$20 LCR ESR Transistor checker project

[denimdragon]

OK, so I have been reading through this encyclopedia in bite-sized chunks (far to much to digest at one time) and I still haven't read the entire thread. I have been playing with my AY-AT for the last couple of days. These little devices are really amazing, well actually the coders (Karl & Markus-Both) are amazing. I'm a rookie but I've been doing analog here and there since I've been in school for the last couple of years. This was a perfect chance for me to get into both arenas.

I have a few questions (I'm an extremist, so bare with me...):

• Could this possibly be ported over to a ST32?
• I've been really playing with the idea of adding a daughter board that uses a LT1021BMH-5 for the Vref. How much trouble would that be and would It make a significant difference in this application? I have a ton of these and some aged LM199AH/883QS and LM399AH. I plan on using external power.
• I'm also curious if I could use external ADC as well.

None of this might be worth the effort but I had to ask 

P.S. I know the aim was to max out what was available with minor hardware adjustments so as not to conflict with the software...

[texaspyro]

The AVR only has (rather crappy) 10 bit ADCs.  Not much to gain with better refs and sub-0.1% resistors.

And yes, it could be ported to other processors and be pimped out seven ways to Sunday... but why bother?   The device is a VERY cute, VERY cheap hack,  and will not ever be a metrology grade universal test instrument.  It's operating principles don't allow it.

[madires]

Could this possibly be ported over to a ST32?

Basically yes, but you would need external switching for the 5V test voltage (IIRC the ST32 runs at 3.3V).

I've been really playing with the idea of adding a daughter board that uses a LT1021BMH-5 for the Vref. How much trouble would that be and would It make a significant difference in this application? I have a ton of these and some aged LM199AH/883QS and LM399AH. I plan on using external power.

That would be overdoing things with no real benefit. Anyhow, it can be done.

I'm also curious if I could use external ADC as well.

Of course you can. The drawback is that the circuitry would become more complex and also the firmware. As texaspyro has put so nicely, the tester's main value is its simplicity. KISS and don't over-engineer things

[denimdragon]

The AVR only has (rather crappy) 10 bit ADCs.  Not much to gain with better refs and sub-0.1% resistors.

And yes, it could be ported to other processors and be pimped out seven ways to Sunday... but why bother?   The device is a VERY cute, VERY cheap hack,  and will not ever be a metrology grade universal test instrument.  It's operating principles don't allow it.

As expected, and thanks for the input. Let me ask you, fellow Texan (assumption from your screen name), what concepts and operating principles would I need to keep in mind if I wanted to try something similar to this small form factor with the previously given criteria (granted I know It would be a larger footprint to make a device that's more accurate simply because of noise considerations and shielding/isolation issues)? As I mentioned, I'm an EE student (sophomore), and 7 ways til Sunday is a perfect project agenda for learning. I just don't want to go TOO far down the rabbit whole (my nick name is "Bugs") 

I think it will be fun to give it a shot and see what can be created with similar simplicity, just on steroids (a disclosure was given about my tendency to be extreme).

Basically yes, but you would need external switching for the 5V test voltage (IIRC the ST32 runs at 3.3V). -------> Correct, I considered this also.

I've been really playing with the idea of adding a daughter board that uses a LT1021BMH-5 for the Vref. How much trouble would that be and would It make a significant difference in this application? I have a ton of these and some aged LM199AH/883QS and LM399AH. I plan on using external power.

That would be overdoing things with no real benefit. Anyhow, it can be done.--------> Point taken and noted 

I'm also curious if I could use external ADC as well.

Of course you can. The drawback is that the circuitry would become more complex and also the firmware. As texaspyro has put so nicely, the tester's main value is its simplicity. KISS and don't over-engineer things --------> Yeah, I get it. I do want to attempt to build something like this but with better hardware and based on the STM32 as they are even cheaper than ATmega328 in most cases. The specs on it are far better and it would be fun to see how that chip could be maxed out. The cool thing about this particular project is the flexibility that comes from the ATmega. You guys have done things with the code that's utterly remarkable and fascinating to me. I'm just interested in attempting this with a bit more core horsepower. It's all in the name of learning.

