MRF90XAM9A interfacing help

[Tabsoft]

Perhaps you need to strip this back to the basics?
Forget trying to configure the MRF registers.
Just try to read them after a POR.
Also, what is your physical connectivity between the PIC and the MRF (Pinouts/Schematic)?
Can you show us your aliases for MRFConfigSel, SCLK, SDO and SDI?
I can't think of any reason why you shouldn't be able to read the config registers.
Using Shiftin/Shiftout for SPI has not been a problem thus far.


Create just a skeleton program to configure the PIC for the connectivity you need to the MRF and Serial port for debug.
Then with the skeleton, just wait for the appropriate amount of time and try to read regs 00 - 07, but do not do this in a For/Next loop. Read each register in a serial fashion.

E.g.
MRFaddr = $40
MRFConfigSel = 0
SHIFTOUT SDO, SCLK, 1,[MRFaddr]
SHIFTIN SDI, SCLK, 0,[MRFSPIdata]
MRFConfigSel = 1
debug "MRF Config data:Register ", bin8 MRFaddr, " Data=", hex2 MRFSPIdata,13
pause 50
MRFaddr = $41
MRFConfigSel = 0
SHIFTOUT SDO, SCLK, 1,[MRFaddr]
SHIFTIN SDI, SCLK, 0,[MRFSPIdata]
MRFConfigSel = 1
debug "MRF Config data:Register ", bin8 MRFaddr, " Data=", hex2 MRFSPIdata,13
pause 50

etc...

[ecoli-557]

Thanks for the continued interest - I will share code with all when the darn thing works <grin>.
the IO assignments below:

        MRFReset        var PORTA.1                                             'MRF module reset
        MRFDataSel      var PORTA.4                                             'MRF module serial interface data chip select, active LOW
        MRFConfigSel    var PORTA.5                                             'MRF module serial interface configure chip select, active LOW
        MRFIrq0         var PORTB.0                                             'MRF module interupt 0
        MRFIrq1         var PORTC.2                                             'MRF module interupt 1

I have the time to work on this now (at home) so I will try just readinga register back - no loops.

[ecoli-557]

OK, I have added the base read-back as you suggested, code below:

'MRF initialization here
debug "@ MRF init", 10
            MRFConfigSel=0                                                      'Select the chip       
'Going to try using shiftout before using hardware SPI
            for i=0 to 31                                                       'Sets up the index var for data
                MRFaddr=(i << 1)                                                'Shifts address 1 bit to left, automatically gets start,write, and stop bits      
                SHIFTOUT SDO, SCLK, 1,[MRFaddr,RF_Init_Values(i)]               'send data to register
debug "Addr=",bin8 MRFaddr,"  Value=",hex2 RF_Init_Values[i],13
            next i    

            MRFConfigSel=1                                                      'Deselect the chip


'Try to read back some registers as a test to see if they were written correctly            
            MRFaddr=$40
            MRFConfigSel=0                                                      'Select the chip 
            SHIFTOUT SDO, SCLK, 1,[MRFaddr]                                'Address to read  
            SHIFTIN SDI, SCLK, 0,[MRFSPIdata]                              'Get data from reg
            MRFConfigSel=1                                                      'Deselect the chip
debug "MRF Config data:  Register ", bin8 MRFaddr, "   Data=", hex2 MRFSPIdata,13
            pause 50
            MRFaddr=$41
            MRFConfigSel=0                                                      'Select the chip 
            SHIFTOUT SDO, SCLK, 1,[MRFaddr]                                'Address to read  
            SHIFTIN SDI, SCLK, 0,[MRFSPIdata]                              'Get data from reg
            MRFConfigSel=1                                                      'Deselect the chip
debug "MRF Config data:  Register ", bin8 MRFaddr, "   Data=", hex2 MRFSPIdata,13

Still doesn't read back. Debug data below:

