Adam Savage on multimeters

[Cubdriver]

Because, as has been repeatedly pointed out to you, there are applications where things are for all intents and purposes cast in stone.  Things like military and aerospace test and maintenance.

And I've said repeatedly that I don't believe that those things are as cast in stone as you think they are. If they were we'd all still be using the Fluke 27FM.

If something suitable and yellow comes along then things will change. Not overnight, sure, but... change can happen.

Nor have I said that things never change - I'm simply pointing out that there are instances where it is complicated, time consuming and expensive to change, and in those instances change is not implemented lightly, especially if what is currently in use works.  If I can buy another of this meter that fulfills my needs and is currently written into my test procedure, or I can buy a 'new and improved' model that, for instance, settles faster or has a better continuity circuit (neither of which has any bearing on what I'm using the meter for), and spend what will likely be tens of thousands of dollars to certify this new meter for use in my testing, WHY WOULD I DO THE LATTER?

You stated earlier "I'm not saying it's no good, I'm saying it's still got a little bit of space for improvements."  Well, I think most anything falls into that category - you can nearly always make some slight improvement in something.  The question is, is that improvement enough to justify the time and effort for those that use the meter to recertify a new device?  It appears that, for the time being, Fluke does not feel that it is.

-Pat

[vk6zgo]

The funny thing is, Broadcasting was a "Regulated industry", but really, at least in Oz, the regulator didn't really give a stuff about how you achieved it, as long as the ultimate product transmitted to the viewers/listeners was within spec for the applicable Broadcast Standards.

TV & sound broadcasting equipment was often subject to major modification over the decades of life which were common in the past, to allow for the unavailability of earlier components.

Replacing all failed components "like for like" would have been incredibly costly, & pretty much untenable.

I work in the broadcast industry as well, and we are a regulated industry, still.

Our license to transmit requires us to do (and not do) a lot of things, and some of those are related to physics, like frequencies, output power, complicance to EBU standards (R 128 for programme loudness comes to mind) et c.

Fortunately, we (and our sub-contractor who is responsible for transmissions) only need to prove that we are right, and that shall be verified by the regulator. We don't have to have a process approval, only an end result approval.

The overwhelming part of "regulation" as a matter of fact controls what you publish, which can be checked with patience, a TV set, and a stopwatch.

In medical, aerospace, and similar industries, the result counts, but the process might cause harm while producing the result, so the process is regulated to much higher degrees.

Back in the day, in Oz, the broadcast regulator was stern but fair, & concentrated mainly upon the technical aspects.

Strangely, sound broadcasting was regulated by the PMG's Dept, (later hived off to a dedicated department), but TV broadcasting was under the control of the "Australian Broadcasting Control Board", who started out as a "programme standards" body, but ended up as a technical regulator as well.

Today, broadcasting is pretty much "self policing", as the current regulator, ACMA, have pretty much lost all its Engineering staff, concentrating upon employing lawyers & "bean counters".

That is why I refer to broadcast regulation in the past tense.l

[floobydust]

I did laugh and then cry that Fluke HQ is next door to Boeing in Everett Washington.
Oh look, they both had "Max" projects as a rehash of some old product because they are unable to make anything clean slate.
The 87 has changed over the years, old IC ran off 9V, now 5V so loss of compliance voltage but who really noticed.
I think Fluke had a great monopoly by patenting multimeter input protection (1993) so nobody could make something that tough without fear of litigation. It did give them a leg up for many years.

Lots of market analysis on the global multimeter market, anticipated to increase 4.3% from USD $847M 2018 to $1,047M in 2024. It grew 5% from 2014 to 2019. Looks worth it to make something new.
https://www.marketsandmarkets.com/Market-Reports/digital-multimeter-market-19694516.html
https://www.credenceresearch.com/report/digital-multimeter-market

But the 121GW Kickstarter 2017 raised AU $645K. Does it really cost that much to make a clean slate multimeter? And Dave's not going down that road again for some reason.

[EEVblog]

Because, as has been repeatedly pointed out to you, there are applications where things are for all intents and purposes cast in stone.  Things like military and aerospace test and maintenance.

And I've said repeatedly that I don't believe that those things are as cast in stone as you think they are. If they were we'd all still be using the Fluke 27FM.

Ironically Fluke had to dumb down the new 28-II meter and make the 27-II meter just to satisfy their military customers.

[EEVblog]

I'm not saying it's no good, I'm saying it's still got a little bit of space for improvements.
But ... the people here say it's untouchable, that no change is ever possible or even desirable.

No one is saying that, we are just pointing out a reason why Fluke aren't that keen to upgrade it. Who knows, they might have been working on a new upgrade for 5 years or something and it's coming out soon, or maybe not.

I find it much easier to believe that if Fluke announces the Fluke 87 VI tomorrow then everybody here would be on the Fluke website within 10 seconds to read the specs and find pricing/availability.

