Handheld meter robustness testing

[joeqsmith]

Back about 10 posts, I linked the FTC document describing the requirements for the various markings.

Yes, but your missing the point about multiple regulation regimes and regulatory risk.  Regulations change.  Multiply that by how many regulatory regimes (eg, countries) this meter is sold in.  Sometimes when they change regulations they do it too fast to clear out all the stock before they come into effect.

I'm missing a point? 

[rsjsouza]

The change in certifications seems to have removed the listing at UL - I know that a listed product requires long term commitments regarding quality inspections and other details, which is a continuous stream of negative revenue.

The UL-CSA mark is equivalent to a UL mark.  There's no need to have both any more as both agencies can cross certify by agreement.  Fluke moved to CSA many years ago for no doubt good reasons which have nothing to do with "long term commitments regarding quality."

I am not sure what are the commitments of the UL or UL-CSA marks, but I was referring the UL listing number which requires annual "audits" on the manufacturing facility and design changes and not your typical "fire-and-forget" certification testing of your pre-release / initial production products. This is a long term committment that prevents the product from having its quality erode over time as Extech, for example.

Quite interesting is how the 87V deals with the "hi-res" mode - does that stay enabled after the meter is power cycled? If not, that would greatly detract from a usability standpoint.

Just , even the UT61E comes with 0.01\$\Omega\$ resolution by default.

No, the old Fluke meter has no sticky settings like Brymens.  IMO, the meter should have an auto mode for that high res.  I doubt many electricians have a need for that feature.  I can't see too many hobbyist wanting it but then again, if you were driven by marketing, maybe. 

I think the only two meters I looked at that can resolve 1mohm are the Gossen and TPI.  I can't really count the UEI meter.  If the Gossen did not have so many problems, it would be a very nice meter.  I wonder if they ever did anything with it or decided it was not worth going after.     

Too bad. My previously owned 179 did not have sticky settings as well and it was quite expensive, although it seemed to be tailored to a different market.

1m\$\Omega\$ is quite a resolution - on par with a number of 5-¹/₂ digit meters.

[joeqsmith]

I wonder if they are looking at it in the ACV setting and measuring the leakage of the cap.   I could check my Brymen in this same way and see if I can get the same accuracy and resolution as the 87V in the volts mode.  I think the Brymen is also AC coupled in this mode so it should perform well.    I could try and measure it but  then again, normally you look at AC with the AC settings and that cap is not going to be the 10 zigga ohms.   

Anyone else care to chime in on this comment as right now, I am a bit lost as to what the OP is describing.   

[HKJ]

High impedance mV ranges will often clamp input below 1V (Through a resistor/PTC), i.e. a normal ohm meter will not always show the correct impedance. In all my reviews I use a source meter where I can control the voltage when measuring ohms.
They will often switch a 10Mohm resistor in for AC mV
I do not have the 87V and cannot check that meter.

[2N3055]

It it probably leakage of input protection circuits plus capacitor leakage (probably not major contributor) when AC coupled..

[rsjsouza]

If that is true, what is unique? Heck, even the UT61E claims >3G\$\Omega\$ input impedance in mV ranges... My Racal Dana measures the UT61E as having 52M\$\Omega\$, but I suspect this may be wrong as clamping may be influencing it (it puts out a couple of V)

[WhichEnt2]

If that is true, what is unique? Heck, even the UT61E claims >3G\$\Omega\$ input impedance in mV ranges... My Racal Dana measures the UT61E as having 52M\$\Omega\$, but I suspect this may be wrong as clamping may be influencing it (it puts out a couple of V)

Sure it is clamped.
UT71E and Fluke 87V (hi Z option on) mV input resistance reads OL while being measured by Agilent U1252A
But if that resistance is measured by F87V it reads about 4.7 MOhm on UT71E and 16 MOhm on U1252A.
Fluke 87V on 11 MOhm load (DC V input resistance) feed about 2.7 V into leads.

[joeqsmith]

Someone wrote me about the Hi-Z mode.  Strange the spec sheet does not reflect the input impedance when this mode is selected.   I suspect this is what the OP was referring to but I am not sure where they are coming up with the 4G.   

