[Thanks Bertrand for pointing me here.]
OK. We need to take this one step at a time. (I am the board's co-designer, so I have intimate knowledge of the circuitry and experience with both building and repair.)
Here's where we're at so far, if I've read everything correctly ...
. Hand-built (black) gruvin9x v4.1 board
. Radio was working for a while, testing model set-ups etc. Then went dead.
. Re-flashed firmware, possibly several times. No joy.
. LCD screen was replaced. Still no joy.
. Battery voltage measures some 8.6V, with power on.
. The LCD back-light does not work.
The last item there, about the LCD back-light is perhaps the most telling -- assuming the EEPROM settings are still configured to have the light come on, in the first place. This suggests that the MCU is not running correctly at all, for whatever reason. (Have you tested the LCD back-light alone, using a series 330 ohm resistor and a 5V supply? Not so important for now. But it might offer another clue.)
Besides that, the first thought when fault finding should always be POWER.
So first off, let us check the output voltages of the two main regulators, located above and left of the MCU. The top one should be outputting 5V (+/- 0.05V) and the lower one 3.3V (+/- 0.03V).
Note that the pin layout is different for each of the regulators, as shown. BE CAREFUL not to short out the adjacent pins. The regulators are fairly hardy and will take a little such abuse. But not allowing it to happen in the first place is the best plan.
Please check that and get back to us, here. I have subscribed to this thread to be notified of all replies, now.
I can confirm that the two halves of the radio definitely do not need to be connected, to make the thing go.
Let's talk a little about dry solder joints in the LCD circuit, which is very common and could be your problem, if the power tests above check out ...
I have built three of these boards by hand. Two of the four had dry solder joints around the LCD screen. I have replaced about 34 LCD connectors, most of them in the first batch of boards from China and the rest to upgrade or repair boards I built. The LCD socket is the FIRST place to look for dry solder joints. Everything can look fine, even under a microscope. But if you apply gel flux and run the iron over the pins again, then unseen problems get fixed. This has been the case for me several times -- certainly more than five times.
When testing the connections using a multimeter (preferably in continuity mode, with the beeper function working) it is important to test from pad to pad -- not pin to pin. That's because the pressure of the meter's probe tip pressed down on the pin, which can make it connect to the pad in the case of a dry joint, where in fact it does not connect during normal operation. Every PCB trace -- pad to pad -- connected to anything to do with the LCD screen should be checked, methodically, scientific-style. Generally, if just one single connection is not present, then the LCD will appear to be completely dead.
P.P.P.S: If all else fails, I might be able to send you a v4.2 board, fully built, in exchange for your hand-built dead one. I'll need to check warranty stocks, etc. I'd prefer we go through some basic testing here first though, if it's all the same to you. Thanks.