Hardware I2C

[BobK]

Hi Toley00,

It may be worth your time to download the I2C Bus Specification manual from Phillips. There are 3 data modes for I2C, Standard Mode = upto 100kbits/sec, Fast Mode = upto 400kbits/sec, and High Speed Mode = 3.4Mbits/sec. I've only used I2C for RTC's using both software I2C and Hardware I2C. I didn't have to set anything up special on my PICs for speed. But you do have to put a 10ms pause after any write command to give it time to catch up or settle down. I have seen some posts on this forum where I2C worked better when it was slowed down.

Don't know if this helps, but I think maybe their manual might give you a better insight into the whole process.

BobK

[Toley00]

Thank you BobK for your reply.
I already know about I2C timing protocol, and I use it in many project already. I just need it to read an EEPROM faster than I2CREAD can do.
Thanks for your help but a 10ms pause will not give me speed...

[SteveB]

I don't know what PIC you are using, so I can't promise anything, but I've attached a file containing some portions of code I have used with the hardware I2C on a PIC 18F4620.

The file is a compilation of cut and past from a few include files, and has only been minimally "cleaned up" for public viewing. It has 4 sections:
1) Basic PIC and HSEROUT setup (mostly to reinforce what pic and how it's configured)
2) The meat of the I2C commands
3) Samples of Large Block Reading/Writing to an EEPROM
4) Samples of Reading/Writing to an RTC

Hopefully this will get you started.
Best of luck,
Steve B

Disclaimer: Looking a little closer I recall this being one of my first PBP-ASM items I tackled. It worked, and with other things pressing, I haven't gone back and cleaned things up, like unused variables (DelayCtr1 and DelayCtr2) and rements of HSEROUT commands used to debug.

[Toley00]

Thank you SteveB, looks like what I need. I will take a closer look and adapt it to my needs. I'm using a PIC16F886, but I think all MSSP modules work the same way, and I only need to read from the EEPROM.

[SteveB]

Thank you SteveB, looks like what I need. I will take a closer look and adapt it to my needs. I'm using a PIC16F886, but I think all MSSP modules work the same way, and I only need to read from the EEPROM.

A couple of things right off the top of my head. The buffer size will need to be reduced for a 16F. Also, the way I allocated variables to "BANKA" or "BANKX" in the declaration statements will also need changing. If I get a chance later I'll try to give it a look-over for anything else. As I recall, I think I got the basics of the ASM routines from a Microchip document/source, and adapted them for the 18F. So that might also be an avenue to persue.

Steve B

[sougata]

Hi,

SteveB your routines are neat (at least than mine) and I learnt from it as well. Thanks to you and Toley for raising the topic.

[SteveB]

I made a quick run through the code (while my kids took a nap). Changed all the ASM to PBP. Also modified the buffer size, and some SFR lables to correspond the 16F. It compiles OK, but no 16F's to test it (nor anything to setup the circut without canibalizing something else). Keep in mind, on many of the 18Fs, you've got RAM to burn, so the buffers reflect that. On the 16Fs, you don't have that luxary.

Hopefully, if you are working it on your end, you can compare/use what I've done to move further down the road.

I did notice a number of areas where things could be improved or I would do thing differently. The error handling is especially weak. However, I never got any errors with the code and my setup, so never needed it to be very robust. So, I will put this on my *LONG* list of things to do when I get around to it.

SteveB