OK, time to post some pictures.
First picture is the PCB out of the CS2.
Slap in the middle of the PCB is the main processor that runs the Linux code. It is of the Intel PXA250 family, and appears to be the highest speed of three possible speed options.
Immediately to the left of that is the boot rom - easily identified as such because it says so on the top of it. Immediately above that is the NAND Flash chip that acts as the internal 'disk drive' for the Linux system, and to the right of that above the processor chip is an FPGA that I suspect does the NAND Flash interface, but I may be wrong on this, I haven't checked through the processor data sheet to see if it natively supports NAND Flash chips.
To the left of the boot ROM is two RAM chips, and above them are pads to fit another two chips. Below these are the two white connectors accessible underneath the CS2 to connect 6015 series boosters and the s88 devices.
To the 'south east' of the processor chip is the cable that goes through to the display. This carries the data that is displayed, there is another cable for the touch screen and another for the screen backlight. Below the white cable there is an 8051 family microprocessor. I believe this is the TPU chip, as a similar chip is used in the add-on board for the CS1 that allows connection of 6015/7 series boosters.
Above and to the right of this chip is the power supply section. There are various regulators to get the voltages required by the processors and various other chips. The two very large capacitors on the very right hand end of the board are the main filter capacitors, and the rectifiers that allow use of an AC transformer are at the very top right. Immediately to the left of the rectifiers is the power connector, with the track and programming track connectors the next two to the left. Below this are various filter components which I presume were designed to originally meet FCC regulations but apparently are not enough to do so seeing later cs2s have an external filter supplied.
Below these components is a large capacitor on its side and then the chip that is the cause of alan's problem. A burn mark is visible in this photo, but I have taken a close up to show some detail.
The result of incorrectly connecting a power supply to the output terminals can be clearly seen. The solder has disappeared off the blackened pads and the split in the casing where the 'magic smoke' escaped can be clearly seen. The transformer on the far side of the resistors beside the large capacitor is used to pick off the mfx signal from whatever is replying out on the track to an mfx information request. The mfx decoder chip is the larger of the two chips on the right hand edge of the picture.
The chip that has blown up is a standard ST Microelectronics L6206 power driver, designed for driving motors and similar loads, which makes it an ideal device for this use. However the fun and gmes is going to be getting the chip off, but I think it s worth a try at replacing it as a replacement chip is less than £6 form Farnell or RS. We have the necessary tools to get the chip off despite it having a heat sink pad underneath that is soldered to the board, and then fitting a replacement chip after cleaning up the burnt pads should be a breeze after that.
I think it is worth a try to repair it at that cost as the chips are readily available. If that doesn't work I haven't laid out too much money if it then needs to go back to Marklin.
The replacement chip should be here later this week, so I will continue the saga then.
