Problem reading 24FC1025 EEPROM with 18f14k22

[JimAvanti]

Based off of an article (Serial EEPROM Part #1) by mackrackit, I have been attempting to read/write to a 24FC1025 EEPROM. I have done this in the past using an 18f4620 and everything worked as expected. I am trying to run the same code on a 18f14k22 and I can't get any read/writes to work when using the I2cread/write commands, but it works flawlessly with the shiftin/out examples. all the bytes I am reading is using the I2cread command come in as decimal 16. I tried slowing down the micro and that didn't change anything. Anyone have any ideas?



' *** I2C EEPROM using Siftin/out
START_Shiftin:
READ_DATA:
lcdout Lcdcmd , Line4, "Reading EEPROM Data "
For ADDR_LB = 0 To 50
GOSUB INT_ROM
GOSUB START_ROM
SHIFTOUT DPIN,CPIN,1,[ExtMem0R]
SHIFTIN DPIN,CPIN,0,[ACK\1,DATI\8]
SHIFTOUT DPIN,CPIN,1,[%1\1]
GOSUB STOP_ROM
lcdout Lcdcmd , Line3, "ADDR:",DEC ADDR_LB," "
lcdout Lcdcmd , Line3+10, "DATA:",DEC DATI," "
PAUSE 250
NEXT ADDR_LB
Return

INT_ROM:
GOSUB START_ROM
SHIFTOUT DPIN,CPIN,1,[ExtMem0W]
SHIFTIN DPIN,CPIN,0,[ACK\1]
SHIFTOUT DPIN,CPIN,1,[ADDR_HB]
SHIFTIN DPIN,CPIN,0,[ACK\1]
SHIFTOUT DPIN,CPIN,1,[ADDR_LB]
SHIFTIN DPIN,CPIN,0,[ACK\1]
RETURN

START_ROM:
HIGH CPIN:HIGH DPIN:LOW DPIN
RETURN

STOP_ROM:
HIGH CPIN:LOW DPIN:HIGH DPIN
RETURN

end


' *** I2C EEPROM using PicBasic I2C commands
START_I2C:
lcdout Lcdcmd , Line4, "Reading EEPROM Data "
For ADDR = 0 To 50
I2CREAD DPIN,CPIN,$A0,ADDR,[DATI]
lcdout Lcdcmd , Line3, "ADDR:",DEC ADDR," "
lcdout Lcdcmd , Line3+10, "DATA:",DEC DATI," "
PAUSE 250
Next ADDR
Return

[JimAvanti]

I have been doing some more testing and found that the data byte returned from the I2CRead relates to the port-pin I have the DPIN set to. I originally had the DPIN set to portA.4 and I was getting 16 as a result on all my reads. I later moved DPIN to A0 and got all 1's. PortB.6 returns all 64's, PortB.7 Returns all 128's... This is very strange. Am I missing something obvious or is there a bug in pBasic Pro 2.60C with I2C eeprom reads using the 18f14k22.

[mackrackit]

I am happy to see that the SHIFTIN/OUT example has possibly helped!

But I have not used the 18F14K22.

Have you checked the analog functions of the pins you are using?
As usual, post the code that is giving you trouble.

[JimAvanti]

Mackarackit,
If I have to use your Shiftin/out method I can get by, but I see no reason the I2C commands shouldn't be working. Especially since I need to read a sting of data on each read and I2Cread does that very easily with STR. All the I/O pins are set up correctly (All digital, all interrupts disables, direction correct…). If someone could try reading one of these eeproms on a 18f14k22 I would love to hear of your results.

