Bad resistors
in [Design]
|
From: DrParnassus on 23 Jun 2010 11:24 On Wed, 23 Jun 2010 19:28:15 +1000, Sylvia Else <sylvia(a)not.here.invalid> wrote: >On 23/06/2010 9:54 AM, DrParnassus wrote: >> On Tue, 22 Jun 2010 21:51:47 GMT, Jan Panteltje >> <pNaonStpealmtje(a)yahoo.com> wrote: >> >>> Got 10 resistors on tape marked 2M7. >>> Used one, circuit did not work right. >>> Took some testing ... before I measured the resistors. >>> Of the 10 several are 620 k and not 2M7. >>> Bought from www.voti.nl, so beware. >>> All are marked with the right color code for 2M7. >>> >>> New to me :-) >> >> A normal handheld DVM barely measures 1MOhm correctly so anything up >> there and over will not likely be easily measured with cheap handhelds. > >Still, you'd expect it to give the same wrong reading for the same >resistances. > >Sylvia. > That should be where their other clue comes from, because the readings are typically not repeatable.
From: DrParnassus on 23 Jun 2010 11:33 On Wed, 23 Jun 2010 07:03:24 -0700, John Larkin <jjlarkin(a)highNOTlandTHIStechnologyPART.com> wrote: > >GR had tube-based meters that would measure gigohms. > >John But not accurately.
From: JW on 23 Jun 2010 11:34 On Wed, 23 Jun 2010 08:14:01 -0700 DrParnassus <DrParnassus(a)hereforlongtime.org> wrote in Message id: <vu8426do63lqaqbvpmlkhu7epjli5mf93o(a)4ax.com>: > > There are no resistors above 1Mohm that are better than 5% save custom >ordered lots that are very expensive. AlwaysWrong strikes again! One of 85 pages worth that are in stock for 27 cents each: http://search.digikey.com/scripts/DkSearch/dksus.dll?Detail&name=RHM4.7MAICT-ND
From: DrParnassus on 23 Jun 2010 11:37 On Wed, 23 Jun 2010 07:05:09 -0700, John Larkin <jjlarkin(a)highNOTlandTHIStechnologyPART.com> wrote: >Extech makes nice stuff. I think FLIR bought them to keep them from >introducing a competitive IR imager. > >John Total bullshit. Extexh did not nor does not have the expertise to make an IR imager. I worked for an IR instrument maker, and FLIR buys calibration equipment from them, and never felt threatened by our IR imager, which has been around since before extech even existed. Raytheon or Square-D is a bigger threat in that market, and I am sure that FLIR has no worries about them either. FLIR has a main market niche that nobody encroaches on. Extech was NEVER a player in IR and especially not ANY military IR, and any gear they did carry, they bought elsewhere and put their name on it. You're a goddamned idiot, as usual.
From: John Larkin on 23 Jun 2010 11:50
On Wed, 23 Jun 2010 08:24:02 -0700, DrParnassus <DrParnassus(a)hereforlongtime.org> wrote: >On Wed, 23 Jun 2010 19:28:15 +1000, Sylvia Else <sylvia(a)not.here.invalid> >wrote: > >>On 23/06/2010 9:54 AM, DrParnassus wrote: >>> On Tue, 22 Jun 2010 21:51:47 GMT, Jan Panteltje >>> <pNaonStpealmtje(a)yahoo.com> wrote: >>> >>>> Got 10 resistors on tape marked 2M7. >>>> Used one, circuit did not work right. >>>> Took some testing ... before I measured the resistors. >>>> Of the 10 several are 620 k and not 2M7. >>>> Bought from www.voti.nl, so beware. >>>> All are marked with the right color code for 2M7. >>>> >>>> New to me :-) >>> >>> A normal handheld DVM barely measures 1MOhm correctly so anything up >>> there and over will not likely be easily measured with cheap handhelds. >> >>Still, you'd expect it to give the same wrong reading for the same >>resistances. >> >>Sylvia. >> > That should be where their other clue comes from, because the readings >are typically not repeatable. But they are, to a fraction of a per cent. In the case of the carbon film resistor, I can hold the resistor body and see the resistor TC effect, also repeatable. High-ohm resistors do tend to have bad TCs. There is nothing different between using and measuring a 10K resistor or a 10M resistor. Or, assuming a little basic cleanliness, a 10G resistor. We recently needed a 100M metal-film resistor to use in a photodiode TIA thing. We tested a bunch of thickfilms and all had too much excess noise; the testing itself is a nuisance. We settled on some axial RC07 types, 50M 1% metal films from Dale Vishay, about $4 each. We put two in series with a mid-air-soldered series connection, floating above the surface-mount parts on the board so's not to waste all that board area. Dale says they're expensive because hardly anybody buys them. Too bad about the noise. A single 0603 100M thickfilm would have been cheap and a lot less hassle. John |