it back to
0.000000463 the result still has more decimal places than the input figures, and I see it ad rounded back to three figures.

Most of us do not say "this is the actual accurate result" most of say "this is what the calculations give me".

The context under which the result is presented makes all the difference. In my opinion.

Often times when we convert to Farads or Henries the figure from the conversion has many decimal places to begin with
and we must convert to use the formula. How can we deal with that without numerous decimal places ?

Cheers