Multimeter Calibration

[james.wilson]

yes that is fine subs and indeed the way we used to do it, but now we have em all calibrated.

I read esp to mean that you could just measure year 1 a resistor and psu, then reuse that for the next n years without anymore calibration

[esp-protocol]

yes that is fine subs and indeed the way we used to do it, but now we have em all calibrated.

I read esp to mean that you could just measure year 1 a resistor and psu, then reuse that for the next n years without anymore calibration

The way our BAFE man spoke that's how i understood it to be....!

In a simple world, if you measure a 4k7 resistor with a meter in year 2000 and every year thereafter using the same resistor, if the meter displays the same value each year then the meter hasn't changed and its callibrated ??

[james.wilson]

assuming the resistor value hasnt chnaged in that time now, but chances are the psu output will have changed

[arfur mo]

hi guys,

must admit i tend to reitre my meter every year out of habit, i do a lot of testing with it not only in alarms. i fond for some reasone even pricey units like fluke one or more range's will start to play up usually nea rthe year end and apart from that always seems a better wizzioer unit on the shelf than the one i have.

iirc the most accurate voltage source is a mecury cell as used in watches, you can make up a gang of them to suit the range being tested. rare we test for more than 20 volts with a need for a high level of accuracy.

by obtaining high spec ressistors from RS, then using ohms law for current will give an interim test of a meter between calibrations. using a freshly calibrated meter to note the results before checking a subject meter will keep you fairly honest.

has to be said as one brought up using a moving coil meter and a load bridge for servicing, i wonder the nee

i question the need for such laboritory accuracy? the way we have to use them in the field simply does not lend itself to gentle treatment, even the most careful will drop their meter sometimes.

for intance, if your looking for intermittents it matters not if the calibrated reading is 4 ohm and shows 6 ohms, if you know the 'error' then its simply the techniques and skills employed that score success.

regs

alan

[hpotter]

Sorry Arf but I think accuracy is important.

I'd want to know why a cct is 6ohms this year but was 4 ohms last year. 50% increase is a lot. I wont know that if my meter isnt the same as the eng who read it lat year.

btw, think you'll find fire guys go over 20v

[SUBS]

hi guys,

must admit i tend to reitre my meter every year out of habit, i do a lot of testing with it not only in alarms. i fond for some reasone even pricey units like fluke one or more range's will start to play up usually nea rthe year end and apart from that always seems a better wizzioer unit on the shelf than the one i have.

iirc the most accurate voltage source is a mecury cell as used in watches, you can make up a gang of them to suit the range being tested. rare we test for more than 20 volts with a need for a high level of accuracy.

by obtaining high spec ressistors from RS, then using ohms law for current will give an interim test of a meter between calibrations. using a freshly calibrated meter to note the results before checking a subject meter will keep you fairly honest.

has to be said as one brought up using a moving coil meter and a load bridge for servicing, i wonder the nee

i question the need for such laboritory accuracy? the way we have to use them in the field simply does not lend itself to gentle treatment, even the most careful will drop their meter sometimes.

for intance, if your looking for intermittents it matters not if the calibrated reading is 4 ohm and shows 6 ohms, if you know the 'error' then its simply the techniques and skills employed that score success.

regs

alan

Couldnt agree more Arf. but thats assuming everyone has common sense and experience....you cant get a degree in either of them !

high spec resistors, plus a few high spec zener diodes would foot the bill, but you need a certified meter so that you can fill in the forms and keep someone in a job filing them away.

I still have my trusty old AVO 8, and I think my old Multiminor somewhere, still in its little wooden box. wouldnt want to carry em round nowadays thpugh.

[arfur mo]

Sorry Arf but I think accuracy is important.

I'd want to know why a cct is 6ohms this year but was 4 ohms last year. 50% increase is a lot. I wont know that if my meter isnt the same as the eng who read it lat year.

btw, think you'll find fire guys go over 20v

checked the forum section 'general security queries' so don't you start

when on service i used to be looking for small ohms readings on kit connected to old 'conventional' panels for especially intermittents a hell of a lot. tbh the better way buY using current measurements as follows -:

remove any eol resistor making the zone a plain loop, check your ohms reading next. then using a 5-10 watt 12 volt bulb run it via the cct loop off a battery through your meter. set meter to current range and any intermittent will show up much better.

if you can watch the meter while tapping the suspect area's (use an assistant if needed but not his/her head ) any 'rot' will show a far bigger reading change in millamps than when on the ohms range on such small ressistance changes.

it used to be a maximum of 22 ohms on any (looped) circuit with a minimum of 1 meg between poles or to earth, now detectors have built in ressistors to limit current through the relay springs from causing damage on an unfortunate 'short so you have to know and allow this value (and yes i have seen several door loops 'smoked' by instal engineers).

better meters then came about (i have a sinclair bench meter - still works) most decerning engineers would not accept less than 20 (or 30 megs depending how high the meter could read), because once you have a leakage its unlikely to go away all by itself so eventually would be another fault call - and 'bouncers' (as then called) are at best very embarresing to your rep.

