I bought one of those under $10 lcr meters. Mine lights up with the logo Mtester.
Over the years I have accumulated a dozen resistors all marked at 1/10 of 1% accuracy so I tested them with my new meter. I hope that I will be able to attach a table of my measurements to this post—I'm not really familiar with such additions.

Anyway nothing comes close to 0.1% accuracy. There appears to be no adjustment on the meter. Do I just chalk this up to $10 junk?

 

Do I just chalk this up to $10 junk?

You're expecting too much from a $10 meter. Did you subtract the resistance of the leads from your measurements?

Does it work better for measuring voltage?

 

Hello,

What is the full marking on the resistors?
There must be a band for tolerance.
Depending on the E-series, the tolerance can be 1, 2, 5 or even 10 %.



Bertus

 

In the first post he stated they were 0.1% resistors.

I am not really surprised that it is off by that much. The voltage reference it is using is likely no better than 1% and then there is the ADC error which will depend on how many values of current source it uses. If it is using decades and a 10-bit ADC that would introduce another 1% error.

Bob

 

Hello,

The 1.6 could be E24 or E192.
The E24 is at least 2 % tolerance:



Bertus

 

Hello,

The 1.6 could be E24 or E192.
The E24 is at least 2 % tolerance:

View attachment 178850

Bertus

These are marked in numbers including the accuracy--and priced as such.

 

Hello,

Here I found a desciption of the Mtester:
https://tkkrlab.nl/wiki/MTester

It does mention nothing about accuracy.

Bertus

 

Hello,

Here I found a desciption of the Mtester:
https://tkkrlab.nl/wiki/MTester

It does mention nothing about accuracy.

Bertus

Thanks for the link. It looks exactly like mine except my display does not show a version number and does not have the signature. It is an atm 238 which is an 8 bit microprocessor.

One of the things that puzzles me is that there is no consistent error--not always one way or the other; not always as far off.
Also, I'm dubious about the use of a ZIF connector--surely they are not made for hundreds of insertions.

 

I know I will offend the OP. I am reading his posts in this thread and they don't seem to be genuine.
- They present data that they collected and only they can verify.
- They mention atm 238. Even simpleton like me knows it is ATmega 328. Dyslexia? Maybe.
- They call it microprocessor. It is not. It is microcontroller. The reason it is used in this product is because it has peripherals that microprocessor DOES NOT have.

In conclusion. It feels like an attention seeking exercise. Makes no difference to me. It just does not feel genuine.

 

Did the meter come with specifications? Was it ±1%, 5%, 10%? How was the specification determined? If you average the percentage errors, you get something like 0.00083 percent. (That is not how I would do it, but who knows for the off-shore vendor?)

Now, for the real test, you might try the $20 meter and see if it is half as bad.

 

Thanks for the link. It looks exactly like mine except my display does not show a version number and does not have the signature. It is an atm 238 which is an 8 bit microprocessor.

One of the things that puzzles me is that there is no consistent error--not always one way or the other; not always as far off.
Also, I'm dubious about the use of a ZIF connector--surely they are not made for hundreds of insertions.

Most likely, the way it measures resistance is to send a constant current through the resistor and read the voltage across it with the ADC of the microcontroller. It will use several ranges of current.

Say it different ranges a decade apart. So it would use, maybe 100uA 1mA, 10mA 100mA until it gets a voltage reading that is at least 1/10th of the ADC max, which is likely 1024. So if the resistor just hits 105 at 10mA say, they cannot up it to 100 mA and your measurement resolution is at best 1 / 105 or about 1%. If the resistor read 1020 then it has a much higher resolution of 0.1%. The accuracy will be lower by at least 1/2 a bit.

Bob

 

I'm dubious about the use of a ZIF connector--surely they are not made for hundreds of insertions.

I wouldn't worry to much about that. I do a lot of eeprom work and am consistently inserting in and out of zif sockets daily with no issue's (accept for the few I've melted because of being lazy and hot air'd the eeprom onto an already inserted adapter). A few of my zif eeprom readers are pretty old as well.

 

I know I will offend the OP. I am reading his posts in this thread and they don't seem to be genuine.
- They present data that they collected and only they can verify.
- They mention atm 238. Even simpleton like me knows it is ATmega 328. Dyslexia? Maybe.
- They call it microprocessor. It is not. It is microcontroller. The reason it is used in this product is because it has peripherals that microprocessor DOES NOT have.

In conclusion. It feels like an attention seeking exercise. Makes no difference to me. It just does not feel genuine.

Well, I'm not offended, just a novice trying to make my way around some pretty complicated stuff.
Yes it is the ATmega 328. And yes, it is a micro controller. I have never used a micro controller except if one is part of something I bought--like the meter. And probably never will, though I did pretty well with Fortran 4-- 50 or 60 years ago.
As for posting data only I have--well, of course. It is my meter and I made the measurements.
Now, since the controller is limited to 8 bits I realize it can't be perfect. But I would certainly expect some pattern to the error measurements

 

Now, since the controller is limited to 8 bits I realize it can't be perfect. But I would certainly expect some pattern to the error measurements

The fact that the microcontroller is limited to 8 bits doesn’t imply that your measurements are also limited to 8 bits. The Arduino ADC on an 8 bit processor still is a 10 bit ADC.

A long datatype on an ATMega328 in an Arduino is 32 bits.

I suspect that error in your measurements is not due to an 8 bit microprocessor.

 

Thank you--I did not know that.
If any reader of this thread has a similar meter, please measure a known value resistor or two--1% is fine--and let me know what you get.

 

The Atmega328 has an internal band gap reference that can be very accurate and resistant to temperature drift but it would require a calibration which is likely skipped because it would require writing this to EEPROM. Who can tell what they did when they made this. I would check against another meter. Might still be useful for instance to check and see if a component works. Maybe not much for accuracy.

 

To be able to measure that accuracy, you need to use a set of Kelvin leads. I suspect that your meter is doing the best with what it has. It you are going to be measuring resistance values to 0.01%, then you're going to need a new, much more expensive meter.

Just exactly what did you expect?

 

The Atmega328 has an internal band gap reference that can be very accurate and resistant to temperature drift but it would require a calibration which is likely skipped because it would require writing this to EEPROM. Who can tell what they did when they made this. I would check against another meter. Might still be useful for instance to check and see if a component works. Maybe not much for accuracy.

Actually, the Band Gap Reference Voltage is not very accurate.
The 1.1 volt Band Gap Reference can vary by +/- 10%, between devices.

 

Actually, the Band Gap Reference Voltage is not very accurate.
The 1.1 volt Band Gap Reference can vary by +/- 10%, between devices.

True between chips it can vary from 1.0 to 1.2V... this is why I mentioned the calibration. With calibration the error is 1mV (0.1%).

 

To be able to measure that accuracy, you need to use a set of Kelvin leads. I suspect that your meter is doing the best with what it has. It you are going to be measuring resistance values to 0.01%, then you're going to need a new, much more expensive meter.

Just exactly what did you expect?

I did not make myself clear. I do not expect 0.01% accuracy or anywhere near it; I used precision resistors in my test so that I would know that any error was not the resistor but the meter. I was hoping for something better than a DVM would give me.
As to Kelvin leads, totally unnecessary at my expected level. There are, actually no leads as such--just a ZIF socket into which the resistor leads are plugged.
An internet search of Ardino ohmmeters was enlightening, if not encouraging. One seemingly competent study indicated worthless readings below about 100 ohms and over about 1000 (I've forgotten the exact range.)
Maybe I'll buy another meter to check that I did not just get a dud. Maybe I'll just pitch it.