With as popular as the meter is, and this Hi-Z mode being a selling feature, strange there are not published specs or at least posts where people have measured it. 

https://www.eevblog.com/forum/chat/fluke-87v-high-z-mode-input-impedance-value/
https://www.eevblog.com/forum/testgear/multimeter-input-impedance-riddle/

[joeqsmith]

I tried using my vintage electrometer to measure the input impedance directly.  This uses a constant current source and can output around 30V.  With the standard DCmV selected, I read 10Meg.  No surprise.   Selecting Hi-Z, the electrometer measure 22.4Gohm.  However,  this means it is putting 2.24V across the meter.   

Using a constant voltage source at 490.03mV (within what the 87V can measure in the DCmV range) and using the electrometer to measure the current, I get about 23pA.  So roughly 21.3Gohm. 

Of course if you really needed higher input impedance for mV measurements, even this old electrometer is 50Tohms and has a resolution of 10uV.  Then again, it doesn't fit in the palm of your hand.

If anyone else has the 87V,  I am interested in hearing how this new one compares with others.   

[joeqsmith]

If that is true, what is unique? Heck, even the UT61E claims >3G\$\Omega\$ input impedance in mV ranges... My Racal Dana measures the UT61E as having 52M\$\Omega\$, but I suspect this may be wrong as clamping may be influencing it (it puts out a couple of V)

My UT61E has had a LOT of mods and I doubt very much it would be the same if it were still a virgin meter.    In the DVmV range, I used a 198.29mV source to keep it in range.  I measure 28.5pA with the electrometer.   So very close to 7Gohm.  At least you know I cleaned it fairly well and their claim of being greater than 3Gohm seems correct. 

For fun, I tried to measure the pre-production 121GW that I damaged a few times, repaired and modified.   Measuring it directly with the electrometer, 0.200Mohms.    I then tried it with a 183.40mV (well within the range of the meter) and measured 918nA or 199.782Kohms.    Again, I doubt the people who have the released meter (on Amp Hour these were referred to as Beta units) would get the same results but it would be interesting to compare. 

[joeqsmith]

For the ANENG fans or in my case the KASUNTEST, looking at the ZT102 that I modified to survive the 14KV transient, in the DCmV with 198.3mV applied, the electrometer measures 14.7pA or 13.49Gohm.   Again, that meter has some major mods done to it and I would expect a virgin unit would be different.  Does show that the dope I used on some of these meters is not causing much of an error.   

[HKJ]

Many meters are high impedance in the DC mV range. How high that is I will not say anything about, my normal setup is way to noise to measure gigaohms at sub volt ranges, I will only say it is in the gigaohm range.
I.e. either the meter is 10Mohm or it is gigaohms, very few meters are lower impedance in the mV range, but overvoltage will usual clamp with a fairly low impedance (kohm).

[joeqsmith]

The TPI (Summit) 194II lime green meter with 490.50mV applied, the electrometer measure 2.22nA or 220Mohm.  Now this meter is damaged and I never was able to find a new front end IC to repair it.  It lost some of the multiplexer.  It is still able to accurately measure the DCmV but the low levels may not be normal.   

I am not sure why the 121GW was so low.  It could very well be because of my mods.  Then again, I damaged that meter a few times as well.  Someone else really needs to check it.

[HKJ]

I am not sure why the 121GW was so low.  It could very well be because of my mods.  Then again, I damaged that meter a few times as well.  Someone else really needs to check it.

I suppose you are going to check it again when you get the production version. I am also going to check it and do my usual multimeter review of it.
I suppose the timing estimate now is late March.

[joeqsmith]

I mentioned a use an Extech meter for work.  Some time back I bought a CEM DT9939 which is identical to my EX540. This is a 40,000 count tri-display meter with all the basics including AC+DC.  The one I use is coming up on six years old and has had seen a fair amount of abuse.   Ruby Electronics was selling these for $120 which was a great deal.    Again, I damaged the meter during my tests and only replaced the parts.  So no modifications for a change. 

The CEM in DCmV with 198.20mV applied, the electrometer measures 1.5pA or 132Gohm.   I then increased the input to 490mV which is outside what the meter can measure.  This yields 4.1pA or 119.5Gohms.   A fair amount higher than I saw with the Fluke 87V but again no where near the 50Tohms of the electrometer. 

