Uni-T launches the UT117C multimeter

[sonpul]

.3% on a 60000 count meter?  That's just silly.  +/-180 counts? 

As they say, it's written on the fence. Reality is different.
I see that UT117C has 0.02%-0.03% accuracy.

[mwb1100]

As they say, it's written on the fence.

Sorry for going off topic, but I don't recognize that idiom.  Is that a literal translation of an idiom from another language?

[ledtester]

As they say, it's written on the fence.

Sorry for going off topic, but I don't recognize that idiom.  Is that a literal translation of an idiom from another language?

I was curious too so I did some investigating. Given that the poster was from Ukraine I had Google translate "it's written on the fence" into Ukrainian: "На паркані написано".

Doing a web search for that phrase resulted in a lot of hits so it appeared that it was indeed a common expression / idiom. Translating some of those search results gave me a sense of what it meant but nothing definitive.

Finally I decided to ask Google Bard (i.e. Google's ChatGPT) for an explanation of the idiom and it responded with:

Звичайно. Український ідіом "На паркані написано" означає, що щось дуже очевидне і не потребує пояснення. Він часто використовується, щоб виразити своє розчарування з кимось, хто не розуміє чогось, що, як вам здається, очевидно. Наприклад, якщо ви говорите з кимось про важливу подію, яка недавно відбулася, а вони просять вас пояснити, що сталося, ви можете сказати: "На паркані написано!" Це означає, що інформація вже доступна, і немає необхідності пояснювати її ще раз.

Ідіом "На паркані написано" також може використовуватися в більш жартівливому ключі. Наприклад, якщо ви бачите когось, хто робить щось дурне, ви можете сказати: "На паркані написано!" Це означає, що людина повинна була знати краще, і немає потреби пояснювати їй, що вона робить неправильно.

Ідіом "На паркані написано" - це корисний спосіб висловити своє розчарування або жарт з кимось, хто не розуміє чогось, що, як вам здається, очевидно.

which translated is:

Of course. The Ukrainian idiom "It's written on the fence" means that something is very obvious and needs no explanation. It is often used to express your frustration with someone who doesn't understand something you think is obvious. For example, if you're talking to someone about an important event that happened recently, and they ask you to explain what happened, you might say, "It's written on the fence!" This means that the information is already available and there is no need to explain it again.

The idiom "It's written on the fence" can also be used in a more humorous way. For example, if you see someone doing something stupid, you might say, "It's written on the fence!" This means that the person should have known better, and there is no need to explain to them what they are doing wrong.

The idiom "It's written on the fence" is a useful way to express your frustration or joke with someone who doesn't understand something you think is obvious.

Perhaps native speakers can comment on the accuracy of this explanation.

[CosteC]

.3% on a 60000 count meter?  That's just silly.  +/-180 counts? 

As they say, it's written on the fence. Reality is different.
I see that UT117C has 0.02%-0.03% accuracy.

NO.
UT117C has 0.3% DCV accuracy. https://meters.uni-trend.com/product/ut117c/
It is stupid high resolution, 0.3% DCV meter. Well build, with couple of nice features like BT. Extra digit or two does not make it more accurate.
Short term accuracy and staiblity is far better than 0.3% - but same is true for any decent meter.
Possibly Uni-Trend is planning other model in same box with increased accuracy, even on "the same" hardware. Thier sweet secret however will be how they select and age references or exact tolerances and quality of components.

[sonpul]

As they say, it's written on the fence.

Sorry for going off topic, but I don't recognize that idiom.  Is that a literal translation of an idiom from another language?

Sorry for initiating an offtopic. The phrase “It is written on the fence” says that you can write anything without any responsibility and correspondence to reality.

There is also a vulgar definition. When hooligans write in large letters on the fence the word for the sexual organ, and behind the fence they will not find either the sexual organ or sex, even if you believe it.  “***** is written on the fence, and there is ordinary firewood”

[2N3055]

As they say, it's written on the fence.

Sorry for going off topic, but I don't recognize that idiom.  Is that a literal translation of an idiom from another language?

I was curious too so I did some investigating. Given that the poster was from Ukraine I had Google translate "it's written on the fence" into Ukrainian: "На паркані написано".

Doing a web search for that phrase resulted in a lot of hits so it appeared that it was indeed a common expression / idiom. Translating some of those search results gave me a sense of what it meant but nothing definitive.