Of course they would, but that doesn't mean it's the overiding reason they upgrade like it is for other companies. The likes of Fluke, HP/Agilent/Keysight, and Tek are very different companies to your average, and have very different customers they focus on.

[EEVblog]

I find it much easier to believe that if Fluke announces the Fluke 87 VI tomorrow then everybody here would be on the Fluke website within 10 seconds to read the specs and find pricing/availability.

You included.

Of course we would go look. But that doesn’t mean we’d rush to actually replace our 87V’s, nor that we would rush to change all our test procedures at work.

And that's the trick.
These meters are so expensive that people aren't just going to rush and out buy the new one. The number of people doing is a drop in the ocean compared to regular sales which very likely aren't drying up because the model is "old".
If Fluke sell say 100,000 87V's a year and continue to do so regardless, what's an extra few thousand who might rush out and buy the new model mean? Nothing really.
The big volume customers like government and military will buy either the existing model or the new model regardless, so that number doesn't really change.

[Fungus]

If they were we'd all still be using the Fluke 27FM.

Ironically Fluke had to dumb down the new 28-II meter and make the 27-II meter just to satisfy their military customers.

The 87IV also seems to have been a climb-down - too big of a change all at once?

[Fungus]

I did laugh and then cry that Fluke HQ is next door to Boeing in Everett Washington.
Oh look, they both had "Max" projects as a rehash of some old product because they are unable to make anything clean slate.

Hah! I hadn't made that connection before. Is the introduction of a "Max" model the sign that a company is now fully institutionalized/fossilized.

[rsjsouza]

You seem upset that Fluke aren't making the meter you want at the price you want. I'd try and get over that, because they probably never will.

No, I'm trying to understand why we'll all still be using the Fluke 87V a thousand years from now.

Makes no sense to me. 

Fungus, we talked about this...

I'm not saying it's no good, I'm saying it's still got a little bit of space for improvements.

But ... the people here say it's untouchable, that no change is ever possible or even desirable.

I understand your point of view but I don't necessarily think that the more swnsible opinions are really saying the 87 is absolutely perfect, but instead that corporate and government pragmatism are the masters of the 87 innovation roadmap.

I find it much easier to believe that if Fluke announces the Fluke 87 VI tomorrow then everybody here would be on the Fluke website within 10 seconds to read the specs and find pricing/availability.

You included.

This already happened with the 87 MAX, which is another proof of pragmatism: let's test the waters by launching a very low risk  rebadge of another product already in production (the 28II). If "sticks", we make a PR campaign explaining why the 28II had to be discontinued.

[tooki]

Because, as has been repeatedly pointed out to you, there are applications where things are for all intents and purposes cast in stone.  Things like military and aerospace test and maintenance.

And I've said repeatedly that I don't believe that those things are as cast in stone as you think they are. If they were we'd all still be using the Fluke 27FM.

Just because you refuse to believe something doesn’t mean it isn’t true. Lots of people believe the earth is flat, but that doesn’t make it so...

[Fungus]

And I've said repeatedly that I don't believe that those things are as cast in stone as you think they are. If they were we'd all still be using the Fluke 27FM.

Just because you refuse to believe something doesn’t mean it isn’t true.

Do you have any evidence to the contrary, eg. en actual example of an official procedure/manual which says "Take the Fluke 87V and select voltage mode, then measure voltage at test point A".

Even if you do, how common would they be compared to one that says: "Using an approved multimeter, select voltage mode then measure voltage at test point A".

Personally I'd hope they'd choose a much safer meter than the 87V for critical systems or life/death situations, eg. one that doesn't have four holes in the front to make mistakes connecting the test leads to it and only has one function per range selector position.

eg. The Fluke 110.

Oh, wait! That meter's only CAT III 600V...!

Why would Fluke do something like that? Is it because people might buy the simple/easy 110 instead of one of their cash cow meters? 

FWIW I just looked at Flukes entire range and there's not a single meter on their site with only two input jacks and more than CAT III 600V rating. That's no accident, it compromises safety, and it shows just how "sold" Fluke are these days.

Edit: No, I missed one! The 113 has a CAT IV 600V rating. It fooled me with fuzzy pixels and because they didn't write "Cat III 1000V" next to it on the meter. See below....

Lots of people believe the earth is flat, but that doesn’t make it so...

Do "lots" of people seriously believe that?

[Fungus]

These meters are so expensive that people aren't just going to rush and out buy the new one.

According to Fluke it's their "Best value" meter!



In other news: The 179 is their "best overall" meter. 

[Fungus]

Actual innovation from Fluke?

I just saw this on their web site. An almost completely automatic meter for $150, CAT IV rated, Low Z check, bar graph, fast continuity buzzer (they actually give a number for that - 500uS response!), only two positions on the range selector switch and one secondary function (capacitance).

