UNI-T UT61E Multimeter teardown photos.

[Wytnucls]

Nice job.

Thank you.

Possibly not much of an improvement

Maybe... Maybe not.

With respect, is it any worse than your 71B? How many spark-gaps has yours got?

because of trace spacing and general construction

Again, with respect, has it been determined with absolute certainty that "trace spacing" was the main factor in downgrading the "GS" models to Cat III 300V / Cat II 600V, and not, perhaps, just it being another simple case of legal "cover-your-ass-itis?"

As for the "general construction," you'd be hard-pressed to find a better solution for isolating the battery (esp. compared with the 71 series!). The 10A fuse is also essentially isolated in its own little corner of the board. The thick plastic input jack "buckets" each act as circular isolation slots, and (to my eyes) the blast protection around the case seems to be well implemented (deep U-groove and thick sidewalls).

Am I missing something?

Not looking to start anything (really!), but a little perspective would be nice.

I have a 61E too. Good machine in its price range.
I don't think the 71B I modified offers better protection than yours. It was just a fun thing to do. The only way to tell would be to put them through some high voltage testing, like what the IEC61010 regulations require. For what it's worth, the original 61E didn't survive 2500V on the volt jack, as per Lightages' recent test. The 71 did.
The GS approved UT-61 is a different animal. The few pictures I saw, showed a PCB that was completely reworked to meet specifications.
I use my meters on the bench for low voltage stuff and that's where the 71 excels. I wouldn't recommend it for anything else. If I had to tamper with appliances connected to the mains on a regular basis, I would buy a Fluke for peace of mind.

[em132]

The only way to tell would be to put them through some high voltage testing, like what the IEC61010 regulations require

Anyone check to see if these new regs require the whole meter to be CAT rated to its lowest common denominator? By that I mean, all (?) the Fluke/Amprobe/etc. meters I've seen don't have any voltage rating printed on the current jacks, only the V-ohms (or a single rating which visibly applies to all the jacks). It's assumed the CAT/voltage rating applies to all jacks equally (rightfully so in the case of Fluke, which all have 1,000V HRC fuses, AFAIK). That's not the case with the "regular" (non-GS) UNI-T's, where the amps voltage rating is stated as significantly lower (250VMAX) than the meter's CAT rating (on the V-Ohm jack), because of the low-voltage fuses used. If these new regs require meters to be max-rated no higher than the fuse voltage, then I can see why the GS UNI-T's had their CAT rating down-rated, even if the voltage input may still technically pass CAT III/1,000V.

For what it's worth, the original 61E didn't survive 2500V on the volt jack, as per Lightages' recent test. The 71 did.

Lightages also said "CAT tests do not require that the meter continue to work after the test."

Hi-voltage surges can be unpredictable beasts. You could do 5 of these kinds of tests and end up with different outcomes in each one (from a "still working or not" point of view).

In the end though, if none of them blow up or catch fire, then they pass.

The few pictures I saw, showed a PCB that was completely reworked to meet specifications.

Very likely because those pics you saw were of either the "C" or "D" model.

[Lightages]

From my interpretation, yes any meter must meet the same CAT rating on all terminals at all settings. The meter cannot have different CAT ratings for different terminals. This is from the most recent CAT requirements but earlier versions allowed it appears. To me, the latest rules make sense because the rules are there for safety when things go wrong. You cannot pick and choose what type of "wrong" happens.

The images that Wytnucls was referring to are from the forums at mjlorton's site:
http://mjlorton.com/forum/index.php?topic=140.msg1052#msg1052

And yes they modified the protection and it still had to be down rated. So I can only assume that the non GS version doesn't even meet the GS specs. Who knows what the real CAT rating of the UT61E normal version is? In this case it is best to keep it for use below 250V and probably no more than CATII.

[Monkeh]

They did nothing but fit a GDT and put HRC fuses in. There are no other changes to that meter.

[em132]

From my interpretation, yes any meter must meet the same CAT rating on all terminals at all settings

Okay, so that explains why you won't see separate voltage ratings on Fluke's Amp/mAmp/µAmp jacks (unlike the UNI-T's).

But my question still stands: if under the "GS guidelines" the CAT rating must apply to all terminals, then why assume that the 61E's "down-rating" is due to it failing the CAT III/1000V test on the V-Ohm side when in fact it could simply be due to the fuses being rated at only 600V (or 690V as the case may be)?

And unless I'm missing something, Torrentula's 61E has the exact same trace layout as the non-GS version.

[Wytnucls]

It could have failed either tests or both for the CAT III 1000V rating. (Destructive pulsed 8000V arcing on the volt/ohm ranges and/or 2000V arcing on the amp ranges).
In the end, UNI-T chose to apply for a CAT II 600V CAT III 300V listing, with pulsed 4000V and 1200V testing with the required 600V HRC fuses.
And yes, the first GS 61 I saw was probably a C/D model with a different layout.
On the GS61E, the bottom part of the PCB I looked at recently, is similar to the older version, except perhaps for beefier PTCs. I don't know if the traces have the same copper thickness.

