0.1 ohm Resistors from Scratch Again

I've been looking, but I can't find my CDROM with the Monty Python dead parrot routine. :(

On Tue, 20 Sep 2005 18:04:48 GMT, "Michael A. Terrell" Gave us:

He should just BUY a precision 0.1 ohm shunt resistor and put a fast-blo fuse in series with it.

Measure voltage across the shunt and calculate the exact current with a simple decimal shift.

I can always borrow a friend's parrot. His name is Pecker and he only stays on with velcro.

Best regards, Spehro Pefhany

How about enclosing in quotes "metric copper wire table" so as to get hits only on pages having the exact phrase "metric copper wire table"?

Once you try that, hits may be low - but then try alternatives, but narrower than finding the maybe .005% of all web pages on this planet that have anywhere in any order with who knows what between them the four words "metric" "copper" "wire" and "table".

Maybe also try "ohms per foot" or "ohms per meter" or ohms/foot" or "ohms/meter" as well as some combination of other keywords above that gave hits too many rather than too few. If these fail, try good keyword combinations plus ohms and also the phrase ((foot OR ft) OR meter)

Please in addition allow tolerance to be so significant that you are probably better off winding a resistor almost on a whim and then pasing through a little current and measuring separately voltage drop and current, and determine resistance to be the ratio of these two if these are known well enough to not have one chance significantly from measurement of the other. Also allow for resistance of copper to vary with temperature, close to proportionately with absolute temperature (which is about .335%-.336% per degree C deviation from 25 degrees C. The temperature coefficient if based on resistance being proportional to absolute temperature is .341% per degree C if baseline is 20 C. As far as I know copper varies even slightly worse than this, and tungsten varies even a little worse still, although nichrome is more constant to some extent).

- Don ( snipped-for-privacy@misty.com)

How many miles of wire get drawn in a day? How many days do they keep on using the same die?

- Don ( snipped-for-privacy@misty.com)

I suspect more likely it is within the manufacturer's tolerance (if stated at all) of something that the manufacturer offers. If the wire diameter (counting only copper) was .067 mm in diameter, then my guess was that this towards the end of usage of the die in question to give 42 AWG wire or being about borderline for having the resulting wire being called AWG 41.

Makes me think now: Since most wire is of even AWG sizes, this makes me think that some odd-AWG wire got that way from well-worn dies - and may be mostly both a little small in size as well as a little out-of-round. By this hypothesis, I am suspecting odd-AWG wire having more of a lack of existence larger than "center" of the AWG gauge. I do suspect that AWG is a wire gauge continuing to have significance worldwide, comparable to inch spacing of IC pins of remaining-popular long-established ICs despite current production of those being highly from metric countries.

- Don Klipstein ( snipped-for-privacy@misty.com)

Oops! :-P

selection.

\"Watt

just

table"?

that

words

gave

pasing

these

to

per

.341%

still,

Thanks, Don. I put 5 of the 1 ohm carbon films in parallel to get 0.2 ohms, but I decided that to make 0.1 ohm takes too many resistors, too much time, and too much space. So winding a few turns of fine wire on a

Howzabout some numbers:

Here are some specs from a manufacturer:

Nominal diameter Max dia Min dia x-section tol AWG 30 0.01" 0.0101" 0.0099" +/- 2% AWG 30.5 0.0095" 0.0096" 0.0094" +/- 2% AWG 42 0.0025" 0.0026" 0.0024" +/- 8%

The tolerance is a +/- a thou for all the above diameters. There is no overlap, so if they use dies for the next largest size, they'd clearly have to do some work on them first, and if the wire is outside spec, it's scrap.

I can't say as I disagree with Watson's conclusions, that the wire can't be AWG42, assuming no serious error in the measurements of length or resistance. Most of the things that could go wrong would tend to increase the measured resistance (heating above 20°C, stretching of the wire during winding). AWG 41.5 is close enough, but that's a fairly odd size, and still biased a bit the wrong way.

Best regards, Spehro Pefhany

I read in alt.binaries.schematics.electronic that "Watson A.Name - \"Watt Sun, the Dark Remover\"" wrote (in ) about '0.1 ohm Resistors from Scratch Again', on Tue, 20 Sep 2005:

Will your very thin wire carry the current without getting too hot? I think you will have constructed a fuse, not a resistor.

\"Watt

the

wire

of

I read that the newer cat5E UTP cables are made using a wire size that's slightly larger than 24 AWG, but it's not 23 AWG.

One thing that I can't seem to conceive is how does the drawer get the wire started thru the die. The end must be tapered somehow, so that it fits thru and can then be grabbed and pulled. Maybe some kind of roller that tapers a few feet down to the hole size?