Thank you both for the input 

[texaspyro]

You would need better ADCs (start with ADS1115's?),   better drivers with regulated (and possibly adjustable) output voltage/current (don't use raw GPIO pins),  higher res timers, etc.

[flodo]

Hey Markus,
Thank you very much!

As far as I understood you replaced the following parts on the TC1/TC-1, right?

U4: post
D2: post
C11/12: post

What exact type have you used for D2?


Another question:

The next m-firmware version will include the settings for the TC-1 (in the "Clones" file).

Is there something like a "nightly build" or something like that? The Mikrocontroller SVN trunk-folder is not yours I guess?

I offer any kind of help (like measurements, testing, ...) for sure.

Cheers!

[madires]

As far as I understood you replaced the following parts on the TC1/TC-1, right?

U4: post
D2: post
C11/12: post

What exact type have you used for D2?

The next best Schottky diode I have in stock, an SB160. But I'd recommend to get something like a SS18 which matches the footprint and has a higher reverse voltage. 

Is there something like a "nightly build" or something like that? The Mikrocontroller SVN trunk-folder is not yours I guess?

Nope, I don't release any snapshots of the current source under development. But in some cases I provide a beta version to someone for testing a specific new feature. And the SVN is managed by Karl-Heinz.

[flodo]

Aii, thanks - I see 

You furthermore listed some alternative options to make the TC1 fit the k/m firmware:

- write a new firmware for U4 (simple power_on signal from the ATmega and button press signal for the ATmega)
- replacing U4 with some transistors (standard power circuitry)

Is there a specific reason why you have choosen the "transistor option" instead of a new firmware?


P.S.: Some SO8 to DIP8 adapters:
https://www.cimarrontechnology.com/browndog-031101b-dip8-to-so8-adapter
https://www.gotronik.pl/adapter-dip8-na-so8-p-2885.html

[madires]

The mod with two transistors is very simple and also reduces the quiescent current. Otherwise I would have to get familiar with the other MCU, get a programmer and set up a complete tool chain.

[flodo]

I understand... okay - makes sense Another option you mentioned is reversing the protocol - what information would you need exactly? Why does the firmware not work on the TC-1? Is it just the "button pressed"-signal?

[madires]

U4 doesn't simply emulate the button signal (low when pushed) and seems to use some serial protocol to communicate with the ATmega instead.

[flodo]

Ah yes I see, I just hooked up some logic analyzer as someone already did here: https://www.eevblog.com/forum/testgear/$20-lcr-esr-transistor-checker-project/msg1553443/#msg1553443

Looks like UART at a first glance, will dig a bit deeper and report back

[flodo]

@madires:
Okay, it is not too much communication happening between those two components.

U4, Pin 7 ---> ATmega PD1 / TXD0
U4, Pin 8 ---> ATmega PD2 / RXD1

I attached the communication i sniffed using one of those Saleae clones.
The comm. at a normal starup is built up of three parts:

1) The first one is on PD2 and is somehow optional, it sometimes is there and sometimes is not - don't know why
2) The second one is the one which was already observed in https://www.eevblog.com/forum/testgear/$20-lcr-esr-transistor-checker-project/msg1553443/#msg1553443 (first on PD1 then PD2), it's exactly the same here.
3) As soon as the detection is finished, there is a last bit at PD1

If you long press the button to shutdown the tester, then there is again some sort of comm. on PD1.


I attached the sniffed data in Salea's format, as csv and pictures.

[Gtx21]

It looks like BC557C's have been discontinued?

[Willem52]

I don't think they are discontinued... Look e.g. here:
https://www.aliexpress.com/wholesale?catId=0&initiative_id=SB_20180907001457&SearchText=BC557C

[madires]

If you have difficulties to find a BC557C any jelly bean PNP with an I_C of 100mA or more and an hFE about 400-800 will do the job.