I had a similar issue on a 18f46k22 where I was trying to compare all 28 ADC pins with each other pin in turn one at a time and I saw a strange result (Incremented after a certain point) and it turned out that there was a bug in PicBasic pro which got fixed. This is looking like a bug to me as well

[JimAvanti]

I still can't get the I2CRead/Write to work with this EEPROM and I have used them with other PICs in the past with no problems. I used my Logic Analyzer to capture both the Shiftin/out and the I2CRead/Write examples. I could follow the working Shiftin/out on the Logic analyzer, but the I2C is either faster than I can capture, or completely messed up. I am using the same program to test both routines.
I just tried uploading the analyzer screenshots, but it said I didn't have permissions.

This code works:
Test_ShiftOut: ' Write "A" byte to EEPROM at Location 0
ADDR_HB = 0
ADDR_LB = 0
Char = "A"
GOSUB INT_ROM
SHIFTOUT DPIN,CPIN,1,[Char\8,%1\1]
GOSUB STOP_ROM
PAUSE 10
Return

Test_ShiftIn:
GOSUB INT_ROM
GOSUB START_ROM
SHIFTOUT DPIN,CPIN,1,[ExtMem0R]
SHIFTIN DPIN,CPIN,0,[ACK\1,Char\8]
SHIFTOUT DPIN,CPIN,1,[%1\1]
GOSUB STOP_ROM
Return

INT_ROM:
GOSUB START_ROM
SHIFTOUT DPIN,CPIN,1,[ExtMem0W]
SHIFTIN DPIN,CPIN,0,[ACK\1]
SHIFTOUT DPIN,CPIN,1,[ADDR_HB]
SHIFTIN DPIN,CPIN,0,[ACK\1]
SHIFTOUT DPIN,CPIN,1,[ADDR_LB]
SHIFTIN DPIN,CPIN,0,[ACK\1]
RETURN

START_ROM:
HIGH CPIN:HIGH DPIN:LOW DPIN
RETURN

STOP_ROM:
HIGH CPIN:LOW DPIN:HIGH DPIN
RETURN


This code doesn't work:
Char = "A"
I2CWRITE DPIN,CPIN,ExtMem0,ADDR,[Char]
Pause 10
I2CREAD DPIN,CPIN,ExtMem0,ADDR,[Char]

The analyzer shows the I2C commands to be much faster. Maybe the EEPROM is too slow?





I am happy to see that the SHIFTIN/OUT example has possibly helped!

But I have not used the 18F14K22.

Have you checked the analog functions of the pins you are using?
As usual, post the code that is giving you trouble.

[JimAvanti]

Trying again to upload the analyzer screenshots:

I2cWrite:
6997

Shiftout (Works):
6996

[Demon]

(Can't believe I have the PBP manual on my phone. lol )

You don't mention the osc speed. The manual mentions up to 20MHz, the 18Fs can go much faster.

Odd that your clock is whacked. What about pull-ups on your pins?

Robert

[Dave]

I bet it's that there are no pullup resistors on the SCL and SDA lines when using I2C command. The SHIFTOUT and SHIFTIN command use the pins in a sink and source mode but the I2C mode reverses the the data pin direction at the end of the byte sequence looking for the ACK bit from the slave device. There must be a pullup on the data line to provide sourcing current back to the PIC. If the PIC does not see the proper ACK signal then the command sequence is aborted.

[JimAvanti]

I am running at 16MHz with the Internal OSC. I do have a 10k pullup on the SDA line. I tried a 2k as well and it did the same thing. I don't have any pullups on the SCL line now, but I did try that before with no luck. I will try it again. I may drop a 16f690 in for testing purposes (Pin compatable with the 18f14k22 and I know they work with this EEPROM).



I bet it's that there are no pullup resistors on the SCL and SDA lines when using I2C command. The SHIFTOUT and SHIFTIN command use the pins in a sink and source mode but the I2C mode reverses the the data pin direction at the end of the byte sequence looking for the ACK bit from the slave device. There must be a pullup on the data line to provide sourcing current back to the PIC. If the PIC does not see the proper ACK signal then the command sequence is aborted.