Thanks for the info.

makes me

may be

By

significance

from

Once they get enough through to grab, the rest can be drawn through. It only decreases in size by a bit each time. Jewelers do this too with a "drawing bench", but copper wire uses purpose-built machines.

The wire plant I was in was in San Chung City, Taiwan (Taipei suburb). The drawing machines had quite a few dies in them, each reducing the wire by a bit, and with some annealing taking place between IIRC.

The cunning factory owner had paid an industrial spy in Japan a fair sum of money to perform espionage on the machine manufacturing process. He was then able to go into business selling the machines rather than being at the low end of the food chain churning out the wire itself. He's probably retired by now with all the machines he's sold to mainland China, wheras the wire companies in Taiwan have probably gone out of that business or moved all their production to China to take advantage of the 10:1 cheaper labor prices.

Best regards, Spehro Pefhany

It seems to me that it's just another logarythmic scale. there's already plenty of others. musical pitch with a ratio of 2^(1/12) between semmitones and of 2 between octaves. decibel with 2^0.1 between units and Richter scale which has base 10 for amplitude and base sqrt(1000) for energy the 5% resistor sequence is based on 10^(1/12)

the same can be done in any logarythmic scale, to square the quantity double the logarithm to multiply add the logarythms, to compute the square root halve it etc..

why?

it's a logarythmic scale and with logarythms the base is immaterial be it base 10, base 2, base e, or base (10^1/12) it still works the same

the same tricks still work, I guess it's a little harder to convert values becase powers of 10 differ by 12 instead of 10

so the base units correspond to 0 1 2 3 4 5 6 7 8 9 10 11

minimize

A tenth of an amp across a tenth of an ohm is a hundredth of a watt. Even at 150 or 200 mA it shouldn't make much heat. The current limited PS should be set for well under 1/4 amp (it's good for only 1/2A max). After reading some of the followups, I think I should be more concerned about heating from the current changing the resistance. A dozen mW around a 1/4W resistor might change it a few degrees. BTW I'd rather let the smoke out of the resistor than a transistor. I'll have to put a few inches of the wire across a PS and crank the current up and see when it starts to smoke and glow red. ;-)

selection.

the

it

roller

Wow. It's a dog-eat-dog world. Even among the Chinese, there's a large descrepancy between labor costs. BTW, I was watching the news last night and they said that WalMart is now starting to build stores in China. They see the country as having a potential sales larger than all of the U.S. A billion more customers. They mentioned a Chinese word of phrase that has something to do with 'who you know', the Chinese have some kind of tradition about this when buying and selling. WalMart is planning on changing that model to one that requires the supplier to reduce costs in order to sell to their company. They had better diversify their sales to other countries, because it seems that more people in the U.S. see their practices as predatory and taking unfair advantage of labor.

Indeed, they are in the mid double digits of stores now, IIRC. I've been in a couple, and a Sam's Club, in mainland China. IMO, they are aiming at attracting a higher class customer base than the big European retailer Carrefour (sp?), who are more into high volume/bargain basement, and are undercutting small retailers.

Walmart obviously have adjusted their retailing to the local tastes- the restaurant serves decent Chinese fast-ish food, you can get 4 different styles of prepared chicken feet in the supermarket section, and the bakery offers the usual Chinese buns and other baked goods. Clothing includes brand-named goods popular in Asia such as Playboy. On things like GP NiMH batteries, they carry the 2nd highest level that you can find in places like HK (eg. 2300mAh AA rather than

Sure, and growing at about 10%/year for the last 20 years, IIRC.

You are probably thinking of "guangxi"

They are highly 'predatory' in China too, screwing their suppliers to the wall there, and playing one off against the other until margins are razor-thin. It's what they do.

Best regards, Spehro Pefhany

The smallest wire size for which I have reliable data is AWG40 whose fusing current is 1.8 amps. It also appears that in these fine sizes that a step of four sizes cuts the fusing current in half -- thus AWG44 would have a fusing current of 0.9 amps, AWG48 fuses at 0.45 amps and so on. THe interpolation for your size should be linear.

Jim

Bear in mind that that table requires the wire to be in 25C air, and the contact ends must be infinite heatsinks at 25C. If the wire is wound in a solenoid, the rating drops quickly.

-Chuck

From my experience dealing with my former parent company based in Europe, AWG is not used. Although, they did their PCB layouts in inches and said that it was common practice.

When in Sweden, I asked what units their construction industry used for wood measures. Ahh, they use "tumme", or thumb. They were quite shocked that one tumme was an inch.