U1272A update

[PetrosA]

I've been using mine at work and also (while at work...) playing around with the switch to see if I could reproduce any of the issues. Some but not all the halfway positions give screwy readings, which may only show up while turning in one direction and not the other - for instance:

Moving from OFF to ACV will give a 0.L readings (no mode symbols) and secondary display flashes between 0L and 301(?) - ACV to OFF just shuts the meter off;

Moving from ACV to AC mV will change my secondary reading from temp to DCV and locks up the volt reading at 7.5V AC, while moving back from AC mV to ACV will make the mV reading jump around until I move the dial far enough that it switches back to ACV;

Between mVAC and VDC both directions seem ok, secondary display remains fine.

Between VDC and mVDC in either direction the display goes wonky until it settles on 0.L, secondary display remains fine;

From mVDC to Ohms it goes into the 0.L state in volts mode - from Ohms to mVDC the reading goes from 0.L Ohms to DCV mode with 21 MOhms displayed (both the DCV symbol and MOhms symbols are displayed), secondary display remains fine;

Ohms to diode test also shows both DCV and Ohms symbols with 21 MOhms. Move it slightly further and the voltage ramps up, the meter beeps a few times and finally displays . 0L V (additional space between decimal and 0), DCV symbol and diode symbol are active, secondary display remains fine;

From diode check to cap check the meter will start a steady beep and displays both DCV and diode symbols with a 0.0004V reading. Move further it will do the same kind of ramp up as between ohms and diode. The opposite direction will activate the DCV symbol and give a dead reading for about 10 seconds, after which it goes to .0L

Between A and uA the display goes wonky again, with readings as high as 12A shown. From uA to A, the display will go to 0L. and the backlight stays on (this seems to be connected to the lead sensors).

The bigger problem for me with the two current settings has to do with the alarm. Going from Off to A, both inputs work fine with no alarm. Once the meter has been in uA range, going back to mA/A setting will produce alarm beeps in one, the other or both inputs until the meter "gets the message" (no standard time) that it's ok for the lead to be in either of the two holes. This continues until I shut the meter off again (see video). It will also sometimes beep a warning in volt mode with the probe in the volt terminal after being in uA mode (I wasn't able to capture this for the video )

[Kiriakos-GR]

Nice find PetrosA , I will start also to believe what some one else wrote ,
in another similar thread : That looks like that some one loaded windows Vista , on a windows XP hardware.  

[PetrosA]

I forgot to add that I haven't had any issues during normal use with my 1272 except for the probes alarm, and that's been rare since I'm normally using a clamp meter (U1211A) for current. I just happened to be measuring current on a SMPS feeding a gang of LEDs at >25kHz and had to use this meter for that.

[Kiriakos-GR]

Just received  an reply from Agilent , about an question that I did related to firmware updates .

It is confirmed that both U1253B OLED DMM  and U1270 series ,
could receive firmware updates, by the use of the IR-USB cable .

For the U1253B there is all ready one update available .
(resolves an issue about the start-up screen design )

Additionally the users evaluating program about HH of Agilent is on Hold ,
and my own demo unit too.

Lets hope that their technicians works faster and coordinates better,
than the postman that they use, for delivering the mail .
  

[PetrosA]

I'm getting a response ready for someone at Agilent who I contacted about this. Can I assume from the lack of responses that no one else has been able to reproduce the mA/A-uA jack warning alarm?

[Kiriakos-GR]

I'm getting a response ready for someone at Agilent who I contacted about this.

PetrosA , as Greek to Greek , I will say to you " to hold an small basket " about the Agilent response.

Personally I have start to have doughs.

[saturation]

The errors occurring are fairly diverse, it appears to depend on how much the firmware is confused by some unstable connection; the worse is a full lockup, with no ranges working, the best is the meter reads inaccurately but its readings are mostly reproducible.

You can get a series of repeatable errors after one glitch, then if you shut it off, and wait 1/2 minute, it will be fine until the next set glitch sets it off.  If you wait for only a few seconds after off, the glitch will not reset.

I've so many various types of errors happening on several 1272a but regardless, they can all be reset once the meter is off for some time.

I have not experienced an error once the DMM boots properly and is working on any range: that is if you are already working on e.g. VDC, then switch to ohms, VAC, mV, temperature etc., all is well, so long as you do not power OFF the unit through the rotary switch.  If the machine goes to sleep, its fine, if you wake it by pressing buttons.

I am about to disassemble one of the meters and examine the switching area and PCB to see if the problem can be user resolved.

[ejeffrey]

Can someone with one of these meters check something I just discovered on the U1253A?

When measuring on the 10 nanofarad (smallest) capacitance range, and measuring a capacitance below 100 pF, the meter is extremely slow to respond.  The display update rate is like 1/s, but if you zero the reading and then hook up a 33 pF capacitor it will take 10 seconds or more to climb to the appropriate value.  Disconnect the leads and it will take another 10+ seconds to fall back to near zero.  This makes it very difficult to get a good reading as you never know if it has actually reached the target value.