[Lightages]

But my question still stands: if under the "GS guidelines" the CAT rating must apply to all terminals, then why assume that the 61E's "down-rating" is due to it failing the CAT III/1000V test on the V-Ohm side when in fact it could simply be due to the fuses being rated at only 600V (or 690V as the case may be)?

They added bigger PTCs, a spark gap that wasn't there before, 600V rated fuses, and it still was rated lower than the original. The original model has only 250V rated fuses and poorer input protection so it cannot be rated at 600V, only 250V. The CATII is a safety margin over what the GS is rated because the original has less input protection. Simple logic.

But the main thing is to realize that the UT61E's original safety ratings are an outright lie and should not be trusted. We can only guess what its true CAT rating is.

[Wytnucls]

The original PTC is a MWZ11 75HV made by Sinochip. I couldn't find the exact reference, but I think it has a resistance of 1.1K Ohm, a withstanding voltage of about 900V, max current of 300 mA and a switching temperature of 75c.
http://www.sinochip.net/eng/wmz11a.pdf

[metRo_]

There are any cheap case for this DMM that I can buy at ebay? Thanks

[ACvolts]

The input protection circuitry looks a bit whimpy but other than that the build quality looks excellent. I like how they didn't use just one of those ICs covered in a blob of plastic. The red board is also a special touch (getting quite tired of all the green ones).

It looks like you've got a really decent meter.

Dude!.. ahhhh, I think that's orange, not red.   

[Monkeh]

The input protection circuitry looks a bit whimpy but other than that the build quality looks excellent. I like how they didn't use just one of those ICs covered in a blob of plastic. The red board is also a special touch (getting quite tired of all the green ones).

It looks like you've got a really decent meter.

Dude!.. ahhhh, I think that's orange, not red.   

You dug up a post from 2011 to say that?

And no, it's red.

[ACvolts]

Who said that is mandatory that 0.025% DMM's should be rated at least CATIII and have HRC fuses?
Maybe there is a market for less protected meters used in let's say, microcontroller bussiness for less than 5V powered boards.

I don't know about that but, your nickname is interesting, I like IpMan I and II but the 3rd
was so/so.  The original actor split, probably because he got hurt and/or not enough scratch.  %-B

I seen this thread too late and the DMM is up to $72.00 now. 

[Monkeh]

I seen this thread too late and the DMM is up to $72.00 now. 

No, it's not.

[ACvolts]

The input protection circuitry looks a bit whimpy but other than that the build quality looks excellent. I like how they didn't use just one of those ICs covered in a blob of plastic. The red board is also a special touch (getting quite tired of all the green ones).

It looks like you've got a really decent meter.

Dude!.. ahhhh, I think that's orange, not red.   

You dug up a post from 2011 to say that?

And no, it's red.

Nope. heh, still looks like an orange but it's probably the
jpeg that did that.  I'm still reading the old messages!
 

[dr_p]

And no, it's red.

No, it's not.

So funny   It's like people are    and you go "ah...no".

[iloveelectronics]

So funny   It's like people are    and you go "ah...no".

Well, at least he provided a link to my eBay store

Thanks for the plug, Monkeh!

[Monkeh]

Thanks for the plug, Monkeh!

Thanks for the plugs, Franky!

[T4P]

Thanks for the plug, Monkeh!

Thanks for the plugs, Franky!

*cough* I see what you did there

[dr_p]

I'm pleasantly surprised by the UT61E. Among the obvious specs:

  - feel more expensive than 60$
  - when on diode range, it still beeps if you short the probes, like in continuity mode. Switch-mode power supply repairs got easier.
  - discharges capacitors under test
  - easy hacks available for Auto-Power-Off and Backlight

[UPz]

Hi guys 
This is my first post here!

I'm having a bit of trouble with my UT61E.

Using the UT61E user interface program V4.01 (UT61E.exe), and loging data to the PC in sample intervals (15 sec for example), I end with time inconsistent results in the log file. Sometimes it jumps one second, sometimes it jumps 9 minutes of log.

I have tested with a third party software, called UltraDMM (v1.0.3), and it has the same behaviour.
I've tested without the sample interval option checked and somehow I got solid two measurements per second...

Two screenshots:





I'm using the RS232C cable, and my PC runs on Win7.
Have you any idea of what can be happening?

Thanks in advance!

[Bored@Work]

Hi guys 
This is my first post here!

I'm having a bit of trouble with my UT61E.

Using the UT61E user interface program V4.01 (UT61E.exe), and loging data to the PC in sample intervals (15 sec for example), I end with time inconsistent results in the log file. Sometimes it jumps one second, sometimes it jumps 9 minutes of log.