@ MRF init
Addr=00000000  Value=28
Addr=00000010  Value=8C
Addr=00000100  Value=01
Addr=00000110  Value=63
Addr=00001000  Value=0C
Addr=00001010  Value=03
Addr=00001100  Value=77
Addr=00001110  Value=64
Addr=00010000  Value=32
Addr=00010010  Value=74
Addr=00010100  Value=62
Addr=00010110  Value=32
Addr=00011000  Value=38
Addr=00011010  Value=00
Addr=00011100  Value=19
Addr=00011110  Value=00
Addr=00100000  Value=A3
Addr=00100010  Value=38
Addr=00100100  Value=30
Addr=00100110  Value=00
Addr=00101000  Value=00
Addr=00101010  Value=00
Addr=00101100  Value=53
Addr=00101110  Value=59
Addr=00110000  Value=44
Addr=00110010  Value=00
Addr=00110100  Value=70
Addr=00110110  Value=BC
Addr=00111000  Value=02
Addr=00111010  Value=01
Addr=00111100  Value=5E
Addr=00111110  Value=00
MRF Config data:  Register 01000000   Data=00

MRF Config data:  Register 01000001   Data=00

Still getting all zeros......... what am I missing?
BTW, where in Texas are you?

[ecoli-557]

I do have a question, in the timing diagram below:



It looks to me as if there is a dummy byte before the real data, am I confusing myself?

[Tabsoft]

I guess I don't see what you're referring to "dummy byte before the real data".
Where specifically are you speaking of?

[ecoli-557]

I've got it. And I am dumb guy. In my design for a wireless sensor for the house, I have the 3.3 rail turned OFF when it is charging - I thought it might be a good idea not to turn on the transmitter while I was re-charging it........
Well, while I am building/designing this, I ALWAYS HAD THE POWER SUPPLY ON, thus, turning OFF the B+ to the MRF - Dumb, Dumb, Dumb.
OK, with code being the same, what I get is

@ MRF init
Addr=00000000  Value=28
Addr=00000010  Value=8C
Addr=00000100  Value=01
Addr=00000110  Value=63
Addr=00001000  Value=0C
Addr=00001010  Value=03
Addr=00001100  Value=77
Addr=00001110  Value=64
Addr=00010000  Value=32
Addr=00010010  Value=74
Addr=00010100  Value=62
Addr=00010110  Value=32
Addr=00011000  Value=38
Addr=00011010  Value=00
Addr=00011100  Value=19
Addr=00011110  Value=00
Addr=00100000  Value=A3
Addr=00100010  Value=38
Addr=00100100  Value=30
Addr=00100110  Value=00
Addr=00101000  Value=00
Addr=00101010  Value=00
Addr=00101100  Value=53
Addr=00101110  Value=59
Addr=00110000  Value=44
Addr=00110010  Value=00
Addr=00110100  Value=70
Addr=00110110  Value=BC
Addr=00111000  Value=02
Addr=00111010  Value=01
Addr=00111100  Value=5E
Addr=00111110  Value=00
MRF Config data:  Register 01000000   Data=28

MRF Config data:  Register 01000001   Data=28

Which looks like the same date being read from Register zero.......
Now armed with this knowledge, moving forward??

[ecoli-557]

Back to cooking with gas........
Code as was before:

'MRF initialization here
debug "@ MRF init", 10
            MRFConfigSel=0                                                      'Select the chip       
'Going to try using shiftout before using hardware SPI
            for i=0 to 31                                                       'Sets up the index var for data
                MRFaddr=(i << 1)                                                'Shifts address 1 bit to left, automatically gets start,write, and stop bits      
                SHIFTOUT SDO, SCLK, 1,[MRFaddr,RF_Init_Values(i)]               'send data to register
debug "Addr=",bin8 MRFaddr,"  Value=",hex2 RF_Init_Values[i],13
            next i    
            MRFConfigSel=1                                                      'Deselect the chip

'Try to read back some registers as a test to see if they were written correctly            
            MRFConfigSel=0                                                      'Select the chip 
            for i = 0 to 7                                                      'Sets up address to read
                MRFaddr=((i << 1) | $40)                                       'Shifts left 1 bit, sets $40 bit, automatically gets start, READ, and stop bits
                SHIFTOUT SDO, SCLK, 1,[MRFaddr]                                'Address to read
                SHIFTIN SDI, SCLK, 0,[MRFSPIdata]                              'Get data from reg
                debug "MRF Config data: i=", dec1 i, "   Register ", bin8 MRFaddr, "   Data=", hex2 MRFSPIdata,13
            next i

And the debug output:

