| View previous topic :: View next topic |
| Author |
Message |
soonc
Joined: 03 Dec 2013
Posts: 215
|
| PIC18LF4520 and I2C not working |
 Posted: Wed Jun 18, 2014 10:13 am |
 |
|
Compiler PCWHD V4.134 also used for other code that works.
Chip as per subject line.
In my code using
| Code: |
#use i2c(Master,Slow,sda=PIN_C4,scl=PIN_C3,force_hw) // hardware
output_float(PIN_C3); // pullup 3k3 to Vdd
output_float(PIN_C4); // pullup 3k3 to Vdd
i2c_start();
i2c_write(value); // code hangs here
|
I used this
| Code: | | #use i2c(Master,Slow,sda=PIN_C4,scl=PIN_C3,force_sw) // software |
and then
| Code: |
set_tris_c(0b10000000);
while(1)
{
output_toggle(PIN_C3);
delay_ms(1);
output_toggle(PIN_C4);
delay_ms(1);
} |
and pins C3 and 4 toggle nicely so the uP and connected hardware appear to be working.
With the force_sw I monitored the pins and they do not show any indication of being other than pulled high. i.e. no clk and no data
Are there any special FUSES concerning I2C on this PIC18LF4520 ?
I've looked in the device .h and found nothing.
I2C normally works fine, this code is working on PIC18F67K22 !
This looks like a chip failure but I doubt it as it never is :-)
What other tests can I do or what did I miss ? |
|
 |
PCM programmer
Joined: 06 Sep 2003
Posts: 21708
|
|
| Posted: Wed Jun 18, 2014 10:44 am |
|
|
What is the Vdd voltage on the PIC, and what is part number of the
i2c chip ? |
|
|
temtronic
Joined: 01 Jul 2010
Posts: 9219
Location: Greensville,Ontario
|
|
| Posted: Wed Jun 18, 2014 11:19 am |
|
|
also...
run PCMP's 'I2C scanner ' program in the code library.....!!!
hth
jay |
|
|
soonc
Joined: 03 Dec 2013
Posts: 215
|
|
| Posted: Wed Jun 18, 2014 11:36 am |
|
|
| PCM programmer wrote: | What is the Vdd voltage on the PIC, and what is part number of the
i2c chip ? |
Vdd is 3V3
I2C parts are
LTC6904 Address 0b00101100 (ADR pin = 0) write only chip
LTC6904 Address 0b00101110 (ADR pin = 1) write only chip
AD5245 Address 0b01011010 (AD0 pin = 1) for write
Digital Vdd is applied to all three chips and AD5245 before analog V is applied to the in/out pins of the AD5245 this is mentioned in the data sheet.
The addresses are different. The digital pot is read/write and as I mentioned I'm using known code from anther project (PIC18F67K22) which also runs at Vdd 3V3 and all works OK. |
|
|
soonc
Joined: 03 Dec 2013
Posts: 215
|
|
| Posted: Wed Jun 18, 2014 11:38 am |
|
|
| temtronic wrote: | also...
run PCMP's 'I2C scanner ' program in the code library.....!!!
hth
jay |
Not sure what that is.
I have a logic analyzer that decodes the I2C bus but there is NO activity on the pins so nothing to analyze ! |
|
|
Ttelmah
Joined: 11 Mar 2010
Posts: 19482
|
|
| Posted: Wed Jun 18, 2014 11:42 am |
|
|
What is 'value'?.
It should be one of the chip addresses.
Shouldn't hang things, but might cause oddities if (for instance) it was the GCA address. |
|
|
soonc
Joined: 03 Dec 2013
Posts: 215
|
|
| Posted: Wed Jun 18, 2014 12:19 pm |
|
|
| Ttelmah wrote: | What is 'value'?.
It should be one of the chip addresses.
Shouldn't hang things, but might cause oddities if (for instance) it was the GCA address. |
Value is as shown here.
LTC6904 Address 0b00101100 (ADR pin = 0) write only chip
LTC6904 Address 0b00101110 (ADR pin = 1) write only chip
AD5245 Address 0b01011010 (AD0 pin = 1) for write
AD5245 Address 0b01011011 (AD0 pin = 1) for read |
|
|
PCM programmer
Joined: 06 Sep 2003
Posts: 21708
|
|
| Posted: Wed Jun 18, 2014 12:23 pm |
|
|
| Quote: | With the force_sw I monitored the pins and they do not show any indication of being other than pulled high. i.e. no clk and no data
|
Does it fail with both hardware and software i2c ?
Also, instead of posting code fragments, could you post a small but
complete (compiles without errors) test program that demonstrates
the problem ? There may be things that you're doing (that are not
shown in your posts) that are causing the problem. A test program
will let us see exactly what you are doing. By complete, I mean the
#include for the PIC, #fuses, #use delay(), #defines, variable
declarations, main(), etc. |
|
|
soonc
Joined: 03 Dec 2013
Posts: 215
|
|
| Posted: Thu Jun 19, 2014 2:44 am |
|
|
| PCM programmer wrote: | | Quote: | With the force_sw I monitored the pins and they do not show any indication of being other than pulled high. i.e. no clk and no data
|
Does it fail with both hardware and software i2c ?
Also, instead of posting code fragments, could you post a small but
complete (compiles without errors) test program that demonstrates
the problem ? There may be things that you're doing (that are not
shown in your posts) that are causing the problem. A test program
will let us see exactly what you are doing. By complete, I mean the
#include for the PIC, #fuses, #use delay(), #defines, variable
declarations, main(), etc. |
Yes that's why I'm lost about what is going on.
I really expected the software version of I2C to work but when it failed also is when I tested the pins by toggling them to demonstrate that the circuit was not shorted etc.
UPDATE:
I did a short program and found it actually works. So clearly I'm having issues in the code.
So I'm breaking the code down into sections testing at each step.
Moving code from one processor to another turned out not to be as easy as I thought it should be.
I'll post my findings when I've nailed down what is happening.
Thanks to everyone for the help so far. |
|
|
soonc
Joined: 03 Dec 2013
Posts: 215
|
|
| Posted: Fri Jun 20, 2014 8:05 am |
|
|
The problem was me. On the new PCB I'd connected the 3K3 pull up resistors to the Analog Vdd (convenient in the pcb layout) which is also 3V3 but from a separate regulator.
AD5245 makes a point about applying Vdd before any analog section is powered up....
Anyway I moved the 3K3 pull up to the Digital Vdd and the code now works as expected. Moved it back to old connection and it fails.
After thought. The Test program switched on the Digital and analog Vdd at the same time, the real program the analog came on later.
More investigation than that I don't have time for.
Thanks for everyone that replied. Sorry to have caused a storm in a tea cup but perhaps some else can learn from this mistake.
Thanks |
|
|
|
FAQ 
Search
Register
Profile 
Log in to check your private messages 
Log in
