$20 LCR ESR Transistor checker project

[madires]

Is there a tester that implements all those new features?

There's no clone with all hardware options available at the moment. DIY time!

[upsss]

The relay contacts are wired to short the tests pins when the relay isn't powered. When powered the shorts are removed. This way the relay can be switched by changing the MCU pin between input and output mode (low). For the external reference we need the input mode (HiZ) to feed the ADC. The pull-up resistor also turns on the relay via the BJT (removes shorts). In output mode the current (limited by the pull-up resistor) flows to ground via the MCU and the BJT switches off the relay (shorts test pins). And with this design you won't BBQ the ATmega when you connect a charged cap to a switched-off tester.

Hi, originally the base of switching transistor was simply pulled up to the +5V, so once the 5V was on the relay got energized.  That is how my relay protection is still working.  Why was it changed to control the switching transistor with PC4, what was wrong with the original concept?

[madires]

I can't recall that the protection relay was controlled differently in the past (I've checked an old schematic from 2013). But I can add an option to control the relay the way you've described (dedicated pin for the relay). Are the test pins shorted when the relay is energized or when it's not energized? The fail safe solution would be the latter.

[timelessbeing]

The relay contacts are wired to short the tests pins when the relay isn't powered...

Got it. Thank you for clarifying.

Hopefully, the probes are shorted when the tester is switched on, but not in the process of measuring.

[upsss]

I can't recall that the protection relay was controlled differently in the past (I've checked an old schematic from 2013). But I can add an option to control the relay the way you've described (dedicated pin for the relay). Are the test pins shorted when the relay is energized or when it's not energized? The fail safe solution would be the latter.

I guess you are right, I looked at my (older) copy of Karl's manual and the transistor controlling the relay IS driven by PC4.  I cannot remember that I ever changed Karl's original circuit, I just remember that when I built my tester about 3 years ago I simply had the base of the switching transistor pulled directly to the +5V.  When the relay is not energized (no 5V)  the relays contacts (normally closed) shorts the test leads and when you activated the tester, 5V is available and that turns on the switching transistor and the relay opens the shorted test leads. 

When I read your explanation how the protection relay works I simply remembered that my protection relay wasn't driven by PC4 but  my circuit operated exactly as you described.  I guess Karl's circuit was never changed, I simply just activated the relay differently with the same final results.

[timelessbeing]

While we're talking about relays ...

The most sensitive small dip relay that I could find ... will draw 30mA at 5v.   Most other small relays I found, drew more than twice that.

The "must operate" current of my 5V Omron is 20mA. 30mA is actually over spec for it.

you might want to consider connecting the collector to Ubat instead of Vcc as it is possible that the relay load would pull Vcc down during the measurement and significantly impact the measurement accuracy.

Would this be an issue for 250mA MCP1702 regulator with 10uF of output capacitance?  A coil has inductance, so it shouldn't draw huge current spike.

[madires]

I'm using (I think Karl-Heinz too) the 4.5V type from the Fujitsu FTR-B3 series (current 30ish mA). No problems powering it also by the 5V LDO. The boost converter for the Zener check is more demanding and the firmware takes this into account by delaying measurements after pressing the test button. But I'd recommend to power the boost converter by a second LDO. And if you do that you can also use the second LDO for the protection relay.

[timelessbeing]

you have to make sure that the voltage at the connection of both resistors (PC4) is not around 2.5V to prevent the tester from thinking it's an external 2.5V reference (k-firmware).

What is the voltage range that it looks for?

[madires]

The range is 2250-2750mV for both firmwares (k & m).

[joeyjoejoe]

Is there a tester that implements all those new features?

There's no clone with all hardware options available at the moment. DIY time!

I might design a PCB and share it, to be populated by higher quality components.

Any thoughts on the "best" design? AY-AT looks decent with TVS diodes, rotary encoder.

https://github.com/Upcycle-Electronics/AVR-Transistor-Tester/blob/master/AY-AT%20J1.3.pdf

[timelessbeing]

the relay can be switched by changing the MCU pin between input and output mode (low).

So in light of this, I had to redesign my protection circuit.

I used a simulator first.
With voltage reference: http://tinyurl.com/y9hw5smz
Without voltage reference: http://tinyurl.com/ycedlh6g

It turns out you don't even need a resistor on the transistor base. All you need is the pullup (without a v. ref that is)

Then I assembled it on a breadboard and hooked it up to the tester, and tried the two firmwares.

Karl (1.13k)
Worked fine. Just be aware of this: The tester only seems to be protected when turned off. For me, the relay energizes as soon as the tester is powered up, and stays on until you shut it off. That means that that if you are testing components in continuous mode (in other words, one after another without powering off in between), you are vulnerable to BBQing the mcu. Then the relay is essentially pointless.

Markus (1.32m)
Worked great. Also, the tester *only* unprotects the probes during actual measurements. When a measurement is complete, the relay releases and doesn't energize again until you press the button for a new measurement. When you're navigating menus and such, the tester is also protected. This is much nicer in my opinion. 

My only wish is that this feature was documented better. 

There's a known bug when using the discharge/protection relay and a rotary encoder. It's already fixed in the new version under development.

This doesn't seem to be an issue for me.


On to the zener extension ...