[joeqsmith]

I am not sure why the 121GW was so low.  It could very well be because of my mods.  Then again, I damaged that meter a few times as well.  Someone else really needs to check it.

I suppose you are going to check it again when you get the production version. I am also going to check it and do my usual multimeter review of it.
I suppose the timing estimate now is late March.

There is one for sale in on this site and the owner has offered to ship to the US.  Maybe other places as well.  I thought about it but with the time it takes me to run the tests, the last thing I want to hear is how what I was showing was a Beta meter.  I would rather hold off until they get everything sorted out and are happy with their testing.   I doubt this particular metric would change much from the Beta units.  It's non-destructive so no problem for people who own them to make the measurement and compare results. 

The pictures I posted of the metallic dust between the switch contacts was discovered when I first damaged the meter.  I had cleaned the circuit board a few times after taking those pictures.   But again, the mods are a strong possibility.   Sub 200K, easy enough to check.. 

[joeqsmith]

Many of the meters I have including the Brymen BM869s, Fluke 196, Gossen M248B, UNI-T UT181A are 10 or 11Meg.  Even the free harbor freight meter was close to 10M.   A few meters were 5M.   I found one that was lower than the pre-production 121GW.   The YX-360TR.  Again, highly modified after being damaged.  This meter measures close to 2Kohms in the DC 100mV range. 

[rsjsouza]

Many of the meters I have including the Brymen BM869s, Fluke 196, Gossen M248B, UNI-T UT181A are 10 or 11Meg.  Even the free harbor freight meter was close to 10M.   A few meters were 5M.   I found one that was lower than the pre-production 121GW.   The YX-360TR.  Again, highly modified after being damaged.  This meter measures close to 2Kohms in the DC 100mV range.

That is a novelty. All M830B clones I have seen have 1M\$\Omega\$ of input impedance in all DC ranges. They even mention this in their manuals.

[joeqsmith]

Many of the meters I have including the Brymen BM869s, Fluke 196, Gossen M248B, UNI-T UT181A are 10 or 11Meg.  Even the free harbor freight meter was close to 10M.   A few meters were 5M.   I found one that was lower than the pre-production 121GW.   The YX-360TR.  Again, highly modified after being damaged.  This meter measures close to 2Kohms in the DC 100mV range.

That is a novelty. All M830B clones I have seen have 1M\$\Omega\$ of input impedance in all DC ranges. They even mention this in their manuals.

This is that yellow one that was given to me.   I wonder if I read it wrong.  ...... Yep.

[joeqsmith]

It seems the released 121GW meters are 10Meg in the mV range.   Looking at the schematic and pictures of the new boards, I suspect that the low resistance I measured is finding it's way through U14 and R91 which are no longer populated.  R91 just happened to be a 200Kohm so this is a strong possibility.   

Sure enough, put it into temperature mode, 10Meg.   It would be interesting to know why they wanted such a low input impedance and then decided to go away from it.   

[joeqsmith]

Do a search some time of the 87V to see what problems people have had with this pinnacle of meters and you may be amazed.   I wanted to see if anyone else had performed some sort of AC line test, intentional or not.  I think I have my answer:

https://www.eevblog.com/forum/testgear/fixing-a-fluke-87v/

[Fungus]

Do a search some time of the 87V to see what problems people have had with this pinnacle of meters and you may be amazed.   I wanted to see if anyone else had performed some sort of AC line test, intentional or not.  I think I have my answer:

https://www.eevblog.com/forum/testgear/fixing-a-fluke-87v/

Jeez. My 13 Euro LIDL multimeter survived that...

[rsjsouza]

Joe, you don't need to look further. Another thread posted today:
https://www.eevblog.com/forum/repair/fluke-87v-105223/

[jordanp123]

Has anyone torn apart/tested one of the new AMES meters from Harbor freight ? Saw them today and they looked decent for an electricians meter, had a ETL listing as well.

[rsjsouza]

MOVs would give up if limiar overvoltage is applied for too long - the circuitry could probably withstand 1.1~1.2kV, but the MOVs would suffer thermal stress. All in all, I don't think any meter would be too different in this regard (unless they used higher voltage MOVs).