Maker Pro
Maker Pro

Lowest power possible for driving segment LCD glass with accurate time

mikebounds

Jan 25, 2017
8
Joined
Jan 25, 2017
Messages
8
What is the lowest power possible for driving segment LCD glass where most of the time the LCD display will just display the time with an accuracy of at least 20ppm.
As I understand the components consuming power would be:
  1. Glass LCD segment display (dependent on number and size of segments)
  2. External RTC chip
  3. Micro-controller
Is is possible to have the total power less than 10uW?

I found this whitepaper:
http://www.nxp.com/assets/documents/data/en/white-papers/S08LL16PWRCMPWP.pdf
which is a good starting point as this would seem to state the most important thing is the standby power mode of the micro-controller and this says the NXP S08LL16 would use 3.3uA @ 3.6V (11.9uW), but this whitepaper is over 7 years old, so I didn't know if there are more efficient micro-controllers now?
I found MSP430FR413x which has a low-power mode with LCD on (LPM3), of 1.25uA @ 3V, but this figure is without the charge pump enabled - is this something that MUST be enabled for a usable display?

Thanks

Mike
 

Minder

Apr 24, 2015
3,323
Joined
Apr 24, 2015
Messages
3,323
Do you mean segment display's which are LED based or LCD pixel based displays such as the Hitachi HD44780 etc?
M.
 

mikebounds

Jan 25, 2017
8
Joined
Jan 25, 2017
Messages
8
Reflective 7/14 segment LCD Chip on Glass (COG).
As I understand LCD COG uses less power than LCD COB with an integrated controller (like HD44780) and LED would use even more power and I am looking for the lowest possible power.
I don't need a graphic display so don't need LCD pixel, rather 4 to 6 digits using 7 or 14 segments.

Mike
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,505
Joined
Jan 21, 2010
Messages
25,505
I would guess the lowest power would be a function of leakage, capacitance, and frequency. The first two would vary from panel to panel, the last of under your control, but limited by the panel.

Have you viewed any datasheets?
 

mikebounds

Jan 25, 2017
8
Joined
Jan 25, 2017
Messages
8
I have looked at lots of datasheets for micro-controllers with:
  1. Integrated LCD driver
  2. Standby mode where RTC is running and CPU is not, and power to LCD can be optionally on
Looking at the NXP S08LL16 datasheet - http://www.nxp.com/assets/documents/data/en/data-sheets/MC9S08LL16.pdf this shows (page 26) the current in Stop3 mode at 25°C with RTC, LCD and Charge pump on is 6.1uA @ 3V (18.3uW)

Looking at the TI MSP430FG461x, MSP430CG461x datasheet - http://www.ti.com/lit/ds/symlink/msp430fg4618.pdf
this shows (page 16) the max current in LPM3 in 4-mux mode at 25°C with RTC, LCD on and charge pump off is 5.5uA @ 2.2V and the charge pump takes 3uA @ 2.2V so this totals 8.5uA @ 2.2V (18.7uW)

But the NXP whitepaper says it's figures come from the datasheets and it has:
NXP: 3.3uA @ 3.6V
MSP 4618: 9.8uA @ 2.2V

so I don't understand how NXP calculates its figures as my figures say the 2 chips both consume about 18uW.

It seems some MCs drive the LCD displays more efficiently, but as I can't reconcile NXP's figures, I don't know for instance if the MSP430FR413x is more efficient than the MSP430FG461x.
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,505
Joined
Jan 21, 2010
Messages
25,505
NXPs figures may be actual figures from a device rather than a maximum specified value.

You'll note that the datasheet value are maximum figures,. Typically typical figures are lower.
 
Top