[JimAvanti]

I am running the Internal OSC at 16MHz. I did try slowing the clock down and it still didn't work. I am not sure if the Clock is whacked or if my Logic Analyzer is just too slow to capture it (500kHz is as fast as it will run on my computer for some reason).

I am still leaning toward a bug in Picbasic Pro 2.60C (Not because it isn't working at all, but because the value being returned is directly related to to the binary value of the pin it is connected to). If I connect to pin ra.0 I get a return value of 1, ra.1 returns a value of 2, ra.2 = 4... This shouldn't be happening unless I either made some strange mistake with variable ussage, or if there is an internal bug in the compiler. I saw a similar wierd sort of incremental value on the 18f46k22 while reading ADC lines and it turnned out to be a bug in the complier (which got fixed with 2.60c).

One other thing to note is that the EEPROM address on the Shiftin uses two byte variables while the I2Cread/write uses one 16-bit word variable for the address.

Jim



(Can't believe I have the PBP manual on my phone. lol )

You don't mention the osc speed. The manual mentions up to 20MHz, the 18Fs can go much faster.

Odd that your clock is whacked. What about pull-ups on your pins?

Robert

[Dave]

This routine has worked for me just fine since PBP2.32 w/4 each 24LC1025's. Just exchange the I2CWRITE for I2CREAD and your done....

CNTRL_BYTE CON $A0 'CONTROL BYTE FOR I2C EEPROM MEMORY (A0/A1 USED FOR CHIP SELECT)

BLOCK VAR WORD '128 BYTE BLOCK ADDRESS
ADDRESS VAR WORD 'ADDRESS WORD FOR SERIAL EEPROM
CNTROL_BYTE VAR BYTE 'CONTROL BYTE TO SEND TO SERIAL EEPROM
EEDEVICE VAR BYTE 'PHYSICAL DEVICE ADDRESS
STOR_DATA VAR BYTE[128]'STORAGE DATA ARRAY

EEDEVICE = BLOCK / 512 'SELECT WHICH PHYSICAL ADDRESS BOUNDRY & EEPROM
ADDRESS = BLOCK << 7 'CALCULATE 12C ADDRESS TO STORE DATA TO
CNTROL_BYTE = CNTRL_BYTE 'COPY CONTROL BYTE
CNTROL_BYTE.1 = EEDEVICE.1 'SET LSB OF HARDWARE ADDRESS
CNTROL_BYTE.2 = EEDEVICE.2 'SET MSB OF HARDWARE ADDRESS
CNTROL_BYTE.3 = EEDEVICE.0 'SET 64K BLOCK BIT
I2CWRITE SDA,SCL,CNTROL_BYTE,ADDRESS,[STR STOR_DATA\128] 'SAVE DATA TO 12C
PAUSE 6 'ALLOW TIME FOR I2C WRITE ~5Ms.

Also a MAXIMUM resistance on each SCL and SDA lines to VCC of 2.2k. The faster the bus speed the smaller the resistance.

Dave Purola,

[JimAvanti]

Dave & DEMON,
Problem solved! I added the 10k pull-up resistor to the SCL line as you suggested and everything started working. I did have a 10k on both the SCL and SDA lines originally, but at that time I must have had something else wrong. I checked the datasheet which only stated pullup of 2k-10k on the SDA line so I removed the SCL pull-up and never put it back. I was able to fully load the EEPROM and now I can read an entire string from EEPROM at a time (Much better than the ShiftIn/Out way). I guess since the Shiftin/out was working I assumed the electronics was correct. I have used these EEPROMS in the past and never had an issue. I guess I should have looked at my old schematics before pointing the finger at Pic Basic!

Thank you all for the help!

Jim


I am running at 16MHz with the Internal OSC. I do have a 10k pullup on the SDA line. I tried a 2k as well and it did the same thing. I don't have any pullups on the SCL line now, but I did try that before with no luck. I will try it again. I may drop a 16f690 in for testing purposes (Pin compatable with the 18f14k22 and I know they work with this EEPROM).