philvr

I've had two MCP23017 digital i/o chips running well at 400kHz. But running at 1MHz is unsuccessful. I've tried tweaking the bus pull up resistors. Has anyone run I2C at 1MHz? Or Faster? I'd like to know what can be achieved. NDP2 with f/w 4.2.2.2 using MultiI2C (thanks NooM)

I've got the following piece of code public struct MAX6651_DATA { public int nr_fans; public char valid; public uint speed; public uint config; public byte [] tachos = new byte[4]; public uint dac; public uint alarm; } MAX6651_DATA myFan = new MAX6651_DATA(); //create an instance of the fans data structure I would expect... myFan.dac = 100; to assign the dac element the value 100.... Please, what is wrong with this? I must be missing something fundamental! (I fear more background work/reading may be required). Thanks for any advice...

I've been experimenting with a couple of Hall Effect Current Sensors and using them to measure large DC currents (up to 60A). The devices are the CSLA1CD by Honeywell. I've had some success but the device has developed an offset! Out of two units one now produces 5.8V when zero current flows and the other 6.3V! Originally they both produced 6V! My question is has anyone here experience of using Hall Effect Current Sensors? Can they become permanently magnetised if exposed to magnetic fields from associated high current DC cables and hence produce an output that is out of specification? I can not think of any other reason for them to develop an offset but I am new to these devices so am hoping someone reading this can offer advice. regards Philvr

The i2c MAX6651 fan controller has a tacho count time register that sets a tacho count period 0.25S, 0.5S, 1S or 2S. The tacho counter is read and the time of a fan rotation derived from the counting period set in the tacho count register and the number of pulses counted by the tacho.Questions is this. As the tacho count time register is in powers of 2, is it quicker to use a bitwise shift ( to shift the tacho counter reading right to multiply by two (left to divide by 2) or to use an integer multiply/divide operation?Does anyone know if the clr c# interpreting system would choose the most efficient operation?

Been playing with this chip but it is not behaving quite as I expect. The data sheet states: "Writing to the GPIOn register actually causes a write to the latches (OLATn). Writing to the OLATn register forces the associated output drivers to drive to the level in OLATn" How does this work in practise? Does it matter which port I write to? Is it rude to answer ones own question? I find that if I write zeros to either register I can write the desired output value to the other obtain an output! I haven't discovered what the difference is between the two registers. Perhaps someone can advice?