Finally I decided to ask Google Bard (i.e. Google's ChatGPT) for an explanation of the idiom and it responded with:

Звичайно. Український ідіом "На паркані написано" означає, що щось дуже очевидне і не потребує пояснення. Він часто використовується, щоб виразити своє розчарування з кимось, хто не розуміє чогось, що, як вам здається, очевидно. Наприклад, якщо ви говорите з кимось про важливу подію, яка недавно відбулася, а вони просять вас пояснити, що сталося, ви можете сказати: "На паркані написано!" Це означає, що інформація вже доступна, і немає необхідності пояснювати її ще раз.

Ідіом "На паркані написано" також може використовуватися в більш жартівливому ключі. Наприклад, якщо ви бачите когось, хто робить щось дурне, ви можете сказати: "На паркані написано!" Це означає, що людина повинна була знати краще, і немає потреби пояснювати їй, що вона робить неправильно.

Ідіом "На паркані написано" - це корисний спосіб висловити своє розчарування або жарт з кимось, хто не розуміє чогось, що, як вам здається, очевидно.

which translated is:

Of course. The Ukrainian idiom "It's written on the fence" means that something is very obvious and needs no explanation. It is often used to express your frustration with someone who doesn't understand something you think is obvious. For example, if you're talking to someone about an important event that happened recently, and they ask you to explain what happened, you might say, "It's written on the fence!" This means that the information is already available and there is no need to explain it again.

The idiom "It's written on the fence" can also be used in a more humorous way. For example, if you see someone doing something stupid, you might say, "It's written on the fence!" This means that the person should have known better, and there is no need to explain to them what they are doing wrong.

The idiom "It's written on the fence" is a useful way to express your frustration or joke with someone who doesn't understand something you think is obvious.

Perhaps native speakers can comment on the accuracy of this explanation.

Meaning of that phrase in this context is "I don't care what they wrote I see meter doing much better than that and that is empirical proof instead of some irrelevant writing on the paper", as in "The proof of the pudding is in the eating"...

@sonpul insists that despite published 0.3% accuracy, meter is actually performing better and that because of that performance he considers it to be in 0.03-0.02% class.

Which is not the way it works. It might be (and usually is) that some product perform much better that spec. But fact that manufacturer specifies very low accuracy of 0.3% means that they deserve right to ship meters as bad as that. It could be that they devised very fast calibration fixture that can calibrate thousands of meters an hour but has wide variability or drifts during day... Even if you guarantee only 3 sigma variation in calibration there will be some perfectly calibrated meters in a population... And some as bad as spec allows.

Or they are just reserving right to put any quality components in there, or meter has large tempco or long term drift, so even after all this it will still perform inside specification.

Best way to guarantee that you will deliver up to spec is lowering expectations...
And you never know which one you're getting.

My Metrix and Brymens are almost an order of magnitude better than specification on some ranges and at the room temp.....And if the planets align.. But I would not trust them for more than specified for serious purposes.

That being said, 0.3% is good enough for most handheld use scenarios.

[sonpul]

2N3055, Totally agree with you. It makes no sense to require an accuracy of 0.03%, but it can be obtained with a high probability. Now the price for this multimeter is about $70. And if the accuracy that we see in the reviews continues, then this is a very good choice.

[alm]

It is stupid high resolution, 0.3% DCV meter. Well build, with couple of nice features like BT. Extra digit or two does not make it more accurate.

Short term accuracy and staiblity is far better than 0.3% - but same is true for any decent meter.

That's true of course, but not everything is about accuracy. Say you are adjusting a trimmer for reading closest to zero. Or you want to see if a value is going up or down with ambient temperature. For that you need resolution, short term stability and linearity, but not accuracy. Quite a lot of measurements are like that. Matching resistors for a differential amplifier for example. Actual value doesn't matter much, as long as they are equal.

Of course more accuracy is better and it's to be proven if the short term stability is there or if the extra digits are just noise.

Possibly Uni-Trend is planning other model in same box with increased accuracy, even on "the same" hardware. Thier sweet secret however will be how they select and age references or exact tolerances and quality of components.

That would make sense to me. Could indeed be selection or just the accuracy of adjustment.

[kloetpatra]

Can you test it on say, rectangular waveform instead of sine?

Yes, but I am not sure how well the Keithley will perform as a reference. Also I don't want to consider crest factor errors.

Don't forget that this is a simple comparison of the readings of 2 multimeters, while the Keithley seems reputable enough and frequencies low enough (at the that point) to stick with kloetpatra's statement that the uni-pee is good up to 1kHz (mV) and 10kHz (V) (sine), we don't know enough about the test setup nor have a known reference to say that one meter or the other is reading high or low into the MHz...

