What would be good is if some large manufacturer were to redesign the package to use a built in Li cell, USB-C charging port, and boost converter, perhaps using a LDO regulator if necessary. But that's a large undertaking.

seems to me you could just rob the circuitry of of one those electronic cigarette things and use it to power the tester. They even have a battery level indicator

Interesting idea! I don't smoke so I never thought about how e-cigarettes work. I just looked it up and they actually pack quite a bit in: a built-in MCU, USB charging, power management, and an airflow sensor. So it could definitely make for a smoking transistor tester!

The main thing to figure out would be hijacking the MCU to sense the tester's button signal. How long does a charge typically last on a normal e-cigarette?

One advantage of this mod is that it leverages the tester's built-in power management, where the battery is completely cut off when idle, so one charge could last months or even years if not used frequently.

Yeah, I dont smoke either but I had friends save them for me when they were done. I have pulled some apart and they are pretty neat. I just had not figured out a good use for them yet. I hate it that people just toss these out anywhere when they finish with them! I have one of those transistor testers and your article came across pretty appealing as I hate 9V batteries!

I do have a few friends who vape, sounds like a good inventory of lithium batteries!

Won't it be better to choose a lower voltage for the boost converter?
As it's followed by the linear regulator U1, I'd choose the lowest output voltage that provides enough margin for U1 to work, maybe around 6.5-7V (you'll need an adjustable one).
Another (maybe better) possibility is using a 5V boost converter, connecting its output to the output of U1 (if the noise in the output of the boost converter is low enough, so the filtering effect of the linear regulator isn't needed)

That's a good point! The tester uses a µA78L05CPKR which has a 1.7V dropout voltage, so 7V should be safe to operate. According to the source code, the tester starts warning at 7.3V and shuts down at 6.3V, so 7.5V might be a better choice to avoid triggering the warning. (https://github.com/kubi48/TransistorTester-source/blob/master/trunk/Battery_check.c)

Directly connecting the boost converter to the MCU could introduce significant noise, which is a concern given how much ADC work the tester does. Replacing the LDO with a lower-dropout alternative would be a cleaner solution, such as the ME6217C50, which has lower output noise and only 0.1V dropout. That said, it would require modifying the firmware to update the warning and shutdown voltage thresholds accordingly. Taking it one step further, reprogramming those thresholds to match a lithium cell's discharge curve would be beneficial by cutting 1 less trace in the first place, since the tester's power-saving circuit would then natively handle lithium voltage levels.

Thanks for the suggestion! I've updated the guide to reflect the voltage improvements, setting the boost converter output to 7.5V provides a safe margin above the 7.3V warning threshold while reducing heat dissipation on the LDO. Great catch! Further modification would require changes to the firmware.

I’ll try to have a look at the source code, I’m sure the starting voltage threshold could be adjusted to work with lithium. Maybe the voltage divider on the ADC input could be changed too, either to improve resolution, or for a hardware-only mod. See schematic here on page 10: https://github.com/Mikrocontroller-net/transistortester/blob/master/Doku/trunk/pdftex/english/ttester.pdf

Wonderful idea! I will prepare a USBasp and probably another LCT-T4 for experiment.

Interesting. I went a different way. I bought one of those lithium cell and boost converter packaged in the 9V battery form factor case that has a USB C charging socket. It will be a swap replacement. Problem is I have to open the case when the battery needs charging. I was considering drilling a hole in the case where the USB C socket is so that I don't have to open the case. Or maybe this doesn't happen often enough to bother. I'm waiting for the current disposable battery to run down first before putting the replacement in.

The USB-C rechargeable 9V packs are a clean drop-in, the main trade-off being the internal boost converter's continuous standby current draw. I actually drilled an ugly hole in it for a USB trigger module.

I tried two approaches before landing on this one. First, using a USB trigger module, it worked but killed the portability. Then I tried a 18650 battery with a TP4056 charging protection module and TPS63070 boost converter. TPS63070 module's quiescent current is around 6 mA or 1 mA with its power-saving feature enabled, which quietly drains the battery over time.

That led me to dig into the schematic more carefully. The LCR-T4 already has a battery-saving circuit built in, so I decided to integrate the solution around that rather than fight against the quiescent current problem. The result is nearly zero drain as the boost converter gets fully cut off just like the original 9V battery would be. The only quiescent current is from DW01A, which is around 4 µA.

Interesting. I'll see what happens with the battery life between charges. I don't mind charging it more often but I don't want any parts outside of the case.

Everything is in the case since portability is important for me. I even hid the boost converter on the bottom of the PCB, and I'm thinking about reusing the ugly hole for charging.

Yup, we agree on portability.