[flodo]

@madires:
Looks like all you need to flash the STC15 is a USB-to-TTL adapter (see e.g. here: http://grauonline.de/wordpress/?p=44).
Furthermore the uC's is 8051 compatible - should be easy to write a new firmware therefore, right?
Maybe we can collaborate somehow? What exactly does the ATmega expect? LOW on a pressed button, right?

[madires]

Of course! The standard firmware expects the button signal to be high when not pushed and low when pushed. It has to stay low as long as the button is pushed. The power-on signal is generated to keep the circuit powered. After the ATmega is powered on the firmware sets the power-on signal high, and when the tester should be powered down the signal is set to low. So U4 would simply sleep and wait for a button press. When the button is pressed it has to enable to boost converter to power the circuit and provide a low signal to the ATmega as long as the button is pressed. At the same time U4 has to check the power-on signal and disable the boost converter when the signal changes to low. Allow some delay for the power-on signal being set to high by the ATmega after enabling the boost converter (startup time of the ATmega).

[seanspotatobusiness]

I assembled my AY-AT clone with a MCP1702-5002 5 V regulator. Since that regulator accepts up to 13.2 V, does that mean I can power my tester with 12 V?

[madires]

Yes, 12V should be OK. But with 9V less power would be dissipated as heat.

[seanspotatobusiness]

If I use two lithium-ion 18650 cells, they would get over-discharged, right? because the regulator will try to take as little as ~5.6 V (2.8 V per cell)? Could I just use a diode with a voltage drop of 0.7 V between the cells and the device so it will stop working when the batteries reach ~3.2 V per cell? Can any of the firmwares detect the input voltage to tell when the batteries are low. It would be cool if it could be told they were a lithium battery and turn off at 6.4 V.

[madires]

You can set the threshold voltages for the battery monitoring. For the m-firmware that would be BAT_WEAK and BAT_LOW in config.h. Also adjust BAT_OFFSET (for the PNP switching power). Both firmwares (k and m) display the battery voltage and switch off when the voltage is too low.

[flodo]

This is the current setup for U4:

Checking target MCU ...
  MCU type: STC15L104W
  F/W version: 7.2.5Q

Current H/W Option:
  . Current system clock source is internal IRC oscillator
  . IRC frequency: 11.051MHz
  . Wakeup Timer frequency: 27.155KHz
  . Do not detect the level of P3.2 and P3.3 next download
  . Power-on reset, use the extra power-on delay
  . RESET pin behaves as I/O pin
  . Reset while detect a Low-Voltage
  . Thresh voltage level of the built-in LVD : 2.42 V
  . Inhibit EEPROM operation under Low-Voltage
  . Hardware do not enable Watch-Dog-Timer
  . Watch-Dog-Timer pre-scalar : 256
  . Watch-Dog-Timer stop count in idle mode
  . Program can modify the Watch-Dog-Timer scalar
  . Do not erase user EEPROM area at next download
  . Do not control 485 at next download
  . Do not check user password next download
  . TXD is independent IO
  . TXD pin as quasi-bidirectional mode after reset
  . P3.3 output HIGH level after reset
  . Testing time: 2017-11-13

  . MCU type: STC15L104W
  F/W version: 7.2.5Q

At least no protection! Going to write a new one now... I will let you know
@madires: Is there anything I need to modify at your firmware? Can you provide me a "ready to flash" hex file - so I can concentrate on U4

[seanspotatobusiness]

The AY-AT clone I bought comes with version 1.12 firmware. Why so old (five years)? I have to wait for my replacement IC to arrive since I want to keep the original one in case I fail every attempt to flash the latest firmware.

[Azure]

This is the current setup for U4:

Checking target MCU ...
  MCU type: STC15L104W
  F/W version: 7.2.5Q

Flodo, great work.  Did you download the current f/w.  Can you post it or PM it to me, I would like to try and look at the code to figure out the data stream it sends and expects to receive from the ATMega. That way we might be able to add routines to the ATMega for the TC1 to work without hardware modification.