Fried LPC17 output

[Highcooley]

Hey everybody

As it seems, I fried the R/W output pin of my LPC17 (p1.27 of the LPC1769 xpresso).

When I moved my Synth from the cardboard housing over to the new wood case, I accidentally plugged the 5V supply into J27 next to J2, the actual supply pins.

The 800mA slow blow fuse blew instantly, so I assumed, that I had shorted 5V to GND of J27 but everything except for the fuse  was alright. Of course, the move from one housing to the other lead to a lot of cables being too short and a bit of debugging due to shorts with the front panel and so on and so forth. The last thing to do, was to solder a longer cable to the KS0108 compatible GLCD, which at least for me is always a hassle to get the specific pinout as well as pullups, pulldowns and trimmerpots correct. After the first power on, the LCD started up normally with the init text, but then things got bad very quickly. After a couple of seconds, it continuously started looking more like a snow storm with random pixels being on or off, firstly changing, whenever the LCD was updated, but becoming static after a while. Meanwhile, with each power cycle, I either get a blank LCD or the mentioned snow storm. Of course, I doubted my skills first, checked the wiring several times, also changing RST high and low, but nothing helped. For a short time, I also believed that the 45cm long cable was too long, but the internet taught me otherwise. So I was left with two possible root causes, either the LCD has to be broken or the LPC. With the built in testlcdpin function, I was able to find R/W always being high (not mattering if an LCD is connected or not). This leads to my conclusion, that I probably fried the output of the LPC, since the pin is directly connected to the micro controller on the lpcxpresso board.

Is there a possibility to assign another pin as the LCD R/W pin in a config file, so I can cut and rerout it to P1.27 on the lpcxpresso board and still be compatible with future updates? If this is not an option, does anybody know if the newly released LPC1769 board is fully compatible with the old version: http://www.mouser.ch/ProductDetail/Embedded-Artists/EAX00242/?qs=sGAEpiMZZMtWuggIubTlfyQKvxcc1sy3C%2fHLDQO4Us4%3d ?

 

Thanks for your help! I am a bit in a hurry to fix this quickly, since I have to give up my electronics lab as soon as possible to make room for our firstborn :-) :-( :-)

Cheers,

Andy

Edited February 14, 2017 by Highcooley

[latigid on]

Damn! The SEQ was looking so nice! Congrats on the new arrival though!

I'm sure there are better people to ask than me, but I'l give it a shot.

In mios32_board.c

 

#if defined(MIOS32_BOARD_MBHP_CORE_LPC17) || defined(MIOS32_BOARD_LPCXPRESSO)

#define J15_AVAILABLE 1

#define J15_SCLK_PORT      1
#define J15_SCLK_PIN       25

#define J15_RCLK_PORT      1
#define J15_RCLK_PIN       26

#define J15_SER_PORT       1        // also used as DC (data/command select) for serial interfaces
#define J15_SER_PIN        28

#define J15_E1_PORT        1        // also used to control SCLK of serial interfaces
#define J15_E1_PIN         29

#define J15_E2_PORT        3
#define J15_E2_PIN         25

#define J15_RW_PORT        1        // also used to control data output of serial interfaces
#define J15_RW_PIN         27

#define J15_D7_PORT        1
#define J15_D7_PIN         19

 

you can see the pin mappings. I think you'd change these MIOS definitions, and every time a new firmware update came out, you'd have to build your own hex for your hardware.

The question is then what pin can be swapped in without conflicting with any of the SEQ functions, and is it special function? Some pins are set as timers for critical processes; I don't know if the R/W pin is critical or not. J10 looks like it's free for remapping? Of course, you may have to initialise these pins first -- it's probably not as simple as substituting the port and pin in the definitions...

 

In the case of buying a new LPC board, I think the newer ones have a different layout, so they probably won't fit into the Core headers :(

 

 

 

 

[Highcooley]

Thank you for the compliment as well as your sympathies for the SEQ. Rest assured, the SEQ is all fine and dandy. It is my second project (a full fledged analog synth, based on Midibox NG) I am talking about, which also got a nice looking case. Pictures will follow, as soon as I get the display running.

In my case, the two J5s as well as the J4s would be unused, as I'm running a SCS at J10. 

Also thank you for pointing me into the right direction with the pin mapping file. It is not very appealing to build own explicit hex files of course. Especially, since I am more the hardware guy than the software specialist. But as a last resort, I am definitely looking into that. 

 

Concerning the new LPC board, Embedded Artists state, that:

Quote

the expansion pins (2x27 positions) are mechanically compatible with the original design

If I compare the schematics, the pinout is also exactly the same. There is only an additional RBG LED onboard, which will flash wildly in different colors, since it is driven by three DOs used for other purposes in the midibox standard. So I tossed my money in and ordered a new LPC. We will see, if it works properly and if all goes well, I'll have a spare LPC for a display less midibox application :-)

 

Cheers, Andy

[Highcooley]

Just a quick followup:

It was indeed a fried pin. After replacing the LPCxPresso board (Rev D) and going through the horror of figuring out how to get the bootloader onto that darn thing (as well as bending all of the header pins on the right side of the board in order to finally ram the displaced row into the headers of the midibox board), my LCD went back to operate perfectly.

In case you need to set up a Rev D LPC17 as well one day, better find this post instead of figuring out the steps by yourself: