Post by thread:[OT] All minus and no plus

To try and help bring everyone gradually down from their adrenaline high, after all that sex-and-religion stuff (what a combination!), and since James has canceled it (with what I would call perfect timing), let me offer the following dry data. I ordered Olin's EasyProg kit. Which is neither here nor there, except that I did some data gathering during the process of putting it together, and I thought the results were interesting. Being an old guy, and consequently having poor vision, I not only looked though a magnifying glass to identify each resistor value, but also checked each one with a ohmmeter. This is a fairly expensive instrument, and so far as I know it is very accurate. So here is the interesting part -- every single resister, with no exceptions, came in below the nominal value. This is not any kind of problem, of course, since they were all within tolerance. But one would think (wouldn't one?) that some would be over, some would be under, and once in a great while one would be right on the money. That is what statistics promises us. Nope. Not so. Every single one comes in below. So all minus, and no plus. What does it all mean? Does it relate to our not using the metric system??? :-) John

On October 7, 2005 04:30 pm, John Nall wrote: > This is a fairly expensive instrument, and so far as I know it is very > accurate. > > So here is the interesting part -- every single resister, with no > exceptions, came in below the nominal value. This is not any kind of > problem, of course, since they were all within tolerance. But one > would think (wouldn't one?) that some would be over, some would be > under, and once in a great while one would be right on the money. > That is what statistics promises us. Nope. Not so. Every single > one comes in below. So all minus, and no plus. What does it all > mean? Does it relate to our not using the metric system??? :-) Almost makes you want to verify the batteries are okay. ;-) You did check that? Have you shorted the leads together to verify it reads zero and not some negative value? Otherwise, let's just consider the conspiracy factor by putting on our tinfoil hats to ponder the possibilities, hehehe ;-) cheers ;-)

Jinx wrote: >>Otherwise, let's just consider the conspiracy factor by putting >>on our tinfoil hats to ponder the possibilities, hehehe ;-) >> >> > >Too bloody right > > OK, armed with the collective wisdom of this group (or at least that rag-tag portion of it that reads [OT], anyway) I did further checking. (a) Batteries -- this is A/C powered. (b) does it zero when shorted -- yep. (c) so let us compare other units -- other (cheaper) units give the same results. Since all three are pretty consistent, I think that we can accept the data as true. I don't know about everyone else, but to me it definitely signifies a lowering of standards. Where it will all end, I do not know. :-( John

Ah, and thus here comes the tinfoil hat topic again.... =) -Mario -----Original Message----- From: piclist-bouncesKILLspammit.edu [.....piclist-bouncesKILLspam.....mit.edu] On Behalf Of Jose Da Silva Sent: Friday, October 07, 2005 7:53 PM To: Microcontroller discussion list - Public. Subject: Re: [OT] All minus and no plus On October 7, 2005 04:30 pm, John Nall wrote: {Quote hidden}> This is a fairly expensive instrument, and so far as I know it is very > accurate. > > So here is the interesting part -- every single resister, with no > exceptions, came in below the nominal value. This is not any kind of > problem, of course, since they were all within tolerance. But one > would think (wouldn't one?) that some would be over, some would be > under, and once in a great while one would be right on the money. That > is what statistics promises us. Nope. Not so. Every single one > comes in below. So all minus, and no plus. What does it all mean? > Does it relate to our not using the metric system??? :-) Almost makes you want to verify the batteries are okay. ;-) You did check that? Have you shorted the leads together to verify it reads zero and not some negative value? Otherwise, let's just consider the conspiracy factor by putting on our tinfoil hats to ponder the possibilities, hehehe ;-) cheers ;-)

As long as they are in specification, what is the problem? Are you expecting too high from the reisistor vendors who need to squeeze every US$0.001 or even US$0.0001 they can save? I know the price of resistors is defined in US$0.0001 increament in our ERP systems. And a lot of the components are pre-selected to different grades. Safety related Zener diode is one of them. 1% zener is selected from 5% zeners. Not so sure about 0.1% resistors. They can be quite expensive so I guess pre-selection is also possible. Regards, Xiaofan On 10/8/05, John Nall <EraseMEjwnallspam_OUTTakeThisOuTgmail.com> wrote: {Quote hidden}> Jinx wrote: > > >>Otherwise, let's just consider the conspiracy factor by putting > >>on our tinfoil hats to ponder the possibilities, hehehe ;-) > >> > >Too bloody right > > > OK, armed with the collective wisdom of this group (or at least that > rag-tag portion of it that reads [OT], anyway) I did further checking. > (a) Batteries -- this is A/C powered. (b) does it zero when shorted > -- yep. (c) so let us compare other units -- other (cheaper) units give > the same results. Since all three are pretty consistent, I think that > we can accept the data as true. I don't know about everyone else, but > to me it definitely signifies a lowering of standards. Where it will > all end, I do not know. :-( > > John

On October 7, 2005 06:02 pm, Mario Mendes Jr. wrote: > Ah, and thus here comes the tinfoil hat topic again.... =) You're just viewing this from a glass half empty perspective ;-) Think about it, suppose all the manufacturers start shorting the material required to create a proper resistance. While those manufacturers make an incrementally small extra sum of money due to less material getting used, the rest of us run equipment hotter than usual, because less resistance is more wattage. More heat, and you know about global warming.... You should be thinking, glass half full and be glad we have concerned citizens looking out for society. ;-) > -Mario > > {Original Message removed}

Andy Armstrong wrote: > > > Everyone knows not to hold the probes on to the resistors with > their hands when they're measuring, right? :) I have always heard that. Since the checking was only to see if I got the colors right, then a ballpark figure was fine, and it is a heck of a lot easier to just hold the probes on the resistors with the hands. But I do plead guilty to having violated that rule. So I went back and did some checking. The results are the same, although I suspect this might vary from person to person. I only checked one resistor, which was nominally 1K. It tested at 994 ohms both with fingers and without fingers. John

William Chops Westfield wrote: > > >Aren't resistors within a particular "batch" expected to come out > very close to each other these days? The tolerance value is more to > handle differences between batches, extremes in temperature/etc, and > so on. I'd be very surprised if ALL the resistors on a particular > reel weren't much closer to one another than they were to the nominal > resistance. So if it's a "low" batch, they're probably ALL low. Yes, I would certainly agree with that. But when you check a lot of resistors, both of different values and of different tolerance ranges (some 5% and some 1%), which I did, then they are presumably from different batches. I raised the issue pretty much tongue-in-cheek, as I think the OP makes clear. But I was thinking about it some last night, and it really does appear that the mean is lower than the nominal value. From what I remember of statistics, a random sample of greater than about 20 should result in values from both sides of the mean. If we assume that the mean should be the nominal value (6.8K, 100K, whatever) then values on either side of that should appear. None do. All of the values -- without exception -- are on the low side. Therefore, for whatever reason (lower costs to the manufacturer, or designed so that as the resistor ages it comes closer to spec, or a conspiracy, or whatever) one has to conclude that the true mean is not the nominal value. It is something lower than the nominal value. John

At 06:55 PM 10/7/2005 -0700, you wrote: >On Oct 7, 2005, at 5:35 PM, John Nall wrote: > >>I don't know about everyone else, but to me it definitely signifies >>a lowering of standards. > >Aren't resistors within a particular "batch" expected to come out >very close to each other these days? The tolerance value is more to >handle differences between batches, extremes in temperature/etc, and >so on. I'd be very surprised if ALL the resistors on a particular >reel weren't much closer to one another than they were to the nominal >resistance. Have you ever checked out that assumption? > So if it's a "low" batch, they're probably ALL low. > >BillW Modern resistors are all individually trimmed to value (yes, for 0.1 cent), so the differences won't be improved by as much as you might think. The measuring equipment (including calibration) should be much more accurate than the resistor. Any correlation would probably be due to systemic mis-trimming of the resistors due to the trimming process. Checking the resistors in an SMT network doesn't yield any especially attractive improvement in typical value. On a 5% film resistor line which I did an evaluation on, they set the nominal value on the trimming machines and the cutter shut off at that value. Naturally there was a tiny lag, so the resistors tended to be a tad high in value, but all were within tolerance. This trimming was done before the coating process, of course, which would probably have shifted the value a bit. If they're within about 1% of nominal, I'd say they are essentially perfect 5% resistors, regardless of which side of the imaginary line they happen to fall on. I *would* expect the TCR to be very similar (they're made from the same batch of raw material), so changes should track once you account for the initial tolerance. In your case, you get a -0.6% tolerance, probably on a 5% part. That's more than acceptable and really nothing you should be complaining about. If it's a 5% resistor it can be as low as 950 ohms and still meet specifications-- and you measure 994 ohms... Reminds me of a great management anecdote- the moral of the story being that as a manager you should take care as to how you reward people. The case study was of a steel rolling mill-- the operators were paid a bonus based on production. Of course, scrap didn't count, and it was measured in tons. It turned out that some operators were able to control the thickness of rolled material better than others, and the best were able to run material that was just barely under the maximum permitted thickness, thus maximizing the output of the rolling mill in tons/hour (and their paycheck). As the customer was actually paying so much per ton, it fit the mathematical model of how the operator should be rewarded, but the customers were not overly pleased with material that was on the edge of being too thick. ;-) Best regards, Spehro Pefhany --"it's the network..." "The Journey is the reward" speffspam_OUTinterlog.com Info for manufacturers: http://www.trexon.com Embedded software/hardware/analog Info for designers: http://www.speff.com ->> Inexpensive test equipment & parts http://search.ebay.com/_W0QQsassZspeff

John Nall wrote: > So here is the interesting part -- every single resister, with no > exceptions, came in below the nominal value. This is not any kind of > problem, of course, since they were all within tolerance. But one > would think (wouldn't one?) that some would be over, some would be > under, and once in a great while one would be right on the money. > That is what statistics promises us. Not quite. That's what statistics promises for a large collection of *independent* events. A bunch of resistors bought at the same time, possibly from the same production run are not independent. Of course different values and different types make them likely to be less dependent. Overall, I would find it surprising if all resistors in the kit were truly low. So much so that I suspect a different explaination. To me the most obvious explaination is your ohmmeter. I know you said it was a good instrument, but what is it's published tolerance? How long has it been since it was last calibrated? If you have a resistance standard availble, try measuring that. ***************************************************************** Embed Inc, embedded system specialists in Littleton Massachusetts (978) 742-9014, http://www.embedinc.com

Olin Lathrop wrote: > >> To me the most obvious explaination is your ohmmeter. I know you said it >was a good instrument, but what is it's published tolerance? How long has >it been since it was last calibrated? If you have a resistance standard >availble, try measuring that. > > You think that is a more obvious explanation than a global conspiracy, huh? :-) Well, you may be right. The published tolerance is plus or minus 0.2%, but they also recommend calibration once a year. I think that I bought it a bit more than a year ago, and have not calibrated it. Somewhere around here I have a high-precision resistor, so I'll check that. John

At 12:19 PM 10/8/2005 -0500, you wrote: >You think that is a more obvious explanation than a global conspiracy, >huh? :-) Well, you may be right. The published tolerance is plus or >minus 0.2%, but they also recommend calibration once a year. I think that >I bought it a bit more than a year ago, and have not calibrated >it. Somewhere around here I have a high-precision resistor, so I'll check >that. > >John Make sure you check it on the same range as your measurements. Best regards, Spehro Pefhany --"it's the network..." "The Journey is the reward" @spam@speffKILLspaminterlog.com Info for manufacturers: http://www.trexon.com Embedded software/hardware/analog Info for designers: http://www.speff.com ->> Inexpensive test equipment & parts http://search.ebay.com/_W0QQsassZspeff

On Sat, 8 Oct 2005, Jinx wrote: >> Since all three are pretty consistent, I think that we can >> accept the data as true > > Do resistor values trend upward with ageing / useage? Maybe > the manufacturers are smarter than you think. I don't measure > them too often but have a feeling my measurements were also > low (just checked a few lying around here and they all are). As > you say though, always in spec. Even a 5% carbon cheapie is > close to 1% ime carbon goes noisy, composition goes every which way plus noisy, and metal film goes up. Glass packed wirewound and metal film stay put for 60+ years when not abused. It depends on what you mean by 'old' and what happened to it meanwhile. Peter

On 8 Oct 2005, at 15:14, John Nall wrote: > I have always heard that. Since the checking was only to see if I > got the colors right, then a ballpark figure was fine, and it is a > heck of a lot easier to just hold the probes on the resistors with > the hands. But I do plead guilty to having violated that rule. So > I went back and did some checking. The results are the same, > although I suspect this might vary from person to person. I only > checked one resistor, which was nominally 1K. It tested at 994 > ohms both with fingers and without fingers. What figure do you get just holding the probes with your fingers? If my fingers are slightly damp I can get down to around 10M so it'd only matter if I was measuring some fairly large values - but it's always worth remembering you're putting yourself in parallel with whatever you're measuring if you hold the probes like that - just in case it makes a difference. -- Andy Armstrong, hexten.net

Peter van Hoof wrote: > My best guess would be that with increasing temperature the resistance > would go up, to be in the safe range at room temperature the resistance > would have to be on the low side. Just a guess though. I'm not sure why nobody seems to pick up on this... John, did you check what the spec says about the temp coeff of your resistors? Many are spec'ed with +- ranges, but some are spec'ed with 0...+ ranges, for example 0...350 ppm/K (pardon me the use of a SI unit here :) 350 ppm/K would be 3.5% over 100 K -- positive. In that case, it would make sense that they are a bit below center at room temperature; at least if we assume that the target temperature range is mostly above room temperature. Gerhard

Gerhard Fiedler wrote: >> John, did you check what the spec says about the temp coeff of your >resistors? Many are spec'ed with +- ranges, but some are spec'ed with 0...+ >ranges, for example 0...350 ppm/K (pardon me the use of a SI unit here :) > >350 ppm/K would be 3.5% over 100 K -- positive. In that case, it would make >sense that they are a bit below center at room temperature; at least if we >assume that the target temperature range is mostly above room temperature. > > No, I did not check that. Makes sense, though. I feel sure there is a rational explanation. Although I am going to check the calibration of my ohmmeter, I really do not think that it is out of spec because all three of them give the same readings. Which does not mean that they are not all reading low, of course, but it decreases the probability. At any rate, I do not want to make a federal case out of it -- just wanted to share what I thought was an interesting observation. (I try to share such interesting observations with my wife, but she has developed this annoying habit of using the remote control to turn up the volume of the TV whenever I try and share such things with her. :-( John

On October 8, 2005 12:20 pm, John Nall wrote: {Quote hidden}> Gerhard Fiedler wrote: > >> John, did you check what the spec says about the temp coeff of > >> your > > > >resistors? Many are spec'ed with +- ranges, but some are spec'ed > > with 0...+ ranges, for example 0...350 ppm/K (pardon me the use of > > a SI unit here :) > > > >350 ppm/K would be 3.5% over 100 K -- positive. In that case, it > > would make sense that they are a bit below center at room > > temperature; at least if we assume that the target temperature > > range is mostly above room temperature. > > No, I did not check that. Makes sense, though. I feel sure there is > a rational explanation. Although I am going to check the calibration > of my ohmmeter, I really do not think that it is out of spec because > all three of them give the same readings. Which does not mean that > they are not all reading low, of course, but it decreases the > probability. At any rate, I do not want to make a federal case out > of it -- just wanted to share what I thought was an interesting > observation. (I try to share such interesting observations with my > wife, but she has developed this annoying habit of using the remote > control to turn up the volume of the TV whenever I try and share such > things with her. :-( FYI, with some wrist watches, the crystal is calibrated to run about the right speed while the watch is on your arm, or about 27degC versus running at room temp. You may have hit the odd statistic where all your resistors are indeed low, but, like Gerhard is describing, you may want to figure out what it would read if you got running hot, let's say it's a 1/4 watt resistor, so maybe run try it at 1/8watt and try reading the temp afterwards... 1/8watt is V=sqrt(wr) = sqrt(1/8watt x 1000ohm) = 11v So, if it's a 1/4 resistor, see what happens after you run it at 12v for a moment. cheers! > John

>So here is the interesting part -- every single resister, >with no exceptions, came in below the nominal value. hah, been there, done the investigation ... Some years back as part of assembling a series of instruments that are now in a highly elliptical orbit on the Cluster II spacecraft, I ran into this phenomenon. I cannot recall what started the investigation, but I ended up figuring we had a whole packet of resistors that had been marked wrong - but it turned out I was incorrect. The circuit used a highish value (around 820k IIRC) carbon resistor (a carbon one rather than metal oxide, was required for various reasons) and I believed we had 680k ones as that is about the value I ended up measuring them as. They were old stock, that had been used before, and we had no reason to believe they were faulty, but IIRC we had some strange anomaly when testing a completed unit that resulted in the test of the components. Someone else on the project found some data that suggested that moisture affected the observed value of carbon resistors, and so I set about doing a test with an oven, heating them to around 40C IIRC. I had been through the whole box of ~100, and found they were all about the same value, so I selected 10, and put them in the oven. Initially I checked the resistance of each sample every hour for the first day, and left them in the oven overnight, and then checked their values again the following morning. The observed resistance of each sample could be seen to move from the initial value towards the marked value during each measurement, with the observed resistance being well within tolerance at the end of the 24 hour bakeout. If the components you have are carbon, then this may be what you are observing. In my case the value was that far out that I was suggesting that the previous run through the machine had been the next standard value down, and the marking paint had been changed before the machine was purged of the previous run.

>>I'd be very surprised if ALL the resistors on a >>particular reel weren't much closer to one another >>than they were to the nominal resistance. > >Have you ever checked out that assumption? > >> So if it's a "low" batch, they're probably ALL low. >> For resistors of >=1% tolerance, yes. Tighter tolerance I haven't checked, but that is what I have observed in the past for leaded resistors. SMD ones never had occasion to check. Ended up doing this once to ensure that we could get good matched pairs of resistors as we wanted 1:1 ratio networks. All the resistors on a tape were matched within very close tolerance, although they were 2% tolerance resistors, and were within that 2%, but not close to nominal. In this case they were metal oxide, and we couldn't get closer tolerance ones readily.

At 12:08 PM 10/11/2005 -0400, you wrote: >I don't know if its true because I have not tested it, but I once heard >/ read that when resistors are made, if they are within 5%, the get >labeled 5%, if they are more than 5% but less than 10%, they are labeled >10% and so on. > >Meaning that a 10% resistor is at least 5% off and as much as 10%. > >Can anyone confirm or deny that? Categorically denied in the case of resistor processes I've seen (carbon film and metal film). It may have been true in the bad old days of composition resistors. Best regards, Spehro Pefhany --"it's the network..." "The Journey is the reward" RemoveMEspeffTakeThisOuTinterlog.com Info for manufacturers: http://www.trexon.com Embedded software/hardware/analog Info for designers: http://www.speff.com ->> Inexpensive test equipment & parts http://search.ebay.com/_W0QQsassZspeff