True! Unfortunately I do not have any calibration standard. I only have two Keithley 2000 and one calibrated Keithley 2015. The Keithleys should be in spec up to 300 kHz. As AC source (sine) I used the 2015 source output up to 20 kHz (Accuracy: ±(0.3% of setting + 5mV)). For higher frequencies I had to use a Rigol DS1074Z source output which is specified as ±(2% of the setting value + 1 mV), but also seems to do better than specified according to the readings.  Due to relocation I only had access to banana leads so higher frequencies can not be tested. Anyway the UT117C AC V reading drops to zero for frequencies > 1.5 MHz.

kloetpatraHow do you create such charts?

The chart was made quick and dirty using anychart javscript library https://playground.anychart.com/h4rynMDs

I also did some DCV mesaurement comparison of the Keithleys using a AD584KH precision reference.
The shown deviation is in regard to the AD584KH chinese factory calibration values written on the back.

[sonpul]

It seems to me that soon the dissatisfied will write a petition to the manufacturer so that they spoil their Uni-T UT117C to the 0.3% declared in the specifications. 

[ledtester]

Mine just arrived -- order was placed on the 12th so technically it took 15 days.

The package was wrapped in Uni-T tape so it stands to reason it came directly from Uni-T -- i.e. the Aliexpress seller simply placed a drop-ship order.

My bundle of the UT117C + clamp meter also came with an oversized mouse pad -- not sure what I'm going to use it for.

[makz]

My ut117c have a precision above 0.1% (compare with philips pm2525). Resistance readings also overestimate on precise resistors (+ 0.1%)

[makz]

comparison measurements of devices

[BILLPOD]

Mine just arrived -- order was placed on the 12th so technically it took 15 days.

     I am also in the USA and ordered mine on Aug 2nd, so should get mine in 15 days?.....I hope

[geb]

Hi! Just joined the forum.

I received my UT117C from Aliexpress a couple of weeks ago. Played with it before leaving for a vacation.

I backed up its 24C02 EEPROM. I tested it against an AD584KH, a pile of 0.01% resistors, a VREF-01, my handmade 1-5-10V reference, and some low-cost reference gadgets (e.g. LB06). When testing voltages, I didn't trust the voltage reference, and instead compared the UT117C's reading to that of my Agilent 34401.

To make a long story short, it seems to be uniformly inaccurate. I mean, voltage ranges are reading roughly 6 digits high and resistance ranges are about 6 digits low.

I want to figure out how to access its calibration menu. Unless someone has already solved that topic and can offer a quick reply, I'll eventually ask that question in a separate thread. I found an unusual menu by holding buttons during power-on but didn't go further.

I've calibrated a small stack of meters using my materials, and been pretty pleased with the result. My ZT-102 is basically perfect (and unlocked to 9999 count using AN8008 EEPROM's limit settings). One of my RM303s is quite good (and its EEPROM stores individual calibrations for each resistance range, unusual for a 12P65-based meter). My Owon B41T+ had the worst rated accuracy but it is a newer revision with a multiturn pot for DC and its DCV is good after a lot of tweaking.

[Electro Fan]

Hi! Just joined the forum.

I received my UT117C from Aliexpress a couple of weeks ago. Played with it before leaving for a vacation.

I backed up its 24C02 EEPROM. I tested it against an AD584KH, a pile of 0.01% resistors, a VREF-01, my handmade 1-5-10V reference, and some low-cost reference gadgets (e.g. LB06). When testing voltages, I didn't trust the voltage reference, and instead compared the UT117C's reading to that of my Agilent 34401.

To make a long story short, it seems to be uniformly inaccurate. I mean, voltage ranges are reading roughly 6 digits high and resistance ranges are about 6 digits low.

I want to figure out how to access its calibration menu. Unless someone has already solved that topic and can offer a quick reply, I'll eventually ask that question in a separate thread. I found an unusual menu by holding buttons during power-on but didn't go further.

I've calibrated a small stack of meters using my materials, and been pretty pleased with the result. My ZT-102 is basically perfect (and unlocked to 9999 count using AN8008 EEPROM's limit settings). One of my RM303s is quite good (and its EEPROM stores individual calibrations for each resistance range, unusual for a 12P65-based meter). My Owon B41T+ had the worst rated accuracy but it is a newer revision with a multiturn pot for DC and its DCV is good after a lot of tweaking.

Wow
Nice first post 
Welcome to the forum!
Looking forward to seeing your new thread(s).

[geb]

Thanks! I'd borrowed an occasional tip from the forums to use in my work up to now. But it was last year, when I found myself single (widowed) and bored, and I idly wondered whether there was anything interesting I should know about my ZT-102. One thing led to another... I've spent about USD1500 at eBay and AliExpress and DigiKey so far. It's been educational, and I'm a bit less bored. 

