EEVblog #1020 - Is A $7 LCR / Component Tester Any Good?

[boffin]

Dave tests the M328 (LCR-T4) model component tester.

[boffin]

And Dave.  "M" is "madires" on the EEVBlog forum...

I've just got my PCBs back for spinning my own serial version of it (I'll be working on adapting the software in the next few weeks)

[Brumby]

A much better assessment!  Well done Dave.

They aren't the be all and end all - but they are definitely worth the few dollars they cost!  (Even if the SMD pads are awkward to use.)

[bjcuizon]

I was actually just looking forward to buying one from aliexpress. Here it is:
https://www.aliexpress.com/item/Digital-ESR-Meter-Mega328-Transistor-Tester-Diode-Triode-Capacitance-MOS-PNP-NPN-LCR-TESTER-METER/32776851185.html?spm=a2g0s.13010208.99999999.271.XyX0bD
It says below its page that the maximum measured value for the ohms measurement is 50M . How come the one Dave has just measures up to 10M ? Is it firmware incapability?

*By the way, I'm not affiliated with that store above.

[Brumby]

There are dozens of different models of this out in the market - and this was shown in the video.  They are going to be differentiated by varying quality, capability and price.

The one Dave received was, it seems, one of the lesser models.  With the price for such a unit, you can't argue much.

.... and I am hard pressed to remember the last time I wanted to check a resistor over 10M.

[b_force]

Most interesting is what a standard atmega328p (or similar) can do.
Are there any known projects that use some better hardware (like ADC/DAC etc)?
That way you can take it easy to another level, for just a few pennies extra.

[madires]

And Dave.  "M" is "madires" on the EEVBlog forum...

The firmware of that clone is a modified k-firmware with a fancy output. The "M" in "MTester" is just a coincidence.

[madires]

WIth a current firmware the Transistor Tester supports:
- L/C/Rs
- special mode for low value L/Cs (k-firmware)
- diodes
- BJTs/MOSFETs/JFETs
- SCRs/Triacs (if they trigger with 5V and up to about 5mA)
- IGBTs
- PUTs/UJTs
- opto couplers
- rotary encoders
- PWM generator
- squarewave generator
- IR RC decoder (m-firmware)
- servo tester (m-firmware)

And there are several hardware options:
- Zener check up to about 30-40V
- frequency counter
- rotary encoder for user interface
- touch screen for user interface (m-firmware)

And I'm sure I missed some.

[madires]

Most interesting is what a standard atmega328p (or similar) can do.
Are there any known projects that use some better hardware (like ADC/DAC etc)?
That way you can take it easy to another level, for just a few pennies extra.

One of the key factors is that the MCU runs with 5V.

[b_force]

Most interesting is what a standard atmega328p (or similar) can do.
Are there any known projects that use some better hardware (like ADC/DAC etc)?
That way you can take it easy to another level, for just a few pennies extra.

One of the key factors is that the MCU runs with 5V.

Explain? 
Key factor of/with what?

[madires]

Key factors of the design WIth a 3.3V MCU you wouldn't be able to check as much components as with 5V without an external switching circuitry.

[RGB255_0_0]

Seems you had your Snickers today Dave 

[Brumby]

Leave Dave alone.  He can't know everything (but missing that thread was a bit of a fail.)

He owned his ignorance - and proceeded to tackle the subject head on.  He has more than redeemed himself - in my eyes at least.

[b_force]

Key factors of the design WIth a 3.3V MCU you wouldn't be able to check as much components as with 5V without an external switching circuitry.

Still don't see the point?

Plus you can easily use a lower voltage ucontroller with proper level-shifting and other techniques.

[madires]

Exactly! WIth a 3.3V MCU we would have to add several external components to be able to switch each probe pin and probe resistor between HiZ, Gnd and 5V. The current design with the 5V ATmega is very simple, another key factor.

[b_force]

Exactly! WIth a 3.3V MCU we would have to add several external components to be able to switch each probe pin and probe resistor between HiZ, Gnd and 5V. The current design with the 5V ATmega is very simple, another key factor.

Ok, I guess. 
Don't really see why that would be an huge 'key factor' to be very honest.
Handful of components would maybe cost you a few cents, plus you get additional robust/sturdy I/O ports.
It's nice if 5V works out, but also not an huge deal-breaker if you can't IMO.

[Kleinstein]

The 5 V µC allows a super simple HW with direct interface to the µC - so no extra level shifter / amplifier needed.  5 V are just enough to turn on most MOSFETs and LEDs. A 3.3 V µC would need amplification and thus considerably more HW, as it would be 3 terminals working as input and output for both high and low currents. With a 3.3 V µC one would even consider the more traditional way with switch matrix and than dedicated drive and measure lines.