Yes, my impression is it was an "old days" technique. These days 5% resistors are generally within 1% and 1%s within 0.3% or so. There is also usually a difference in temperature coeficient (and power rating) between them. RP On 12/10/05, Spehro Pefhany <spamBeGonespeffspamBeGoneinterlog.com> wrote: {Quote hidden}> At 12:08 PM 10/11/2005 -0400, you wrote: > > >I don't know if its true because I have not tested it, but I once heard > >/ read that when resistors are made, if they are within 5%, the get > >labeled 5%, if they are more than 5% but less than 10%, they are labeled > >10% and so on. > > > >Meaning that a 10% resistor is at least 5% off and as much as 10%. > > > >Can anyone confirm or deny that? > > Categorically denied in the case of resistor processes I've seen > (carbon film and metal film). > > It may have been true in the bad old days of composition resistors. > > Best regards, > > Spehro Pefhany --"it's the network..." "The Journey is the reward" > TakeThisOuTspeffEraseMEspam_OUTinterlog.com Info for manufacturers: http://www.trexon.com > Embedded software/hardware/analog Info for designers: http://www.speff.com > ->> Inexpensive test equipment & parts http://search.ebay.com/_W0QQsassZspeff > > > -

Hi John. New world order conspiracy perhaps ? :) Thinking seriously I guess this is China's attempt to reduce total amount of different resistors manufactured. In simple words if you are on 5% tolerance and technology is pretty cheap to manufacture 1% resistors ( not much cost involved that 5% technology ) you rather ( that is China's point of view :) will produce half amount of resistors with value in between of two 5% values with tolerance 1% rather than produce two different values ( twice amount ) with 5% precision. Nobody uses thick film 5% resistors for anything precise anyway. WBR Dmitry. John Nall wrote: {Quote hidden}> > To try and help bring everyone gradually down from their adrenaline > high, after all that sex-and-religion stuff (what a combination!), and > since James has canceled it (with what I would call perfect timing), let > me offer the following dry data. > > I ordered Olin's EasyProg kit. Which is neither here nor there, except > that I did some data gathering during the process of putting it > together, and I thought the results were interesting. Being an old > guy, and consequently having poor vision, I not only looked though a > magnifying glass to identify each resistor value, but also checked each > one with a ohmmeter. This is a fairly expensive instrument, and so far > as I know it is very accurate. > > So here is the interesting part -- every single resister, with no > exceptions, came in below the nominal value. This is not any kind of > problem, of course, since they were all within tolerance. But one would > think (wouldn't one?) that some would be over, some would be under, and > once in a great while one would be right on the money. That is what > statistics promises us. Nope. Not so. Every single one comes in > below. So all minus, and no plus. What does it all mean? Does it > relate to our not using the metric system??? :-) > > John

Hi John. I suggest to measure the reference ( calibration resistor ) with 0.1% precision at least and look at the result. All the guesswork about how precise meter is will be eliminated. WBR Dmitry. John Nall wrote: {Quote hidden}> > William Chops Westfield wrote: > > > > > >Aren't resistors within a particular "batch" expected to come out > > very close to each other these days? The tolerance value is more to > > handle differences between batches, extremes in temperature/etc, and > > so on. I'd be very surprised if ALL the resistors on a particular > > reel weren't much closer to one another than they were to the nominal > > resistance. So if it's a "low" batch, they're probably ALL low. > > Yes, I would certainly agree with that. But when you check a lot of > resistors, both of different values and of different tolerance ranges > (some 5% and some 1%), which I did, then they are presumably from > different batches. > > I raised the issue pretty much tongue-in-cheek, as I think the OP makes > clear. But I was thinking about it some last night, and it really does > appear that the mean is lower than the nominal value. From what I > remember of statistics, a random sample of greater than about 20 should > result in values from both sides of the mean. If we assume that the > mean should be the nominal value (6.8K, 100K, whatever) then values on > either side of that should appear. None do. All of the values -- > without exception -- are on the low side. > > Therefore, for whatever reason (lower costs to the manufacturer, or > designed so that as the resistor ages it comes closer to spec, or a > conspiracy, or whatever) one has to conclude that the true mean is not > the nominal value. It is something lower than the nominal value. > > John