[makz]

So, how to calibrate ut117c?

[geb]

So, how to calibrate ut117c?

That *is* the question.

I can't find my note at my workbench about which keypresses launched the hidden menu. And the menu merely shows "0" through "9", which advances by a keypress.

I could swear that I'd reached a sign of success once. I pressed some sequence and reached a DCV range, though I'd started the meter in ACV. Never reproduced that since.

The meter has a separate MCU driving Hycon chips. If it was 100% Hycon chips like the HP-770D I last played with, I'd expect a calibration experience like ZT-102 and RM-303 and so on, because the Hycon MCUs have very little space and their firmware never seems to vary much. But with the separate MCU, anything goes.

[makz]

this combianation of sel+rel+ multimeter on. But what this 0 is mean 

[LeonR]

The backlight on those is a pretty poor (couple LEDs around main LCD). Does yours behave the same?

[makz]

Hi! Just joined the forum.

I received my UT117C from Aliexpress a couple of weeks ago. Played with it before leaving for a vacation.

I backed up its 24C02 EEPROM. I tested it against an AD584KH, a pile of 0.01% resistors, a VREF-01, my handmade 1-5-10V reference, and some low-cost reference gadgets (e.g. LB06). When testing voltages, I didn't trust the voltage reference, and instead compared the UT117C's reading to that of my Agilent 34401.

To make a long story short, it seems to be uniformly inaccurate. I mean, voltage ranges are reading roughly 6 digits high and resistance ranges are about 6 digits low.

I want to figure out how to access its calibration menu. Unless someone has already solved that topic and can offer a quick reply, I'll eventually ask that question in a separate thread. I found an unusual menu by holding buttons during power-on but didn't go further.

I've calibrated a small stack of meters using my materials, and been pretty pleased with the result. My ZT-102 is basically perfect (and unlocked to 9999 count using AN8008 EEPROM's limit settings). One of my RM303s is quite good (and its EEPROM stores individual calibrations for each resistance range, unusual for a 12P65-based meter). My Owon B41T+ had the worst rated accuracy but it is a newer revision with a multiturn pot for DC and its DCV is good after a lot of tweaking.

what error did you end up with in comparison?

[geb]

LeonR, agreed, the UT117C's backlight isn't great. I reached for my nearest DMM, an HP-770D, for a side-by-side check. I'd guess the UT117C had about 60-70% of the lumens of the other.

makz, SEL+REL sounds right. As for comparison, I'm not sure what you're asking.

Out of the box, all the DCV and resistance ranges I tested on my UT117C are about six digits off. This is oddly consistent and made me think they copied the EEPROM from another unit without recalibrating .

As for some of my other calibrated meters...

The RM303 is rated for better accuracy (0.05%+-3 in DCV) and is now within a digit or two on multiple DCV and resistance ranges after a brief warmup. It's the best resistance accuracy I've seen in a cheap meter so far. I modded the voltage reference but it was nearly as good before this. (Changed ICL8069D to a handpicked ICL8069A with low drift and added a 4.7uF cap across it per datasheet recommendation. Was hoping to get rid of its wobble, 10.000V is 9.999-10.001 every second or two, but the mod only helped a little. Paralleled caps in a few spots also didn't help. Didn't want to try adding a VLCD supply as another modder did.)

The B41T+ is only accurate to 1-2 digits on a short range of DCV. Its linearity as tested across its 2.2-22V range seems a bit off (several digits IIRC) and some other modes don't have their own calibration (no EEPROM, just a few pots). The B41T+ is a good performing meter otherwise, with a fast-responding analog bar graph and fast auto-ranging.

Incidentally, the bar graph on the UT117C is responsive but "noisy", at least during a quick check of around 5.0VDC.

[makz]

LeonR, agreed, the UT117C's backlight isn't great. I reached for my nearest DMM, an HP-770D, for a side-by-side check. I'd guess the UT117C had about 60-70% of the lumens of the other.

makz, SEL+REL sounds right. As for comparison, I'm not sure what you're asking.

Out of the box, all the DCV and resistance ranges I tested on my UT117C are about six digits off. This is oddly consistent and made me think they copied the EEPROM from another unit without recalibrating .

six digits off - 5.0006 instead of 5V? I understand correctly?

[EEVblog]

BT is nice too.

I'm disappointed that Brymen didn't release the Bluetooth option in the BM786 
It was supposed to be ready for release but it never happened.