AFAIK the SW is more or less open source with extra functions added incrementally. In part it's rather tricky to measure very large an small resistors with such simple HW - even getting a ballpark figure is not simple.
With this rather good SW the cheap Chinese HW appeared - usually without giving credit to the original source.

However the super simple HW is also one of the big downsides: there is a chance to damage the µC with a charged cap or maybe ESD.

The capacitance is not tested at 1 KHz - it is more like a charge / discharge with the given resistors in the circuit.

The measured capacitance on MOSFETs might be high, because it is measured at low voltage. Data-sheet values are usually for a higher voltage.

The only thing disappointing is the lack of instructions.

[madires]

Yep, it would be nice if the clone sellers would add a note with the project's URL for documentation and firmware updates.

[Cerebus]

Yep, it would be nice if the clone sellers would add a note with the project's URL for documentation and firmware updates.

Which is?

[Brumby]

Yep, it would be nice if the clone sellers would add a note with the project's URL for documentation and firmware updates.

Which is?

Good point.

I think it would be good to include that in this thread.  I've found a lot of EEVblog gets indexed on Google - and people looking for some info could easily end up here.

[madires]

Here you go:

Project: https://www.mikrocontroller.net/articles/AVR_Transistortester
k-firmware: https://github.com/svn2github/transistortester/tree/master/Software/trunk
m-firmware: https://github.com/svn2github/transistortester/tree/master/Software/Markus (read the README!)
documentation: https://github.com/svn2github/transistortester/blob/master/Doku/trunk/pdftex/english/ttester.pdf

[b_force]

However the super simple HW is also one of the big downsides: there is a chance to damage the µC with a charged cap or maybe ESD.

The measured capacitance on MOSFETs might be high, because it is measured at low voltage. Data-sheet values are usually for a higher voltage.

Those are a few things I was referring to.
5V is the bare minimum.

Don't get me wrong btw, I am not saying it's crap, on the contrary.
I just don't see why 5V is "a key point".

That being said, I think it wouldn't be so difficult to lift this project to the next level.
With a few addons that means you can actually get very reliable results.
In my opinion the software is pretty decent in that side, but the HW could just use a few more bits.
That way you can easily come up with something that's being close or even equal to professional gear for just fraction of the price.
But maybe some people like to keep it simple and am I to ambitious?
I just see potential that's not fully being used.

[firewalker]

If I am not mistaken there is calibration for this? Shorting the three probes and starting it? I could make more accurate. Mine is spot on compared with two flukes DMMs.

Alexander.

[Kleinstein]

Building a more accurate / capable version would likely mean changing the concept. With higher voltage one can no more use the digital IOs with tristate. So one would need kind of external multiplexers (e.g. 4 channels 3:1 mux) to connect measurement functions to the terminals. With dedicated channels for ADC and driving one could have higher resolution ADC / DAC or current source. However the software would likely need to be rewritten in many aspects. The software might be similar, but the HW would be considerably more parts. The next step up would be something like 5 more chips.

[b_force]

Building a more accurate / capable version would likely mean changing the concept. With higher voltage one can no more use the digital IOs with tristate. So one would need kind of external multiplexers (e.g. 4 channels 3:1 mux) to connect measurement functions to the terminals. With dedicated channels for ADC and driving one could have higher resolution ADC / DAC or current source. However the software would likely need to be rewritten in many aspects. The software might be similar, but the HW would be considerably more parts. The next step up would be something like 5 more chips.

I don't think you would need to use an higher voltage on all aspects.
But I see your point, although I doubt if you would 'completely' change the concept.

It's nevertheless doable, and I think worth a shot.

[dardosordi]

Thanks Dave for doing this, it was great!

[D3f1ant]

When's that pretty new eevblog meter going on sale?

[djos]

What an awesome little tool, just bought one with plastic shell for $15! 

http://www.ebay.com.au/itm/LCR-T4-M328-ESR-Meter-Transistor-Tester-Diode-Triode-Capacitance-SCR-Inductance/112259065141?ssPageName=STRK%3AMEBIDX%3AIT&var=412673841412&_trksid=p2060353.m2749.l2649

[T3sl4co1l]

How the heck did this get screwed up?



Didn't catch it?  Look at the diode!

Which further proves my favorite rant about drawing MOSFET symbols.  The diode's already there, in the middle!  You're just making it worse by drawing in the other redundant one!

