UT71D weird Peak Hold values

[3nigm4]

I'm testing this meter and I have strange Hold Peak values; while on shorted probes I get about 0VDC, with Hold Peak the values are:
- 4VDC scale: -0.170~0.139 V
- 1000VDC scale: -63~32 V
- 1000VAC scale: 0~30 V

Is this normale or I get a faulty meter?

[3nigm4]

More tests made: there is a constant backgroud value.
With probe shorted:
- 1000VDC scale: peak -63 to 31 V
- 400VDC scale: peak -6.3 to 3.1 V
- 40VDC scale: peak -0.63 to 0.31 V
VAC scales changes only for min value that is about the same as the max

I'm thinking it's better to return it

[.rpv]

Hi, I got this values with a 61e:

range----------peak
2.2v............. -0.0493 - 0.0208
22v.............. -0.027 - 0.063
220v............ -0.07 - 0.52
1000v.......... -0.0 - 6.9

[joeqsmith]

Maybe you need to turn off the radio transmitter or try it in a screen room.  Maybe ditch those long leads and try again with a short jumper and see if it makes a difference.

On my UT181E (damaged, repaired then slightly modified) with the inputs shorted in manual range with peak enabled,  allowing 1 minute collection between range settings:

DC ranges PeakMax/PeakMin
0.014 / -0.007
0.01 / 0.00
0.1 / 0.0
1 /  0

DCmV range PeakMax/PeakMin
0.73 / -0.92

AC ranges PeakMax/PeakMin
0.013 / -0.006
0.01 / 0.00
0.1 / 0.0
1 /  -1

ACmV range PeakMax/PeakMin
0.27 / -0.08

[joeqsmith]

My UT61E has some major mods done on it and it may effect the results of this test. 

DC ranges PeakMax/PeakMin
0.0399 / -0.0204
0.016 / -0.007
0.15 / -0.04
1.4 /  -0.2

DCmV range PeakMax/PeakMin
5.70 / -4.34

AC ranges PeakMax/PeakMin
0.0272 / -0.0209
0.015 / -0.006
0.14 / -0.03
1.4 /  -0.3

ACmV range PeakMax/PeakMin
7.55 / -6.87

[3nigm4]

Latest tests (those with -63~31 peak) was made with the short alligator clips.
This mornig I've made some misuring inside with my oscilloscope and found floating voltage on U4 pins (but that is the room temperature reference circuit): about 150 mV 3 Hz oscillation.
I had no time to investigate more but I thing there's something wrong: your values are more acceptable with ±0.3% (rpv UT61E) and ±0.2% (joeqsmith UT181A) tolerance, but mine reach a ±1.6%.

[.rpv]

I'm not expert but, unless you have this offsets on your real readings, this is just a "funny fact", isn't?

[3nigm4]

I think that there's an internal noise that on normal reading is filtered but is captured on peak mode.
In the datasheet of the controller IC is reported a note on peak mode: tracks to MCU must be shielded to avoid interferences. Do they shielded those tracks? Maybe not.

[.rpv]

I think that there's an internal noise that on normal reading is filtered but is captured on peak mode.
In the datasheet of the controller IC is reported a note on peak mode: tracks to MCU must be shielded to avoid interferences. Do they shielded those tracks? Maybe not.

According to Dave's tear down pictures there is shield on the back of the meter and additional shielding for the precision network resistor and some other components, it's for the E version but should be the same for the D.

Have you noticed if your readings are being affected?

[3nigm4]

Have you noticed if your readings are being affected?

The normal readings (for what I test 'till now) are fine... only peak function is affected. But this give me an idea of something not working properly...
I decided to send it back... I'm too picky to accept that a function is not working properly, even if probably I'll never use it

[3nigm4]

Looking at Dave's pictures I tracked the "must be shielded" tracks, as per ES51966 datasheet

Edit: in page 24: "In PEAK mode, the wire of SCLK STATUS EOC must be shielded to prevent
from the noise."

[Wytnucls]

UT-71D
Offset voltages are taken in account in Peak Hold.
Testing AC voltage without any input is meaningless because of the limitations of the RMS converter below 5% of range.

Proper test of Peak Hold for 71D:
AC 231.0V 50Hz  Vpp: 653.0V Quoted AC accuracy 1000V range: 1%+30 (3.26+30=3.56V)
Calculated Peak Hold: 326.50V Measured: 323.00V Difference: -3.50V(measurement is within limits)

There is no fault or flaw with your 71D.

[3nigm4]

UT-71D
Offset voltages are taken in account in Peak Hold.
Testing AC voltage without any input is meaningless because of the limitations of the RMS converter below 5% of range.

My concern is about DC voltages, with crocodiles shorted and 0V displayed on 400V DC range, if I switch to Peak Hold it is always reporting a min -6,3V and a constant max of 3.2V, values that becomes -63V and 32V on 1000V DC range and -0.63V and 0.32V on 40V DC range... those values a repeatable and added to the measured value: i.e. if I measure a 9 V battery in 40V DC range with a normal 8.9V value, I get a constant peak of 9.22V (8.9 + 0.32) with a minimum 8.27V (8.9 - 0.63)...
That is not normal for me... there is something oscillating in the last stage of the input or in the reference circuit

[Wytnucls]

The Peak Hold function is only used to detect fast transients, something that you're unlikely to encounter measuring a 9V battery.
If you suspect a slow fluctuation on a DC source, use the Min/Max function.
If you suspect a fast ripple on a DC source, use the AC+DC function and the Peak Hold function.