@ MRF init
Addr=00000000  Value=28
Addr=00000010  Value=8C
Addr=00000100  Value=01
Addr=00000110  Value=63
Addr=00001000  Value=0C
Addr=00001010  Value=03
Addr=00001100  Value=77
Addr=00001110  Value=64
Addr=00010000  Value=32
Addr=00010010  Value=74
Addr=00010100  Value=62
Addr=00010110  Value=32
Addr=00011000  Value=38
Addr=00011010  Value=00
Addr=00011100  Value=19
Addr=00011110  Value=00
Addr=00100000  Value=A3
Addr=00100010  Value=38
Addr=00100100  Value=30
Addr=00100110  Value=00
Addr=00101000  Value=00
Addr=00101010  Value=00
Addr=00101100  Value=53
Addr=00101110  Value=59
Addr=00110000  Value=44
Addr=00110010  Value=00
Addr=00110100  Value=70
Addr=00110110  Value=BC
Addr=00111000  Value=02
Addr=00111010  Value=01
Addr=00111100  Value=5E
Addr=00111110  Value=00
MRF Config data: i=0   Register 01000000   Data=28
MRF Config data: i=1   Register 01000010   Data=8C
MRF Config data: i=2   Register 01000100   Data=01
MRF Config data: i=3   Register 01000110   Data=63
MRF Config data: i=4   Register 01001000   Data=0C
MRF Config data: i=5   Register 01001010   Data=03
MRF Config data: i=6   Register 01001100   Data=77
MRF Config data: i=7   Register 01001110   Data=64

Which matches perfectly.

I did scope the data in, out and noticed the pulses were not right at all.
Then double-check the schematic and board to the datasheets - cuz, you know sometimes there is a mistake.........
I appreciate your help, your going back to basics post was done after I had done it so we think alike - I am just DUMB!

Your help in finding the correct way to address the registers was of immense help, I just missed that one completely.
I'm off to the races, well, off to get the code for a wireless smart sensor system for the house.......
When working, will post code to help others.

Regards.

[Tabsoft]

Okay! Now you're starting to cook.

Stick with just reading the 1st two config registers 00-01($40/$41), but reset your MRFSPIdata to a known value between the two reads.
Read register 00
display data out to debug port.
MRFSPIdata = $BE (or some other value besides $28, $FF, $00, etc)
Read register 01
display data out to debug port.

[Tabsoft]

Looks like we cross-posted.

Glad you moving forward now.
Hope to see your final version when you have it working.

[Tabsoft]

I'm in the Houston Area.

I'm not sure, but it looks like from your posting this is still part of a larger program.
When I am faced with this kind of issue, I go back to square 1.
Create a new program with only the absolute minimum in it.

MCU Config Fuses
Only the VARS needed for this operation (Reading the MRF registers and printing out to the Debug serial port).
TRIS, ADCON, etc for only the pins for the above operation.
Disable the PIC modules correctly.
Aliases only for the SPI pins and the debug pins.
NO Interrupt handling at all. If you are using DT Interrupts, don't do it now.

Do not try to write to the MRF Config registers, only try to read from them after POR + 10ms minimum.

I mean BASIC functionality only.
No ADC, No Interrupts, no nothing.

You are having a fundamental issue just trying to read from the device.

Do you have a logic probe or OScope?
If so, monitor the CSCON line and make sure it is going low as it should.
Monitor the SCK and SDI (on the MRF) to make sure it is being pulsed by the PIC.

Also, and I am not trying to insult you, but how about your wiring between the PIC and MRF for the SPI interface?
PIC MRF Module
MRFReset (PORTA.1) ----- RESET (Pin 2) for MRF89xAM9A (Leave this disconnected for these tests)
MRFConfigSel (PORTA.5) ----- CSCON (Pin 3) for MRF89XAM9A
MRFDataSel (PORTA.4) ----- CSDATA (Pin 8) for MRF89XAM9A (Should be High for all of these tests)
SCLK (PORTx.y) ----- SCK (Pin 6) for MRF89XAM9A
SDO (PORTx.y) ----- SDI (Pin 5) for MRF89XAM9A
SDI (PORTx.y) ----- SDO (Pin 7) for MRF89XAM9A

I didn't see your aliases for SCLK, SDO and SDI.

I know all this probably sounds elementary, but that is what I would do.