Tim

[mauroh]

What an awesome little tool, just bought one with plastic shell for $15! 

My 2 cents:

There are so many hardware variants based on the exact same core schematic and before everyone rush to buy the variant that Dave reviewed, I think it is better to evaluate which will better fit your needs.

The main difference I considered are:

Thru hole            vs   smt components        If you forget to discharge a capacitor before testing it the THT version can be fixed quickly
Assembled          vs   DIY Kit                      a DIY kit shouldn't be an issue in this forum
With push button vs  with rotary encoder    whith the encoder it is easier to navigate in the menu to select specific measurements options (opto, IR, Servo...)
With case            vs  without case              I like the case one

For instance my personal preference is the one that is called Hiland DIY M12864 Kit at Banggood, because it is THT, it has a nice case and it is faster than the color display variant.
I modifyed it to have:
External input voltage
ICSP connector
UART for measurement logging also with Bluetooth....

Mauro

[Urs42]

I have a variant that is powerd by aa sized lithium cell that can be recharged over a USB port:

https://www.aliexpress.com/item/OOTDTY-12864-LCD-Mega328-Transistor-Tester-Capacitance-ESR-Meter-Diode-Triode-MOS-LCR/32825169218.html

You can find some nice cases for this model on thingiverse.

[Barny]

OK. Looks like the prices are mystically are ramping up.
Could it be because Dave talked good about it? 

[HKJ]

I like the large display on the one Dave reviewed, but I do not like loose PCB lying around.

My current favorite is this one:


They are fine for identifying components, but the Peak's are better (and much more expensive).

[Oskar]

Interesting info here:

https://github.com/svn2github/transistortester/blob/master/Doku/tags/english/ttester_eng112k.pdf

[Fungus]

I like the large display on the one Dave reviewed, but I do not like loose PCB lying around.

My current favorite is this one:

Does it have the same graphics as Dave's or just text?

[glarsson]

The TC1 has a Graphic color/colour LCD. Powered by internal LiIon battery and charged via USB. Also decodes many IR remote control protocols.

[HKJ]

Does it have the same graphics as Dave's or just text?

Graphic:

I did a review of it: http://lygte-info.dk/review/ComponentTester%20Multifunction%20Tester%20T1%20UK.html

[Fungus]

Does it have the same graphics as Dave's or just text?

Graphic:

Color! Oooh, shiny!

[JPortici]

There are dozens of different models of this out in the market - and this was shown in the video.  They are going to be differentiated by varying quality, capability and price.

The one Dave received was, it seems, one of the lesser models.  With the price for such a unit, you can't argue much.

.... and I am hard pressed to remember the last time I wanted to check a resistor over 10M.

I was making order in the lab at the workplace! argh! all the components i had to sort... many of them were about 20 years old. there even were 10 Meg resistors. brown black blue???

[bjcuizon]

Does it have the same graphics as Dave's or just text?

Graphic:

I did a review of it: http://lygte-info.dk/review/ComponentTester%20Multifunction%20Tester%20T1%20UK.html

Ooo..looks classy as. Its also a good idea to have colour as it can help you to identify the pinout quickly (like which coloured pin on the screen is connected to which coloured lead) because your not just looking at numbers.

[joeqsmith]

I assume you dropped it to 7V because the mux.  That's a tough problem to solve.  Personally, I would like to see the 15V for testing high voltage diodes. 

[floobydust]

Out of MCU codespace on these, the ATmega328 is 32KB and pretty much maxed out.
Some builds you have to use different fonts or watch the compiler version or else the binary is too big.

Look forward to next gen, something with better than 10-bit A/D and lots of FLASH for graphic LCD.

[nomy]

Hi,

Is it possible to to connect it with the PC to get serial data?

Thanks

[madires]

The k-firmware supports output via a serial interface (software UART).

[mauroh]

Yes, the k-firmware supports the serial output, but this means that you need to enable the UART output in the makefile, compile it and program the Atmega328. None of this testers on the market are sold with the uart enabled AFAIK.

There are 2 option to enable the UART (documented in the makefile)
CFLAGS += -DWITH_UART
and
CFLAGS += -DWITH_UART=2

The second option adds the line number at the beginning of each line.
This is very useful and simplify a lot the program on the PC/Mobile side
I've written a super simple interface with Processing that replicate the tester output on the PC

See also my previous post in this page, where I connected the enabled UART output to a mobile phone with a Bluetooth module.
Have fun
Mauro

[FearTec]

Mine arrived from eBay broken