Since the UT61E can't be programmed, i.e. the data communication is one-way UTE61e --> PC only, and since it therefore always sends with a fixed sample interval, which the PC software can't change, the reason for the occasional 1 second jumps seem obvious. Any sample interval that isn't a multiple of the UT61E's native data rate (probably 2 sec.) will sooner or later end up with such a jump, due to accumulating errors when the software tries to map the time of received data to the nearest time for the intended sample rate. Even a sample rate set to a multiple of the native sample rate will sooner or later have such a jump, depending how the UT61E's clock rate and the Windows timer differ.

Nine minutes of data loss, however, experienced with two different pieces of software, means there is something wrong outside of the applications. E.e. the driver, Windows, hardware ...

Anyhow, you could try yet another software, http://www-user.tu-chemnitz.de/~heha/hs_freeware/UNI-T/dmm.zip Don't be surprised that the software is so small, the guy who did that software doesn't believe in .NET rubbish. Maybe that software does what you want.

[Rick Law]

Hi guys 
This is my first post here!

I'm having a bit of trouble with my UT61E.

Using the UT61E user interface program V4.01 (UT61E.exe), and loging data to the PC in sample intervals (15 sec for example), I end with time inconsistent results in the log file. Sometimes it jumps one second, sometimes it jumps 9 minutes of log.

Since the UT61E can't be programmed, i.e. the data communication is one-way UTE61e --> PC only, and since it therefore always sends with a fixed sample interval, which the PC software can't change, the reason for the occasional 1 second jumps seem obvious. Any sample interval that isn't a multiple of the UT61E's native data rate (probably 2 sec.) will sooner or later end up with such a jump, due to accumulating errors when the software tries to map the time of received data to the nearest time for the intended sample rate. Even a sample rate set to a multiple of the native sample rate will sooner or later have such a jump, depending how the UT61E's clock rate and the Windows timer differ.

Nine minutes of data loss, however, experienced with two different pieces of software, means there is something wrong outside of the applications. E.e. the driver, Windows, hardware ...

Anyhow, you could try yet another software, http://www-user.tu-chemnitz.de/~heha/hs_freeware/UNI-T/dmm.zip Don't be surprised that the software is so small, the guy who did that software doesn't believe in .NET rubbish. Maybe that software does what you want.

I have done a good bit of logging since I got my UT-61E around New Year time.  I can confirm that Bored@Work is spot on for both counts.

I deal with the +-1 tick by adding a "calculated time" column in the spreadsheet, and made that
=CellAbove + N Seconds  (where N is my sampling interval)
Of course, the first row has no CellAbove, you have to copy-paste the start time onto that cell.

You will find that a few rows down, it will get out of sync (calculated time != time stamp) and then in a few more rows it will get back in sync, then later, it will be out again and back again.  The time delta could be 2 clock ticks: Mathematics saids when you truncate and round, you have a +-1 on each.  When PC and UT both have a +-1, your max error is 2.

I just add that extra column and plot the graph based on that extra.

* * *
As to the 9 minute lost...

After encountering the same thing, I ignore it at first, then I saw it attempting to download new anti-virus file while logging data and not having a good time doing it.  So, now when I am logging, I disable WiFi and hard-IP the NIC with no gateway and no DNS.  Hard-IP means type it in yourself.  These auto-updates are deadly.  Flash, JAVA, whatever.  With the NIC having no DNS and no gateway, I can talk to it inside (so I can monitor progress remotely) inside the house, but it has no official way to talk to the outside world.  It has not happened since.

If you are not sure what IP to use, turn your machine on normally and do a:
ipconfig /all
It will tell you what IP you are using.  You can use the same IP and just erase the gateway and DNS entry.  That IP assigned to your machine will stay with your machine for (typical) 24 hours).  Without gateway and DNS, it will disable most normal network connections to the outside world.  Hackers may hard-code their target IP (no need for DNS) and try various gateways (typical setup makes it easy to find).  Those you will need a firewall setting to stop which is hard, or just pull the darn RJ45 plug which is easy.  Make sure you set it back when done.  The hacker's communication target IP is likely not their own IP but a zombie machine.

Rick

[T4P]

You seem to be rather knowledgeable on the topic of network security ... studied network security? 

[Rick Law]

You seem to be rather knowledgeable on the topic of network security ... studied network security? 

If you are referring to me, the answer is NO and YES.

No, I do not know a lot of about network security.  Network is something that cannot be secured 100%.  The more I know, the less secure I feel.

Yes, I studied enough to know the basics. On multiple occasions, I was either responsible for the operation of a network for a group or responsible for network for an enterprise (>100 nodes, <5000 nodes) - and of course, my wife appointed me the official "tech support" at home.

[hgg]

Hi, question for the Uni-T UT61E.

When you switch on to Volts the display will initialize and show 0.000V but when you switch on
and have the blue button pressed the display will show 0.0000V until you let go the blue button.

Does anybody know if there is a hidden mode or some hidden function?
For example in the UT71E if you do the same, the multimeter will start with one digit less resolution,
but with a very fast refresh.   

Thanks.