Unwell Fluke 1652

[JohnW2]

Have you tried removing the batteries for a few hours and see if it sorts itself out.

I think that's one of the things I tried early on, but I'll try again - just to be sure!

Anyway I made a special trip for you .....

That's incredibly kind of you. I've got to pop out for an hour or two, but on my return I'll reply properly to your post and will look for your PM.

Thanks again!!

Kind Regards, John

 

[JohnW2]

Have you tried removing the batteries for a few hours and see if it sorts itself out.

As I said briefly before, I think that this was one of the first things I tried. However, since, I’m not totally certain of that, I’ll make very sure that the batteries are disconnected overnight and see what the situation is tomorrow!

Anyway I made a special trip for you , unfortunately the repair man who also is the Managing directer was not there and only seems to do certain hours on certain days, the receptionist was helpfull and said it wont be a problem if you want to phone him and he may give you some pointers or some quotes on moving forward, she said most repairs they do are about 30 odd quid and more in-depth faults around 130, ...

As I said before, that’s very kind of you. If, as you seem to imply, this chap is clued up about repairing these things, I would imagine that, given a description of the manifestation of the problem, he ought to have a reasonable idea as to whether or not it’s likely to be repairable and, if so, at what cost.

If there were some ‘assurance’ that I was going to end up with a working/’mended’ machine, even £130 is by no means unthinkable, since I don’t think that any of the alternative strategies is going to cost me any less than that. What I obviously want to avoid is paying a substantial amount of money only to end up being told that it is not repairable!

.... though they may need to send it to fluke for about 160.

Maybe they have a special/‘secret’ relationship with Fluke, then! As I told you, Fluke told me that my machine could only be ‘repaired’ by replacing it with a new 1662, and that that would cost £430+VAT (not to mention their substantial ‘optional calibration charge’).

I will private message you his number anyway and you can decide, he may also have some faulty or refurb stuff possibly for sale, as I imagine some companys just scrap them if there uneconomical to repair, who knows he may even give you some secret fluke info.

Who knows - he sounds like a potentially valuable contact, in more than one potential way! I’ve now seen your PM and will respond to it shortly, and will then attempt to contact this chap when he is next going to be around. I’ll obviously keep you posted.

Thanks again for all your trouble.

Kind Regards, John

 

[JohnW2]

Talking of calibration, something has just occurred to me which might interest the 'techies' amongst us ....

... having now become quite familiar (at least visually) with the innards of a 1652, it has just occurred to me that, unless I need a trip to Specsavers, there is absolutely nothing 'adjustable' in the hardware.

There is obviously no problem with 'calibration' in the sense of confirming (or not!) that the machine is producing sufficiently accurate results. However, if those tests indicate that one or more of the measurements are not sufficiently accurate, then I can but presume that this can only be corrected by altering constants in the firmware - very probably at the level of digital processing, although it's not impossible that there are chips doing analogue processing which can be controlled by (digital) firmware.

Kind Regards, John

 

[Iggifer]

Talking of calibration, something has just occurred to me which might interest the 'techies' amongst us ....

... having now become quite familiar (at least visually) with the innards of a 1652, it has just occurred to me that, unless I need a trip to Specsavers, there is absolutely nothing 'adjustable' in the hardware.

There is obviously no problem with 'calibration' in the sense of confirming (or not!) that the machine is producing sufficiently accurate results. However, if those tests indicate that one or more of the measurements are not sufficiently accurate, then I can but presume that this can only be corrected by altering constants in the firmware - very probably at the level of digital processing, although it's not impossible that there are chips doing analogue processing which can be controlled by (digital) firmware.

Kind Regards, John

I think calibration is the wrong term really. I think the term should actually be "verification". My Megger had a "mareginal pass" on Low Ohms @ 10Ω. It read as 10.20Ω with an allowable tolerance of 220mΩ. You would think if there was actually any "calibration" to be done, it would have been corrected.

Bizarrely on a 230v test (Applied value of 236.9V) the allowable tolerance is 23.7V! I make that 10%!! I would have hoped that would be a much lower threshold - mine recorded 235V so I can live with that, but still, if you're testing a 230V supply and your meter is reading 203v you're going to think you've got issues!

 

[JohnW2]

I think calibration is the wrong term really. I think the term should actually be "verification".

That's what I was implying. However, many of the 'calibrators' seem to talk in terms of "... and adjustment if necessary" - and, as I said, it looks as if that could only be done in firmware, at least in my machine.

Once the signal has become digital, little short of total failure of the processor or corruption of the firmware will cause anything to change. However, particularly when using off-the-shelf components, the analogue-to-digital conversion is surely not always going to be perfect 'as initially built', nor necessarily totally stable over time, so their surely ought to be some facility for 'tweaking?

My Megger had a "mareginal pass" on Low Ohms @ 10Ω. It read as 10.20Ω with an allowable tolerance of 220mΩ. You would think if there was actually any "calibration" to be done, it would have been corrected.