Same electrical safety as an 87V but much safer operation. Maybe Fluke's safest multimeter is a $150 multimeter...?

I really wasn't expecting a meter like this from Fluke. It's more like something Aneng would come up with.

https://www.fluke.com/en-sg/product/electrical-testing/digital-multimeters/fluke-113



OK, it's not much use to an EE but it shows Fluke hasn't fired all the engineers yet.

Over to joe... 

[joeqsmith]

Do you have any evidence to the contrary, eg. en actual example of an official procedure/manual which says "Take the Fluke 87V and select voltage mode, then measure voltage at test point A".

Even if you do, how common would they be compared to one that says: "Using an approved multimeter, select voltage mode then measure voltage at test point A".

Personally I'd hope they'd choose a much safer meter than the 87V for critical systems or life/death situations, eg. one that doesn't have four holes in the front to make mistakes connecting the test leads to it and only has one function per range selector position.

The Fluke 189 that a friend gave to me was owned by a friend of his who had passed away.  The meter was actually own by a company that is no longer around.  This company wrote software specifically for that meter (using the interface cable).   The software helped semi-automate some of the measurements.   I understand that when the meter was reaching it's end of life, the company basically bought much of the stock or had some exclusive deal.  I doubt very much safety was considered in their choice to base their testing on the 189 as much as the features it offered. 

The US DOD used to have a website where they had scanned many manuals for the equipment they used.   It was a gold mine for people like me who bought obsolete equipment for my hobby.   After 9/11, they locked it down and I'm not sure if it still exists.    Many of the scanned manuals you find today were taken from that site.  Have a BAMA for example. 

Anyway.... reading these manuals, in many cases the test procedures would call out specific equipment to be used.   I own an very old HP8640B RF generator.   Large, heavy, LED readout, mechanically tuned....  It seems like it was a staple in the military due to how clean the output signal was, long after I would have expected the product to be obsolete.   

https://bama.edebris.com/download/hp/8640b/tm%209-4935-601-14-7&p.pdf

[tooki]

And I've said repeatedly that I don't believe that those things are as cast in stone as you think they are. If they were we'd all still be using the Fluke 27FM.

Just because you refuse to believe something doesn’t mean it isn’t true.

Do you have any evidence to the contrary, eg. en actual example of an official procedure/manual which says "Take the Fluke 87V and select voltage mode, then measure voltage at test point A".

I’ll leave it up to (ex-) military folks to answer that, as they know it better than I. I have seen (non-military) procedure manuals specifying a particular model of test gear, though.

Lots of people believe the earth is flat, but that doesn’t make it so...

Do "lots" of people seriously believe that?

While they’re still a minority, it’s a shockingly large number of people. (I mean, it should be zero. Not several percent. Some surveys put it at over 5%!!  ) If you have Netflix, there’s a glorious documentary about them called “Behind the Curve”. Be judicious about eating or drinking while watching it, as those people are inadvertently hilarious.

https://www.netflix.com/title/81015076?s=i&trkid=255824129

[Fungus]

Anyway.... reading these manuals, in many cases the test procedures would call out specific equipment to be used.

Is just that so that the grunts don't try to measure volts on the signal generator or is detailed operation of the equipment used contained inside each test procedure?

ie. Does it say "measure the voltage using the HP1234A" then assume you're going to read the HP1234A operators manual to find out how to measure voltage, or does the test procedure include pictures of the HP1234A along with where to put the cables and which buttons to push.

The "read the device manual" makes most sense but I've seen some military paperwork and it's measured in shelf-feet so the other option is entirely possible. OTOH it makes you incredibly dependent on availability of the devices so it seems like a really bad idea. It's also a lot more work to produce the documentation, but hey... taxpayer money!

[dcac]

According to the multimeter spreadsheet....

The Bryman BM869 & Greenlee DM-860A have a single 9V battery.

The Fluke 87V has a 9V battery

Fluke 289 & 287 use 6 AA batteries. Those meters need an update for battery power draw.

I personally prefer 9V batteries due to leakage issues with AA/AAA

9V batteries can leak too - but perhaps not as violently. On some leaky 9V I’ve encountered it started at the terminals then over to the connector and then it managed to creep inside the battery leads to the PCB.

[Fungus]

I’ll leave it up to (ex-) military folks to answer that, as they know it better than I. I have seen (non-military) procedure manuals specifying a particular model of test gear, though.

Yes, I've seen lists of "Equipment to be used" followed by a list of model numbers. What happens if the device breaks and they can't get another one? Work can't stop, there has to be a substitute device.

(Pass me a pencil and I'll fix the manual...)

The Army plans for availability and obsolescence. In the case of the Fluke 27FM I believe they made a deal with Fluke to garantee production for X number of years which is why Fluke was still making them well into the 21st century.