For instance: a sine wave 50Hz 1Vrms with a DC offset of 3V, will read correctly as Vmax 4.4V and Vmin 1.6V in Peak Hold mode.

In fact, the 71D has one of the most sensitive (10us) Peak Hold function of all handheld meters:
https://www.eevblog.com/forum/testgear/uni-t-ut71d-review/25/

[3nigm4]

The Peak Hold function is only used to detect fast transients that you're unlikely to encounter measuring a 9V battery.
If you suspect a slow fluctuation on a DC source, use the Min/Max function.
If you suspect a fast ripple on a DC source, use the AC+DC function and the Peak Hold function.

Yes, ok, but with my base noise I'll detect a transient but I'll get a wrong value... so, I may use it only to know there's a transient

For instance: a sine wave 50Hz 1Vrms with a DC offset of 3V, will read correctly as Vmax 4.4V and Vmin 1.6V in Peak Hold mode.

In my case with 0.31V base error in AC I'll get 4.7V as Vmax and 1.9 as Vmin

In fact, the 71D has one of the most sensitive (10us) Peak Hold function of all handheld meters:
https://www.eevblog.com/forum/testgear/uni-t-ut71d-review/25/

Yes, I read all your topics about this meter 

[Wytnucls]

'In my case with 0.31V base error in AC I'll get 4.7V as Vmax and 1.9 as Vmin'

You won't know until you test it with a function generator. I suspect that yours will read exactly the same as mine.

[Wytnucls]

Here is the set-up with the same sine wave, but with an offset of 2.5V:
No base noise effect: 4.00V and 1.16V

My 71D with probes shorted:
- 1000VDC scale: peak -0003 to 0031 V
- 400VDC scale: peak -000.4 to 003.1 V
- 40VDC scale: peak -00.05 to 00.31 V

[3nigm4]

Ok, I understand what you mean

By now, I sent it back using the Amazon return policies... meanwhile I try to see if there's a better alternative with good frequecy counter support... looking to:
- UT171B/C: they use the same HY3131 ot the Agilent, but no maximun frequency counter is reported in the datasheet... in the manual they say 10MHz, while pinsonne-elektronik.de reports 60MHz
- UT181A: it's more complete with ES51997 that supports 60Mhz
- VC890: the same as UT181A
- VC880: use the same ES51966 of UT71x with 400 MHz support
- a cheap UT61E with its 220MHz

I never thought to find hard in searching on a good multimeter with near 100MHz
Brymen are good but stops to 1MHz
Other such Agilent stops to 10MHz

I don't know... if I don't find nothing I may try another UT71... maybe not from chinese seller but from German seller

[Wytnucls]

Keysight U1282A supposedly can measure 100MHz. I don't think there is another one that goes that high.
Accuracy may not be good enough though.

[Wytnucls]

For what it's worth, I tested the 71D up to 20MHz with 1Vpp sine wave and 5MHz with a 1Vpp square wave. The frequency readings are spot on. No errors with a 1KHz resolution.

[joeqsmith]

If your buying a meter to use as a 100MHz counter, seems like the wrong choice.  No input termination selection, no trigger level, inaccurate, poor resolution ....   I had recorded the data I collected from my last UT61E out of the box.  It could read up to 130MHz in the Hz/% setting and up to 1.01MHz in the ACV setting.

[joeqsmith]

I wonder what happens to the meters that are returned like this.   There is no tamper seal so the distributor may not know that the item was taken apart or how it was treated.  Do they just resell them to...   I wonder if my Extech EX540 was in such poor condition because it had been sold to someone first then returned.   I would expect it to at least be marked as used.  The Extech was purchased from a fairly large distributor that many of you buy from.

[3nigm4]

Keysight U1282A supposedly can measure 100MHz. I don't think there is another one that goes that high.
Accuracy may not be good enough though.

Oh, I've read only first part related to both 1280 and 1282 and left that one. Thanks for the info

[3nigm4]

If your buying a meter to use as a 100MHz counter, seems like the wrong choice.  No input termination selection, no trigger level, inaccurate, poor resolution ....   I had recorded the data I collected from my last UT61E out of the box.  It could read up to 130MHz in the Hz/% setting and up to 1.01MHz in the ACV setting.

Yes, you are right.
I'm a developer and I'm writing my own MCU libraries for my home projects and I need a fast and handy way to check the average output speed I get on a 85 MHz MCU... no need for precise readings at that speed 'cause not even my software is precise in outputting data
I prefer good reading and stability but a basic counter too

[3nigm4]

I wonder what happens to the meters that are returned like this.   There is no tamper seal so the distributor may not know that the item was taken apart or how it was treated.  Do they just resell them to...   I wonder if my Extech EX540 was in such poor condition because it had been sold to someone first then returned.   I would expect it to at least be marked as used.  The Extech was purchased from a fairly large distributor that many of you buy from.

This is a good question I made to myself too. Mine is gone back exactly as I received it... and if it was already returne by someone? We will never know I think... but the possibility that some seller use discounts to give away returned items... not all buyers will test all functions like people on this forum