A 330 pF capacitor on the same range is no problem.  It gives a good reading immediately and has the normal update rate of several Hz.  I checked on a couple of other varieties of (less expensive) meters and they had no problem with the small capacitance.  Anyway, I am curious if the 127X series shows this behavior as well.

[Kiriakos-GR]

Hi ejeffrey .

It would be nice to post this question at the Agilent forum too.

https://forums.tm.agilent.com/community/viewforum.php?f=552&sid=ac542c0140e499099b6d22a1931f405e

By letting them know about it , they will investigate if there is a problem.

[saturation]

I have a 1252a, it easily measures 2.4pF quickly and easily, just rechecked.

Can someone with one of these meters check something I just discovered on the U1253A?

When measuring on the 10 nanofarad (smallest) capacitance range, and measuring a capacitance below 100 pF, the meter is extremely slow to respond.  The display update rate is like 1/s, but if you zero the reading and then hook up a 33 pF capacitor it will take 10 seconds or more to climb to the appropriate value.  Disconnect the leads and it will take another 10+ seconds to fall back to near zero.  This makes it very difficult to get a good reading as you never know if it has actually reached the target value.

A 330 pF capacitor on the same range is no problem.  It gives a good reading immediately and has the normal update rate of several Hz.  I checked on a couple of other varieties of (less expensive) meters and they had no problem with the small capacitance.  Anyway, I am curious if the 127X series shows this behavior as well.

[EEVblog]

Just got off the phone with Agilent who have gotten back to me after investigating this in their multimeter group in Penang (they took it very seriously).
They have found the problem and confirmed it is the ADC result locking up during certain range switch conditions.
They can fix it in firmware and are doing this now, or have already done so.
They will provide me with a formal email outlying the exact company position on this, so will keep you posted.

Dave.

[saturation]

This is fantastic news.  No previous reviewer caught this issue until it was reviewed on eevblog.  I recall the safety problem you found with a tweezer DMM that was of equal concern.  Likewise, after your Atten 858 hot air reworkstation review, other owners found issues with its build quality and posted on the forum.

It shows the eevblog reviews are more than just entertainment: they have been more thorough whether its by you or your audience.

There are reviews on the 1272a on element14.com that preceded eevblog by 6 months or more, and were trite and elementary regardless of the length of them in words.  Reviews in trade magazines are nearly worthless, they all read like advertising for a product. 

Just got off the phone with Agilent who have gotten back to me after investigating this in their multimeter group in Penang (they took it very seriously).
They have found the problem and confirmed it is the ADC result locking up during certain range switch conditions.
They can fix it in firmware and are doing this now, or have already done so.
They will provide me with a formal email outlying the exact company position on this, so will keep you posted.

Dave.

[PetrosA]

That is great news! I agree that this is a superb place to figure out issues like this and kudos to Agilent for keeping their head in the game. I wouldn't have figured out what was happening with my probe alarm if I hadn't tried to recreate the other on-off-on switch problems that I read about here.

[EEVblog]

Thanks.
Often I don't actually get to play with a product much before a review, so real time spent actually using the instrument can be rather scarce.
In this case I didn't even do a review, just a teardown, and it just happened to play up. Luck, or the EEVblog review curse?  
There is always great follow up stuff from viewers after my reviews, finding and pickup stuff I overlooked. So the whole community is working well toward making better instruments.
Agilent actually commented on this to me today, and as many big companies are, are a bit unsure what to make of the whole new social media thing. But they do love and appreciate the feedback and like the way it's all working to produce better product.
After all, I think we all understand issues can slip out with gear getting more complex these days and design/release cycles getting shorter.
And it's the smart companies that know the ones who are seen to respond and fix things when problems arise, will be the ultimate winners.
Some of the design and management teams from various companies (not just Agilent) do actually read this, and presumably other forums, so they are listening!

Dave.

[saturation]

Yes, that's good to know and I hope it also helps in keeping you self employed now.

With the fixes, the 1272a can indeed challenge the Fluke 87V easily, and after some weeks of added evaluation, its both easy to use ergonomically, and tough.  I'd surmise a users only concern is the digits are less easier to see on the 1272a vs the Fluke 87V, the Fluke has a 'lifetime' warranty vs 3 years on the Agilent, but otherwise, the Agilent is superior as a measuring instrument in all functions that the Fluke supports, and has more.

Just to add, added items were found at eevblog that have made a difference:

sleep power consumption bug in the 80 series III Flukes, that went unreported for almost 10 years
GSM interference bug in the Fluke 87V unreported since the series V was released in 2006

Unique items that were born in the forum and that no forum has been unable to match [ I know, I check!]:

The Rigol 1052e hack and support threads
Hantek/Tekway scope threads

Just the big one's that come to mind.

Here are some samples of the 1272a after a temperature test.  I took it down to 0F then up to 120F, and these are sample images of the DMM warming and cooling down.  RH% exposure to 100% caused the amp alarms to turn on on the 1272a that stopped once all the condensation dried up.  Rechecking the DMM shows all functions well.  During the temp stress the meters are tested for VDC, DCA, and ohms.  Since the meters are often very wet, I don't do any AC tests as it a shock hazard for me.