[dcac]

But the 121GW Kickstarter 2017 raised AU $645K. Does it really cost that much to make a clean slate multimeter? And Dave's not going down that road again for some reason.

I believe the 121GW was an unique situation where the manufacture (UEi) approached Dave with the idea of making a EEVBlog multimeter from scratch based on his input and to some degree his supervision - but as I understand it - he couldn’t really make any demands on the performance of the final product. So perhaps not really an ideal situation.

[tooki]

I’ll leave it up to (ex-) military folks to answer that, as they know it better than I. I have seen (non-military) procedure manuals specifying a particular model of test gear, though.

Yes, I've seen lists of "Equipment to be used" followed by a list of model numbers. What happens if the device breaks and they can't get another one? Work can't stop, there has to be a substitute device.

The kinds of customers that have that level of strictness aren't exactly the kind who only buy just one. They have contracts with stipulations and will stock up if need be.

(Pass me a pencil and I'll fix the manual...)

Yeah, and then get your ass fired/discharged for gross negligence.

The Army plans for availability and obsolescence. In the case of the Fluke 27FM I believe they made a deal with Fluke to garantee production for X number of years which is why Fluke was still making them well into the 21st century.

Ummmm, this proves the need for long-term product availability.

The existence of the 27II (and the 83V) are almost certainly due to the need to be drop-in replacements for old procedures designed around average-responding multimeters. (It's not cost savings, since they actually often cost more than the mainstream TRMS 28II and 87V.)

[joeqsmith]

Anyway.... reading these manuals, in many cases the test procedures would call out specific equipment to be used.

Is just that so that the grunts don't try to measure volts on the signal generator or is detailed operation of the equipment used contained inside each test procedure?

ie. Does it say "measure the voltage using the HP1234A" then assume you're going to read the HP1234A operators manual to find out how to measure voltage, or does the test procedure include pictures of the HP1234A along with where to put the cables and which buttons to push.

The "read the device manual" makes most sense but I've seen some military paperwork and it's measured in shelf-feet so the other option is entirely possible. OTOH it makes you incredibly dependent on availability of the devices so it seems like a really bad idea. It's also a lot more work to produce the documentation, but hey... taxpayer money!

For example they may have and alignment procedure for a radio and will call out the equipment to use and how to set it up.  It would seem typical.   

I have worked on a few designs for the DOD.  Everything was tightly controlled.   Say you had a 10uF 16V Alum Elec from some vendor and you decided to change to the same part from another supplier....  Nope..   Want to change your code, nope.   Also I understood that if something were to happen where we could not supply parts, they had some ways to make sure that was resolved.   
Some time ago I posted about Northrup and the MX missile.  Takes it to a new extreme. 

I understand some civilian industries are very procedure driven.  Makes sense really.  Medical and nuclear power generation to name a few.   

[mansaxel]

I understand some civilian industries are very procedure driven.  Makes sense really.  Medical and nuclear power generation to name a few.   

The B-17 Flying Fortress is very complicated to get in the air and back down again, because there is a sequence of events that need to happen just right, or your flight ends prematurely in a bad way.

Not until the check list was invented and mandated did the accident frequency drop enough that they could go to war and get shot down by external causes.  Source: USAF

When you have people that might be under a lot of pressure performing complicated procedures, you want the procedures to be unambigous, as fail-safe as possible, and free from traps. 

Imagine being the radio operator in the movie Crimson Tide, trying to fix a VLF receiver with the two highest-in-command officers on the boat standing over him.  The EE with their fantasy and skills might make a Doc Brown appearance and conjure up a better-performing receiver from a chewing gum, a steel comb, and some cling film from the galley since their Brymen can measure the MF thanks to its frequency response, but the petty officer will reasonably reliably swap a component or sub-assembly, adjust the PSU regulation to perhaps 2% precision using a Simpson 260, slap the receiver together, and receive the message.

Likewise, the only non-manufacturing organisation I've ever worked in that took ESD seriously was the military. Probably because they have so much gear, run it so long, and have a lot of paperwork to get a FRU issued that there actually is some follow-up on causes of failure. Consequently, there is procedure.

[Mortymore]

SpaceX also likes the yellow stuff 

November 2020


May 2020

[bd139]

That’s because they can easily get it in the right mode wearing those clothes 

[EEVblog]

Do "lots" of people seriously believe that?

While they’re still a minority, it’s a shockingly large number of people. (I mean, it should be zero. Not several percent. Some surveys put it at over 5%!!  ) If you have Netflix, there’s a glorious documentary about them called “Behind the Curve”. Be judicious about eating or drinking while watching it, as those people are inadvertently hilarious.
https://www.netflix.com/title/81015076?s=i&trkid=255824129

I agree, one of the funniest docos I've seen in a long time.