Precise Resistor Values

On a sunny day (Sun, 1 Sep 2019 09:09:27 -0000 (UTC)) it happened Cursitor Doom wrote in :

Not a 'high end' designer here, but I guess (wild) somebody did something like:

Volts 5 LED drop 1.65 Current required 10 mA

So resistor is (5 - 1.65) / 10e-3 = 335.000000 Ohm (Linux wcalc) then put 335 .001% in the circuit diagram Spice freaks.

No sense of reality, Where the world goes, F35 stealth, LOL, any passive RF radar can see it.

We, the neural nets, REALLY need to build and test circuits and get a feeling for real values, else it all is lost.

AI will defeat us and robots will rule.. oops those will make the same mistakes. one AI will invent math and an other one spice, and 1224.5 Ohm .001 % will be required.

Take a hot air gun, heat up your circuit, so much for monte carlo and put it in the fridge,.,, Drop if from some height, , high volts, low volts.

Nice theory, Jan, but it's not correct in this instance at any rate. The actual physical component has markings backing up the precise value and it's stated as such on the parts inventory too.

LOL! I like your thinking! :-D

--
This message may be freely reproduced without limit or charge only via  
the Usenet protocol. Reproduction in whole or part through other  
protocols, whether for profit or not, is conditional upon a charge of  
GBP10.00 per reproduction. Publication in this manner via non-Usenet  
protocols constitutes acceptance of this condition.

Another possibility is that somebody once designed a product that used

1250 ohms. The resistor itself was probably marked 1251. Somebody else copied the design into another product and, looking at the resistor markings decided that it was a 1251 ohm resistor. Millions were ordered and this became a company standard part. All future designs that needed about 1250 ohms then used this part because there would be too much administrative hassle in specifying a 1250 ohm resistor instead.

I have encountered a similar situation where I needed to design a 600 ohm resistor into a product. The nearest value in the company database with the right power rating was 601 ohms. Nobody could explain why this value had been chosen. I think the tolerance was +/-5%. The resistor marking was 601, so I am sure it was a mistake.

Large companies will put each resistor value needed for a product out to tender and the manufacturers will happily quote for any value if the quantity is large enough. Once this has been done, all future products that need a similar value will be forced to use what is already there unless there is a compelling reason to change. It can take weeks to get a new part into the "system" and there is significant cost involved. The "large company" I refer to is one that everyone here will have heard of.

John

HP perchance?

-- This message may be freely reproduced without limit or charge only via the Usenet protocol. Reproduction in whole or part through other protocols, whether for profit or not, is conditional upon a charge of GBP10.00 per reproduction. Publication in this manner via non-Usenet protocols constitutes acceptance of this condition.

Think of blue bat wings.

Motorola? Why not just say the name?

-- This message may be freely reproduced without limit or charge only via the Usenet protocol. Reproduction in whole or part through other protocols, whether for profit or not, is conditional upon a charge of GBP10.00 per reproduction. Publication in this manner via non-Usenet protocols constitutes acceptance of this condition.

You had no trouble working it out, but it is less obvious to search engines.

Jan Panteltje wrote in news:qkg2tk$bud$ snipped-for-privacy@dont-email.me:

Not if it destroys the entire radar site 150 miles before it even reaches the radar site's view horizon base.

Oh and that 'passive radar' 'picture' appears as if to be a small bird. Usually ignored by the passive radar's operator or computer analyzers. Way past "too late" time and also guess what... ignored by radar guided air-to-air missiles.

The detection schema change requisite to track them would have missiles flying off after non existent noise signatures. Way too loose to call guided any more.

Jan Panteltje wrote in news:qkg2tk$bud$ snipped-for-privacy@dont-email.me:

Then you can toss it off the back dock and if it still works you can then get your CSA cert.

In these days of laser trimming, 1% is free. We buy all 1% resistors, except when we need something better. I think the best we now buy is

0.05% 10PPM, which costs 20 cents in modest quantities. 1% surfmount resistors cost under a penny by the reel. 1251 ohms is an oddball value. "1251" could mean 1250 ohms, also oddball.

--

John Larkin   Highland Technology, Inc   trk 

jlarkin att highlandtechnology dott com 
http://www.highlandtechnology.com

Yeah, but this is mid eighties we're talking here.

There are *plenty* of oddball values I've since discovered, having gone to the trouble of seeking them out on the schematics:

316 19.6k 61.9 56.2 23.7 17.8k 3.83k

To name but a few of very many such examples!

In fact these weird values seem to outnumber the regular ones by some measure. Weird!

--
This message may be freely reproduced without limit or charge only via  
the Usenet protocol. Reproduction in whole or part through other  
protocols, whether for profit or not, is conditional upon a charge of  
GBP10.00 per reproduction. Publication in this manner via non-Usenet  
protocols constitutes acceptance of this condition.

The Susumu thinfilms seem to come in oddball values.

--

John Larkin   Highland Technology, Inc   trk 

jlarkin att highlandtechnology dott com 
http://www.highlandtechnology.com

Those are all standard E96 1% values.

Michael, thanks for the link! I had always wondered where those values came from - that will help with our resistor and capacitor stock.

Wikipedia is most useful when it is accurate!

John :-#)#

Right, very common, if they were E192, not quite as common.

--
 Thanks, 
    - Win

Win, I haven't looked particularly for this in your book but do you have the equations for calculating 1% 5% and other standard resistor values in there ?

I remember the old basic program to find the nearest value standard resistor value from way back when.

No, sorry, we don't, but it's readily available.

--
 Thanks, 
    - Win

Google for E192 "E series of preferred numbers"

--
Uwe Bonnes                bon@elektron.ikp.physik.tu-darmstadt.de 

Institut fuer Kernphysik  Schlossgartenstrasse 9  64289 Darmstadt 
--------- Tel. 06151 1623569 ------- Fax. 06151 1623305 ---------

It is nothing more than rounded log-spaced (geometrically spaced) numbers o ver a decade. For m-code, E96 is:

po = 96 n=0; p = (round(10.^(2:(1/po):(3-1/po))))'; p = (10.^n).*p(:,ones(size(n,2),1))

yielding: p = 100 102 105 107 110 113 : : 887 909 931 953 976

for E192 (po=192) it yields:

p = 100 101 102 104 105 : : 931 942 953 965 976 988

The 5% class (E24) and 10% class (E12) are tricky because there are wrongly rounded values, and you have to fix them with special treatment. Here is t he compare of "correct rounding" versus what we actually have:

10 10 11 11 12 12 13 13 15 15 16 16 18 18 20 20 22 22 24 24 26 27 29 30 32 33 35 36 38 39 42 43 46 47 51 51 56 56 62 62 68 68 75 75 83 82 91 91

It is interesting that the "bad rounding" of E12/24 works to our favor in c ertain resistor ratio problems. We can actually get some ratio combinations with the distorted E24 family that we can't with the E96 family. I mean th ere is less error in the ratio for some desired ratios. This is because the "always correct" rounding (to discrete values) tends to cluster the availa ble ratios at "ratio-spectral lines" and leaves gaps elsewhere.