[Kiriakos-GR]

I can confirm too , what Dave said about Agilent and the U1272A firmware update,
but even if they have everything under control, they need some time to prepare their strategy,
about making public the update.

Some times we forget the size of this companies, and the paper work,
that they need for every single step of action.

Agilent are a bit unsure what to make of the whole new social media thing.
Their confusion are totally justified , they play the game with the rules of the market,
and we ( reviewers and  testers )  some times we exaggerate.

1) Dave started to call the device as piece of Shit.
And kicking it around as to was a football.    

2) Saturation, instead of examining if this product had useful functions,
or if it is practical to use, or if it needs improvements ..
He started to analyze the DNA of the unit, and did blood exams to it,
and at the end I am very hard pressed to believe that he acted wisely so far.  

By evaluating our behaviors, we become better characters,
and if we maintain one positive climate with out exaggerating,
this positive climate it will raise the needed trust,
that those cooperation's or team works needs so to grow.

According to my own information, one evaluation unit shipped today ( Friday 3)  to me,
directly from the factory, with the latest firmware on it.

Even so, I am hopping,  Dave to receive an similar unit even faster than me.
So to finish the review, and restore the negative climate, if the product proved to be now as spotless one.

About Saturation : Now that there is a new firmware, he must get in the line and wait so to load the new one .
The firmware in his samples, today are called as outdated.
( at list you will have the joy to re do all the testing from the start )  

Smack me in the face if you dislike what I have write so far.  
I do not care ..

 

 

[PetrosA]

I don't have any controlled way to test it, but after reading Saturation's post, I suspect the issues I've had with the probe alarms are related to humidity, and that less than 100% will cause the problem. From what Agilent has explained to me, the split plates in the socket are used to sense the presence of a probe in the jack by measuring for a short between the two plates. I speculate that depending on how that's done exactly, if more than one socket is detecting low enough resistance under high humidity levels, that could cause the sensor circuits to go whacky. Whatever the cause, it needs to be ironed out since many regions in the world have regular seasonal high humidity levels and the meter becomes unusable when the alarm is going off.

[Kiriakos-GR]

My opinion is that there is no need for further speculations,
until the arrival and testing of the new firmware.

Your own video and thread , plus the findings of Dave, was known to Agilent.
And hopefully this update will cure all the known bugs.

About the split plates in the socket, lots of DMM are made by the same way and from other brands,
never ever had be reported an issue about humidity problems with this design.
Even the Fluke 87V uses this design !!

[PetrosA]

About the split plates in the socket, lots of DMM are made by the same way and from other brands,
never ever had be reported an issue about humidity problems with this design.
Even the Fluke 87V uses this design !!

Yes, but do they implement the same probe sensor solution?

[Kiriakos-GR]

Yes, but do they implement the same probe sensor solution?

Personally I think that they do.
But due the fact that the U1272A uses an secondary PCB for the probe connections,
its easier to see the traces on the PCB.

Only the Fluke 28II is an exception about the leads sensing circuitry, it uses optical sensors,
because the leads holes they must be waterproof too.

For the Fluke 87 ( all generations ) it is known, that if it enters in leads holes humidity or oils or dirt,
you just have to clean the split plates in the socket, with a bit of cotton and alcohol.

 
 
  
 

[PetrosA]

...you just have to clean the split plates in the socket, with a bit of cotton and alcohol.

More shit to carry on the truck

[toli]

Can someone with one of these meters check something I just discovered on the U1253A?

When measuring on the 10 nanofarad (smallest) capacitance range, and measuring a capacitance below 100 pF, the meter is extremely slow to respond.  The display update rate is like 1/s, but if you zero the reading and then hook up a 33 pF capacitor it will take 10 seconds or more to climb to the appropriate value.  Disconnect the leads and it will take another 10+ seconds to fall back to near zero.  This makes it very difficult to get a good reading as you never know if it has actually reached the target value.

A 330 pF capacitor on the same range is no problem.  It gives a good reading immediately and has the normal update rate of several Hz.  I checked on a couple of other varieties of (less expensive) meters and they had no problem with the small capacitance.  Anyway, I am curious if the 127X series shows this behavior as well.

I did notice the same thing a few days after I got it. I've tested a 33pF cap and it took a long time for it to settle. For 220pF it was way faster. I'm not sure what caused it, but I didn't pay much attention as I almost never use such small caps in my projects. Anyway, posting on the Agilent forum as Kiriakos  suggested could be a good idea.

[firewalker]

After all, I think we all understand issues can slip out with gear getting more complex these days and design/release cycles getting shorter.

Dave.

We should also keep in mind that companies use peoples. Peoples like us, electronics engineers. We are not robots we make mistakes.

[Kiriakos-GR]

...you just have to clean the split plates in the socket, with a bit of cotton and alcohol.

More shit to carry on the truck

Lets hope that you have an exclusive space for the flammable materials.
Like the plastic compartment that you store your comb - cologne - etc etc      

[PetrosA]

Actually I'm thinking of trying a dielectric grease in the jacks. That should keep any moisture to a minimum.