Again, that's what I was implying, but it's interesting to hear that they did not 'correct' that error.

Bizarrely on a 230v test (Applied value of 236.9V) the allowable tolerance is 23.7V!

That does seem a bit ridiculous. A £5 multimeter will do a lot better than that!

Kind Regards, John

 

[JohnW2]

I think calibration is the wrong term really. I think the term should actually be "verification".

Further to what I've just written, I see that Fluke offer the folowing...

Calibration Levels
CalNet
Traceable calibration with certificate. Adjustments are carried out if points are at 70% of specification or worse. If points are out of specification, a report with measurement data is supplied before adjustment.
CalNet+
Traceable calibration with certificate and a report with measurement data. Adjustments are carried out if points are at 70% of specification or worse. If points are out of specification, a report with measurement data is supplied before adjustment.
Calintl
Accredited calibration (RvA, etc.) with a certificate and report with measurement data. No adjustments are performed. For each point, the measurement uncertainty is provided.
Calintl+
Accredited calibration (RvA, etc.) with a certificate and report with measurement data before and after possible adjustment. For each point, the measurement uncertainty is provided.

.... so definitely 'adjustments', but only if results are "at 70% of spec or worse" - which doesn't soon too great to me. I don't even fully understand what they mean by "at 70% of spec or worse"!

Kind Regards, John

 

[John D v2.0]

Well the calibration could just be a sheet of paper telling you to multiply all the readings by 1.1. The meter would be hard to use but it would be calibrated!
Mine is analogue so I could calibrate it with a fineline marker pen..

 

[John D v2.0]

I don't even fully understand what they mean by "at 70% of spec or worse"!

Of the spec is plus or minus 10% then they would adjust if it's plus or minus 7 or more %

 

[JohnW2]

Of the spec is plus or minus 10% then they would adjust if it's plus or minus 7 or more %

That sounds like a credible interpretation, and not to bad.

In any event, my point was that they clearly are able to 'make adjustments', however they do it (and not just issue a piece of paper with 'correction factors'!).

Kind Regards, John

 

[DetlefSchmitz]

My company makes calibrated compliance equipment, and also calibrates it after sales. Calibration consists of writing values into eeprom. So of course you wouldn't see (e.g.) potentiometers to adjust.The eeprom may be inside a micro-processor, so not visible as a separate chip.

 

[JohnW2]

My company makes calibrated compliance equipment, and also calibrates it after sales. Calibration consists of writing values into eeprom. So of course you wouldn't see (e.g.) potentiometers to adjust.The eeprom may be inside a micro-processor, so not visible as a separate chip.

Yes, in the absence of any adjustable hardware (like potentiometers) that's exactly the sort of thing I was thinking about when I talked about "altering constants in the firmware". I've yet to notice anything immediately identifiable as a standalone EEPROM so, as you say, it could well be some non-volatile memory within the (probably custom) processor chip. If only I knew where/what this data was, and how to get at it, it would not surprise me if there is a byte or two that could be re-programmed to say "don't do the leads-present test"!!

Kind Regards, John

 

[333rocky333]

Most wholesalers now have calibration days and they do it there and then in front of you, next time i hear theres one in your area I will let you know, and you may be able to go and see how its done.

 

[JohnW2]

An update ... patience eventually paid off.

I observed quite lot of eBay auctions for 1652s (and participated in a good few of them), but was thwarted by seemingly daft people, who persistently bid anything up to £100 more than what it would have cost them to get a "Buy it now" one! However, totally due to luck rather than judgement, I eventually stumbled across an auction which not many people seemed to notice. I was therefore able to by it at an acceptable price.

It's in good aesthetic condition, and appears to work fine on all functions. If my 'testing box' is to be believed, it is well within spec for everything.

I have by no means given up attempts to mend (or get mended) my previous one. If I could, then I could either sell it to 'get my money back', or keep it as a spare. If repair proves impossible/impractical, I will at least have quite a few useful spares for my 'new' one.

Kind Regards, John

 

[EFLImpudence]

Very good.

 

[stillp]

An update ... patience eventually paid off.

I observed quite lot of eBay auctions for 1652s (and participated in a good few of them), but was thwarted by seemingly daft people, who persistently bid anything up to £100 more than what it would have cost them to get a "Buy it now" one! However, totally due to luck rather than judgement, I eventually stumbled across an auction which not many people seemed to notice. I was therefore able to by it at an acceptable price.

It's in good aesthetic condition, and appears to work fine on all functions. If my 'testing box' is to be believed, it is well within spec for everything.

I have by no means given up attempts to mend (or get mended) my previous one. If I could, then I could either sell it to 'get my money back', or keep it as a spare. If repair proves impossible/impractical, I will at least have quite a few useful spares for my 'new' one.

Kind Regards, John

Good result John, but how much did you pay in the end?