After having some very useful discussions with a few people (special thanks to ComTexeStarter) I came to the conclusion that there was nothing fundamentally wrong with my initial approach – ie using the FTDI board – and that the reason it hadn’t worked was down to software problems.
So I tried again today, the main difference being I was using a new laptop with a clean install of Windows 10. The driver for the serial port was automatically installed, and the Texecom firmware flasher picked it up as a ‘USB-COM’ cable as before. I took a deep breath, pressed ‘Flash’ and…
Success! Everything worked perfectly and the process was complete in about 3 minutes. Following this (as per the manual) I factory reset the panel by removing all power and then powering on whilst holding down the ‘Reset defaults’ button on the main panel PCB.
This caused a bit of a racket with the tamper alarms going off and you then have to enter the default code (1234). Once it settled down, I used Wintex to restore a backup of the configuration data taken a few days ago and the panel is back to normal. You also need to change the firmware version in the configuration screen so that it resets all the new settings to sensible defaults. If you don’t do this you can get some odd behaviour.
ComTexeStarter has been doing some great work on understanding what the programmer PCB does – and the answer seems to be, ‘not much’. Here is my (rather blurry) photo with their annotations (NB this is the correct version, I had previously posted an earlier one with different resistor numbering):
Most of the components and the chip seem to be concerned with feeding 12V to JP2 pin 1 which we think is used to power one of the older Texecom (non-USB) serial interfaces and so is redundant for the USB ones. This means that it should be possible to do the firmware flash with just the UART plus a few resistors to set the panel to programming mode.
ComTexeStarter has worked hard on this and has come up with this schematic which separates off the chip and the 12V supply (in the red box) from what is actually needed to do the programming:
Given that this is now a reverse engineered CDH-0001 this has been christened the 1000-HDC!
The other thing I did whilst I had the panel open was install one of the new ComPort+ boards into the header pin sockets.
You can see it just under mains supply cable, plugged into the header with the (old) firmware version of LS1 v.211.09 on it. It is dead simple in construction:
At a quick glance the pin assignments appear to be (reading the pins from left to right when the board is installed – so pin 1 would be rightmost pin on the above picture:
Pin 1 – NC
Pin 2 – 12V
Pin 3 – GND
Pin 4 – NC
Pin 5 – NC
Pin 6 – NC
Pin 7 – NC
Pin 8 – NC
Pin 9 – RXD
Pin 10 – TXD
So if you don’t want to pay £7 for 5 of these boards (like I did!) and you want an additional serial port, I think you could just connect up your UART to pins 3, 9 and 10 directly. There are a couple of other components – the three pin SMD device is marked Z2X and is I think a Zener diode which is connected between to RXD, and has one leg grounded and the other not connected. The same track feeds goes to an SMD resistor (47k) and then to the 12V line. I don’t really know enough electronics to understand why this might be. However given the RXD pin appears isolated from 12V and connects directly to the header I suspect you could connect directly up to them without a problem. I’d be interested to hear if anyone tries it… meanwhile I have 4 more of these boards to get rid of so let me know if you’d be interested in one.
I haven’t tested Com3 yet but I have another UART on order to test it.