cablescapacitancei2cwire
I'm using an I2C interface for my e-compass and controller. The capacitance of the e-compass I2C pin is 20pF, and that of the I2C controller is approximately 40pF.
But the problem is: I want to use a 100cm long wire, but don't know how to measure wire capacitance, and there is a PMDC motor which injects noise into the line.
The question is how to ensure my I2C capacitance is not exceeded,
and what I2C buffer I need to use (any specfic IC?).
Where did you get the idea that SDA and SCL lines should be twisted?
This is the second worst thing you can do to I2C communication next to cutting the wires.
Twisting is suitable for differential lines such as CAN or RS485/422.
With slow enough communication you probably won't even need repeater/extender on 2.5m. Especially, if this a 5V system. Just untwist those I2C lines.
If you have some spare lines on the ribbon cable try to put some GND lines in between I2C (non-twisted )pairs or even between SDA and SCL of same channel.
EDIT: there is a way where twisted-pair might come in handy for I2C. It is where each of the two signals is transmitted differentialy (some kind of 4-wire I2C).
Driving I2C-bus signals over twisted pair cables with PCA9605 transmits SDA bidirectionally over a twisted pair of wires.
Sending I2C-bus signals via long communications cables with P82B96 or PCA9600 explains the advantages of transmitting SDA using a unidirectional "4-wire driving method".
You definitely need external pull-up resistors. The internal pull-ups of a microcontroller are too weak. Note that with a high capacitance bus it's usually not a problem to have fast falling edges, because the FETs which pull the line low have low impedance, and then the wire's capacitance can be discharged quickly.
But for the rising edge we only have the relative high resistance of the pull-ups, and the RC time of that combined with the wire's capacitance determines the rise time. A smaller resistance value is better, but there's a limit to this too: an I2C device doesn't have to draw more than 3 mA, so at 5 V bus voltage the resistance shouldn't be lower than 1.7 kΩ.
The graph shows the maximum pull-up resistance for a given wire capacitance. At 1 m that capacitance won't be high, if you use signal wire probably less than 50 pF, and even for 100 pF you could go up to 12 kΩ. I'd stay near the lower limit though, not only will you have faster edges, but a lower impedance also makes the wire less sensitive to noise. 2.2 kΩ resistors will probably do.
Further reading
I2C Specification
I2C termination versus capacitance
Best Answer
Despite the capacitance specifications being what they are, you are unlikely to see any ill-effects of cable capacitance at typical I2C signaling rates up to 400 KHz Fast Mode, for a cable length of 1 meter.
Also, as per I2C specifications by NXP, the maximum supported cable capacitance is 400 picoFarad. This apparently applies to the original I2C specification (100 KHz), there seems to be no update for the newer I2C versions.
For 1 MHz Fast Mode Plus, 3.4-MHz High-speed mode (Hs) and 5 MHz Ultra fast mode (UFm) a long cable may be unreliable, or will not work at all, at worst. If the peripheral in question supports and is to be used in these high speed modes, a recommended approach would be to avoid off-board I2C altogether, or add a Long Distance I2C extender.
I2C buffers/extenders: NXP publishes an array of suitable parts on this page, specifically designed to allow extending the I2C bus over kilometers. P82B715, P82B96, and PCA9600 are the parts to look at, in this context.
There are also suitable parts from Texas Instruments, and perhaps others, for such bus extending purposes.
Also see Samuel's answer for some further insight into this matter.