Tickless and CORTEX_ENABLE_WFI_IDLE

[dflogeras]

Hi, on a previous (Chibi-2.x) project I was able to decrease current significantly by enabling the WFI_IDLE option. Now that I am using dynamic ticks in Chibi-3.x, is there any point to WFI_IDLE? I empirically tried this and didn't see much difference in current consumption, but I am not sure that could not be for some other reason I am not aware of.

Thanks,
Dave

[Giovanni]

Hi,

It should still make a difference, I am surprised that you don't see differences, are you sure it is really enabled?

Giovanni

[dflogeras]

I think the reason I wasn't measuring any difference is from clocking my uC lower and lower, I've now gotten it to ~1mA which is now a small portion of my power budget. I was mistakenly measuring my total power supply current and since the uC has gone from being about 80% of my total current, to <10% of my current I wasn't seeing a difference. Trying to measure iDD directly is now down to the LSB of my DMM so I'll have to wait until my INA219 current probe arrives

Good to know that it is still useful to use WFI though.

[electronic_eel]

Trying to measure iDD directly is now down to the LSB of my DMM so I'll have to wait until my INA219 current probe arrives

You may also want to take a look at the AD8428. With it you can use your scope to monitor the current consumption of your controller in high bandwidth and correlate the functions and modes it is working in with the actual current used. I find that very helpful when optimizing power consumption.

See this blog entry, that is where I first read about it: http://www.dorkbotpdx.org/blog/paul/measuring_microamps_milliamps_at_3_mhz_bandwidth

[dflogeras]

Whoa, yeah I like that idea! Thanks again for the insight electronic_eel, you're like my guardian angel