4 Wire SPI with Shiftin/Shiftout ?

[CuriousOne]

Hello.
I'm trying to connect GT21L16S2Y Font ROM IC to PIC16F886.
I have soldered connections, but have not written code yet, because this chip has separate IN/OUT pins for SPI.

Below is the draft code which I think should work (have not tried it yet).



SCLK VAR PORTB.5 'CLOCK
TC VAR PORTC.0 'TO CHIP - WRITE
FC VAR PORTC.6 'FROM CHIP - READ
RST VAR PORTB.4 'RESET
I VAR BYTE 'COUNTER VAR
X VAR BYTE 'TEMP VAR
FONT VAR BYTE [8] 'FONT ARRAY






Reader: 'Font reading outline
LOW RST ' Ready for transfer
Shiftout TC, SCLK, 5, [$03,$0] ' Send write command, set start offset
For I=0 to 7 'read 8 bytes of data
Shiftin FC, SCLK, 5, [x]
FONT[I]=X 'WRITE TO ARRAY
next
HIGH RST ' Reset IC
Return



And here is screenshot from the datasheet. It is in chinese, but diagrams show things clearly.

So I'm on the proper way or not?

9101

[HenrikOlsson]

I have soldered connections, but have not written code yet, because this chip has separate IN/OUT pins for SPI.
Of course they have separate in/out pins, that's how SPI works.


So I'm on the proper way or not?
I think so. But for starters, looking at the image you posted you should send an 8 bit command followed by a 24 bit address (32bits in total), you're only sending 16.

[CuriousOne]

Well, MAX7219, DS1302 we use only 3 pins.
Yes, I'm missing two more bytes there, will add them later and post back :)

[HenrikOlsson]

The MAX7219 is not capable of outputting any data, therefor no data out pin.
DS1302 might work with SHIFTIN/SHIFTOUT but its interface isn't really SPI.

[richard]

So I'm on the proper way or not?

not. the shiftout shiftin commands leave the clock line high on completion{mode 5} so when you swap between them the slave device will see
a false rising edge and you lose a data bit .

nearly all respectable pic16/18 chips have a mssp module why not just use it ?
or write a bit banger to suit


Well, MAX7219, DS1302 we use only 3 pins
which are not spi anyway , its a stupid argument

[CuriousOne]

Yes, I tried with mode 5 and it does not works, it also does not works with mode 4 or others.
MSSP or whatever it is, might be good, but are there direct support statements in PBP ?
And also, I already have pre-wired devices and they're meant to be in that way.

The code below does not works - it returns all 255.



;----[16F886 Hardware Configuration]--------------------------------------------
#CONFIG
cfg1 = _INTRC_OSC_NOCLKOUT ; INTOSCIO oscillator: I/O function on RA6/OSC2/CLKOUT pin, I/O function on RA7/OSC1/CLKIN
cfg1&= _WDT_ON ; WDT enabled
cfg1&= _PWRTE_OFF ; PWRT disabled
cfg1&= _MCLRE_OFF ; RE3/MCLR pin function is digital input, MCLR internally tied to VDD
cfg1&= _CP_OFF ; Program memory code protection is disabled
cfg1&= _CPD_OFF ; Data memory code protection is disabled
cfg1&= _BOR_OFF ; BOR disabled
cfg1&= _IESO_ON ; Internal/External Switchover mode is enabled
cfg1&= _FCMEN_ON ; Fail-Safe Clock Monitor is enabled
cfg1&= _LVP_OFF ; RB3 pin has digital I/O, HV on MCLR must be used for programming
cfg1&= _DEBUG_OFF ; In-Circuit Debugger disabled, RB6/ICSPCLK and RB7/ICSPDAT are general purpose I/O pins
__CONFIG _CONFIG1, cfg1


cfg2 = _BOR40V ; Brown-out Reset set to 4.0V
cfg2&= _WRT_OFF ; Write protection off
__CONFIG _CONFIG2, cfg2


#ENDCONFIG




'chip configs
TRISA=%01000000 'SET A TO OUTPUT 1=input
TRISC=%00000000 'set half C for in/out
TRISB=%00000000 'set PortB to output
ANSELH=%00000000 ' ADC OFF B
ANSEL=%000000000 'configure PortA as digital except first 2
ADCON1=%10000000 'adc justify
OSCCON=%01110101 'SET FREQUENCY TO 8MHZ
WPUB=%00000000 'turn off Pullups
CM1CON0=0 'DISABLE COMPARATORS
CM2CON0=0 'SAME HERE


DEFINE OSC 8
DEFINE ADC_BITS 10
DEFINE ADC_CLOCK 3
DEFINE ADC_SAMPLEUS 50
'lcd config
pause 10
' Set LCD Data port
DEFINE LCD_DREG PORTC
DEFINE LCD_DBIT 4
DEFINE LCD_RSREG PORTB
DEFINE LCD_RSBIT 0
DEFINE LCD_EREG PORTB
DEFINE LCD_EBIT 1
DEFINE LCD_BITS 4
DEFINE LCD_LINES 2
DEFINE LCD_COMMANDUS 1500
DEFINE LCD_DATAUS 44




SCLK VAR PORTB.5 'CLOCK
TC VAR PORTC.0 'TO CHIP - WRITE
FC VAR PORTC.6 'FROM CHIP - READ
RST VAR PORTB.4 'RESET
I VAR BYTE 'COUNTER VAR
X VAR BYTE 'TEMP VAR
FONT VAR BYTE [8] 'FONT ARRAY


Reader: 'Font reading outline
LOW RST ' Ready for transfer
Shiftout TC, SCLK, 5, [$03,$100,$00,$0] ' Send write command, set start offset
For I=0 to 255 'read 8 bytes of data
Shiftin FC, SCLK, 5, [x]
LCDOUT $fE, $C0, DEC I,32, DEC X
Pause 200
next
HIGH RST ' Reset IC
GOTO READER

[HenrikOlsson]

I'd try mode 5 for SHIFTOUT and mode 6 for SHIFTIN.
I would also add some code to make sure #CS is high for a while during startup. Then I would add a scope or logic analyzer to verify that what you think is happening is actually happening at the hardware level.

[richard]

I'd try mode 5 for SHIFTOUT and mode 6 for SHIFTIN.

You can learn something everyday
that seems to work with spi 25aa1024 eeprom

[CuriousOne]

Shiftout 6 changed nothing, so will hook up scope in logic analyzer mode and report.

[CuriousOne]

Hooked scope to pins.
There is series of pulses on SCLK and RST gets high and low as specified.
However, there is no activity on serial input/output lines.
To check the wiring, I wrote this simple code:



krd:
high sclk
high tc
high rst
high fc
pauseus 500
low sclk
low tc
low rst
low fc
pauseus 500
goto krd


And it works fine - I can see square pulses with scope on each pin, so there are no wiring issues.

Any ideas?

[HenrikOlsson]

One issue might be the fact that you have LCD datapins overlapping at least one of the pins that you're using with SHIFTIN/SHIFTOUT.
This code does not match comment: TRISC=%00000000 'set half C for in/out.
Also, note that my suggestion was to try mode 6 on SHIFTIN - not SHIFTOUT. Not that it would make any difference untill SHIFTOUT is outputting data.

[richard]

not to mention. a 32bit address attempt ?


Reader: 'Font reading outline
LOW RST ' Ready for transfer
Shiftout TC, SCLK, 5, [$03,$100,$00,$0] ' Send write command, set start offset
For I=0 to 255 'read 8 bytes of data
Shiftin FC, SCLK, 5, [x]
LCDOUT $fE, $C0, DEC I,32, DEC X
Pause 200
next



5.75 SHIFTOUT
SHIFTOUT DataPin, ClockPin, Mode, [Var{\Bits}...]
Synchronously shift out Var on ClockPin and DataPin. ClockPin and DataPin may be a constant, 0-15, or a variable that contains a number 0-15 (e.g. B0) or a pin name (e.g. PORTA.0).
SHIFTOUT is a software-based command and does not require that the target device have synchronous serial capability. The ClockPin and DataPin parameters may be set to any digital I/O pins, and may be different in different commands within the same program.
\Bits optionally specifies the number of bits to be shifted out. If it is not specified, 8 bits are shifted out, independent of the variable type. The Bits shifted out are always the low order bits, regardless of the Mode used, LSB or MSB. Up to 32 Bits can be shifted out of a single (long) variable. If more than 32 Bits are required, multiple variables or constants may be included between the square brackets.

[CuriousOne]

OK I moved these ports to C0 and C1.

SCLK VAR PORTB.5 'CLOCK
TC VAR PORTC.1 'TO CHIP - WRITE
FC VAR PORTC.0 'FROM CHIP - READ
RST VAR PORTB.4 'RESET

Set all ports to output, except FC, which is set as input.

And I don't get, what's wrong with $100 ?
I'm sending 3 bytes, that's 24 bits.

[CuriousOne]

It still does not works.
I removed all LCD code
I mapped output to different port - still can't capture anything with scope.
And there are no issues with hardware - setting these pins high-low works perfectly.
So it seems like shiftout not working?

[HenrikOlsson]

Yes, you're sending 3 bytes (four atually if you count the command byte ($03)) but what Richard is saying is that $100 does not fit within a byte and according to the manual it will then get truncated to 8bits so in effect your shiftout statement send 3,0,0,0. Are you SURE you don't see a short pulse on the FC-pin? Try

DoIt:
LOW RST
SHIFTOUT TC, SCLK, 5, [$AA,$AA,$AA,$AA]
HIGH RST
Pause 2
Goto DoIt

Set the scope to trig on the falling edge of RST.

[richard]

And I don't get, what's wrong with $100 ?

how big is the biggest number that can be represented in a byte ?



Shiftout TC, SCLK, 5, [$03,$100,$00,$0] ' Send write command, set start offset

an attempt to read from address 0x1000000 a non valid 25 bit address, except that shiftout will send address of 0x00 since the 8 bit default not over written

maybe [3,1,0,0] or [3,100,0,0] or [3\8,$100\16,0\8] or [3,$100\16,0]

[CuriousOne]

Guys, issue is different - shiftout does not output any data on data pin. So for sure, there will be nothing to read with shiftin.
This is the issue.

[richard]

should we just guess about the new connections and code ?

[CuriousOne]

The code is posted above and connections are clearly seen from there and described :)

took brand new all parts, will try on breadboard later :)