[upsss]

It turns out you don't even need a resistor on the transistor base. All you need is the pullup (without a v. ref that is)

To turn on a BJT all you need is to provide enough base current times the transistor current gain to support the collector current.  So of course all you need is one resistor to limit the base current and that resistor can be a simple pull-up, and like I mentioned earlier, I simply have the pull-up resistor going directly to the +5V.

[timelessbeing]

Thanks. I know how a transistor works. I said "it turns out ..." because I didn't know how the mcu was controlling the relay. Neither did you, it seems.

[upsss]

So why were you surprised that "It turns out you don't even need a resistor on the transistor base. All you need is the pullup".  Of course you don't need a base resistor if you have a pull-up resistor to ANY positive voltage when the tester becomes active providing enough base current!  You don't need a silly circuit simulation to figure this out.

[timelessbeing]

Maybe you don't need it, but electronics is just a small pastime of mine.  I find that visualizing a circuit helps me a great deal to wrap my head around how it works. And if it were simply a matter of one transistor, I certainly wouldn't need it, but there are other things going on. You have to watch the voltages and currents on the PC4 pin, the relay etc.  This topic is frequented by beginners, so I posted my findings in case it helps somebody else.

I wouldn't say I was surprised about the reisistor. I just remember madires saying ...

What is the procedure if not using the voltage reference, but relay extension is used? No pull-up resistor?

A pull-up resistor at PC4 (ATmega 328) and the BJT's base resistor should be replaced by a higher value.

... so I was following up about that.

It's too bad this upsets you so much, but hopefully calling it silly made you feel better about yourself. 

[bugi]

While perhaps not quite the same issue as to this particular transistor base resistor case, but, this reminded me of another, old case, where sometimes seemingly "silly" things can after all be required. Beginners (or even somewhat more experienced persons) might not immediately realize it. Testing in practice is then the only way to make sure / to find out. Or asking someone else for opinion/facts.

As an example, an old MCU project. I had put small series resistors on all main "bus" digital I/O pins, connected to another MCU about 10cm away directly (no pull-ups or anything). I.e. 5V AVRs talking to each other, easy stuff. Even usual simulations would not have revealed why I did so... and seemingly they are totally useless resistors, right?  The person converting the schematics to a PCB and assembling the units thought so, too, and decided to leave the resistors out.

The end result was that the MCUs couldn't talk with each other properly. After plenty of poking around, the other person looked at the schematic again, glued the resistors (or were they array chips, don't remember), hand-tweaked the circuits on PCB etc.  And the MCUs finally talked.

In that particular case, those "totally useless resistors" were to handle heavy noise levels... crude, but worked wonders.

(No, it wasn't really any smart idea on my part; I just had read about such possibility and solution some time before, and decided to put them in, just in case, thinking the resistors aren't that expensive, considering that all the other parts were already going over 30€ or something.)

(EDIT: sorry, so off-topic... nothing about LCR meter...)

[lordstein]

today i was updating my transistor tester, change the 4mm conector to allow use the multimeter probes, and wiring and installing a conector for icsp program, i use the tl866a to program my transtor tester, this is the diagram.


this is the wiring



programing

[madires]

Someone has asked how much flash memory the remote command option will need. It's nearly finished and uses a little bit less than 3kB flash and about 250 Bytes EEPROM (just the remote commands without the TTL serial driver).

[KD0CAC John]

I guess multiple post are needed for some questions , they get missing in all the firm-ware posts
Still looking for suggestions on the most up to date kit to buy ?
I have one from a few yrs. ago , I seem to run into lots issue with updates etc. , not a programer etc.
So thought that finding a 2nd one , my 1st had some pads for surface mount components and a rotary switch [ I get the impression that allow for more ] .
Also would like to find one that may have a box to put in , and still have the SMD pads etc.
I've some searches and do not seem to come up with answers - that are not almost like reading from page one .
I gotten replies from others trying to do the same --- no responses ?
Maybe this thread should be called the firmware only thread , no hardware replies

[madires]

Usually you'll find the answer on the last 10 pages. Same question over and over again And the same answer: AY-AT

[timelessbeing]

I guess multiple post are needed for some questions

I guess some people would rather keep tapping you on the shoulder, rather than to do a bit of reading

[KD0CAC John]

192 pages , is not a bit of reading .

I also did some reading , searching , but most of the last couple of yrs. is about firmware .

So if its easier to be an --- than help never mind , hopefully that the internet is not getting that bad ?

[timelessbeing]

 

[madires]

Get your programmer ready!

v1.33m 2018-05
- Fixed orientation of TRIAC symbol in symbols_32x32_hf.h.
- Added remote commands for automation (via TTL serial).
- The x & y offsets for the ST7735 driver can be changed now.
- Entering the menu by a short circuit of the probes is an option now
  (UI_SHORT_CIRCUIT_MENU).
- Fixed problem with discharge relay when using rotary encoder.
- Added configuration switch to disable MCU sleep modes.
- Added RX support to TTL serial driver (bit-bang & hardware USART).
- Fixed error in serial text output and added serial output for results of
  the optocoupler check.
- Danish texts (provided by glenndk@mikrocontroller.net)
- Settings for capacitor correction factors.

[timelessbeing]

Great work! Can't wait to try this out.

This has interesting potential for further development. I envision adding a bluetooth module to the tester, and using a smartphone app with GUI to control it.