simple test for earth leakage is to test from each side of the panel voltage rail to a decent earth, anything over a volt is worth looking at, leav meter in circuit and pull cables until reading drops off.

those tests need a decent trusted meter but imo do not need an accurately calibrated meter, i'd even suggest noticing a 2 ohms variance between calibrated meters over a prologed date period would not help you find any fault better than the above tests even using a right old clanker of a meter - and i know i've used many such company supplied meters

problem is, engineer one might do his tests and mark them down all well and good, or he might guestimate readings due to time pressure or laziness. most i'd vouch sdo not appreciate why ts done and see it as a 'nag', so leading you up the gum tree when your readings differ.

and who's to say he has not done some damage to his meter beforehand? say he has had it several months have you never seen a guy check mains with leads still in the 10 amp socket and the resultant bang , fits wrong replacement fuse and you no further forward there are other various misuses.

even checking charge voltages, you willsoon suss if you meter is 'off', one or 2 panels might be higher or lower than expected but all high or all low should ring the caution bells of concern to the trained, observant and definately the experienced engineers.

so unless your leaving these tests to a dedicated supervisors or charge hands trustedto take care of their meters between calibrations, i can't see any advantage of having calibrated meters ad hock - but perhaps i just can't trust another engineer to be cabable of doing a job as good as i do , you will be the same if you want your repairs to 'stick' and not hear "that alarm you went to last week has gone off again". worst words other than 'that alarm did not activated'

regs

alan

[McP]

The way we do it at my company is we have a rig set up for the general readings ohms, volts and amps. These readings are taken with a meter calibrated every year and then checked against the readings from the engineers meter and this is excepted by the inspectors as ok.

[hpotter]

Sorry Arf, I still think accuracy is important.

No matter for security/fire/cctv etc.

Though I take the point abt "laboratory" accuracy. I guess we just need to know for sure (and can prove) that everyone who works on the same gear is getting the same reading to an acceptable tolerance - and I dont set the goalposts I play within them.

At what point a meter looses accuracy and wether once a year/week/day checking I think would depend on the type of work you do for one thing. (but I'm not that pretensious to suggest my work requires greater care /accuracy/proof than anyone elses - I do once a year and again if I drop them).

I agree also that many eng will fill out RM forms in the cafe, - alls well till something goes belly up. Those forms cover the eng & his firm. Same with spurs etc (but lets not get excited).

Like McP, We calibrate one, and check others against it.

[arfur mo]

Sorry Arf, I still think accuracy is important.

No matter for security/fire/cctv etc.

Though I take the point abt "laboratory" accuracy. I guess we just need to know for sure (and can prove) that everyone who works on the same gear is getting the same reading to an acceptable tolerance - and I dont set the goalposts I play within them.

At what point a meter looses accuracy and wether once a year/week/day checking I think would depend on the type of work you do for one thing. (but I'm not that pretensious to suggest my work requires greater care /accuracy/proof than anyone elses - I do once a year and again if I drop them).

I agree also that many eng will fill out RM forms in the cafe, - alls well till something goes belly up. Those forms cover the eng & his firm. Same with spurs etc (but lets not get excited).

Like McP, We calibrate one, and check others against it.

hi HP,

i would not any obtainable accuracy should be discarded, its the practicalitiesand the real need for scientificly acuratel calibrated meters carried by all.

scenario -:

say you do the handover tests, so you measure all the zones as you should and write it all down, your meter is infinatelly accurate having just been delivered from RS with a gold calibration cert.

now i go to site 6 months or later on the pm, and i also test all the zones as i should with my now ancient but beloved Sinclair Bench DVM (the size of the 1st panasonic mobile complete with satchel ), and i find my reading is 'off' from yours by 2 ohms, so then the same error margin also on each of the other readings. so this naturally sticks out like a sore thumb that my meter differs to yours and is established early on no probs.

now if there is a h/r it will be say example 12 ohms on my meter instead of the 10 ohms (which i lnow its 8 ohms due to the known error). so the reading is still accurate enough for our needs once extrapulated.

think of it this way, we could measure in in units of 'fartifacts', as long as we know something has changed near any meter serves its purpose regardless of its basic accuracy to aid finding faults, tbh when fault tracing do we actually care if its 5 ohms or 15 ohms to start off with? we have concern if it has altered without a known reason the whole point of the test.

obviously, make a rough estimate of 1 ohm per 10 meters + 1 ohm per contact or joint so at least we start off wth an acceptable reading, but we don't need the actual reading to be exactely the same on 2 or more meters so we don't need calibrated meters imo - well i definately don't.

trust me, and i and many 'old school' experienced service engineers, could match and i'd bet beat just anyone knob job just armed with an old fashion swing needle meter, against any young gun using a calibrated DVM, especially on fault finding and prevention.

modern tools will aid and help make life easier, but they simply don't circumvent true know how

imho i think all trainees should be taught using a swing needle meter - no dvm's, so they can better understand the instrument and what it can actualy tell them.

regs

alan