> Hi Jinx. > > Wearing gold and silver metal ring on the head > wouldn't it protect in the better way ? :) > ( looking towards all those crowned folks :) Yes, an aluminium hat does seem a bit....well, plebian. I was thinking I could get one like Archduke Franz Ferdinand had Hmmm, maybe not OK, what about one like King Birendra's............. Oh dear, perhaps something a bit plainer then. Say, a simple felt hat like Abe....... Looks like any sort of hat on a famous head just leads to trouble. I'll stick with no hat, just like JF... (sigh) or RF.... (sigh) Maybe aluminium isn't so bad after all. If people think you're a looney they'll leave you alone (great for getting a seat to yourself on the bus too)

> It is always good to keep the sense of humor :) > Microwave weapon ( skin heating and the rest of yada yada ) > was tested and comes shortly even to police I believe. I have a dangerous microwave weapon in the kitchen. My ancient oven rusted through somewhere (can't find where) and produced pleasant heat sensations up to 1 meter away from it in b*** height (it radiates from below but there is no hole ?!). A quickly soldered microwave tester (1N23+6cm rigid copper wires+d'Arsonval meter) confirmed the diagnosis (by nearly wrapping the d'Arsonval's needle around the end stop). I would still like to know how 12 cm wavelength microwaves get out through holes much smaller than that. Peter

On October 12, 2005 04:13 pm, Dmitriy Kiryashov wrote: > It takes _more_ material to produce _lower_ value resistor I believe. > Perhaps everything in total is averaged anyway. If the distances are the same, yes you would need more material, but if the distances are shorter, then you use less material.... so instead of building a 1206 smt resistor, suppose it's a 1106 smt resistor. Shorter distance & less resistance. Another possibility is that the paste used costs more for the resistive material versus the conductive material, so maybe more conductive material vs resistive material got used. hehehe. Anyhow, joking aside, another person indicated that it may be moisture, and that baking-out the moisture resolved their problem, so that may be a good possibility. Another person suggested temperature, and I know that some watch makers calibrate their crystals for body temp instead of room temp, so it may be a possibility that the resistors may indeed fall better in range when warmed up (warm-n-toasty, used in a valid circuit). ... and someone else suggested that it may just be the meter, and it may be a good idea to check the meter against a 0.1% resistor so that you know if it is the meter or the resistor.... last, but not least, the person originating the thread may have run into the statistical probability that he did get a batch of resistors that are "all" on the minus side of rated value. ;-) > Jose Da Silva wrote: > > Think about it, suppose all the manufacturers start shorting the > > material required to create a proper resistance. While those > > manufacturers make an incrementally small extra sum of money due to > > less material getting used, the rest of us run equipment hotter > > than usual, because less resistance is more wattage. More heat, and > > you know about global warming....

Hi Jose. Concluding. Moisture and aging are primary reasons for resistor value change. ( conditions components were kept in storage ) Temperature is the second I believe. ( it changes with it ) Human body resistance is the third :) ( finger-body-finger ) Meter battery, zero-in, calibration and precision are the last. WBR Dmitry. > Anyhow, joking aside, another person indicated that it may be moisture, > and that baking-out the moisture resolved their problem, so that may be > a good possibility. > Another person suggested temperature, and I know that some watch makers > calibrate their crystals for body temp instead of room temp, so it may > be a possibility that the resistors may indeed fall better in range > when warmed up (warm-n-toasty, used in a valid circuit). To good to be true :) I don't think anybody design components pre-adjusted to get to normal values after one run of solder wave machine or reflow oven :) Too much complications it implies.