Hi! I have idea to use stepper motor as incremental encoder due to its nice feel with some feedback via op-amp. But it has to be very low power, sub 60uA when in sleep mode. I've seen stepper connected to arduino via LM358P op-amp (https://www.instructables.com/Use-a-Stepper-Motor-As-a-Rotary-Encoder/), but it looks that LM358P op-amp draws to much current. I found OP290 op-amp which uses 20uA per amplifier. Would it be possible to use OP290 instead LM358P?
For wireless communication I thinking about nRF24l01.
Next would be MCU. What to choose? Arduino itself is power hungry as I know. Attiny85 hasn't enough pins. Atmega328? For now I would be happy without display, for future it would be nice to have 3 digit 7 segment led to display 0-100% state.
Everything should be powered with single CR2032 battery.
Principle would be like this: when MCU gets signal from op-amp it wakes up, starts counting increments and sends data to arduino via nRF24l01. If MCU got no increments for lets say 10s it shuts down nRF24l01 and goes to sleep mode.
Is it feasible to get it working with low power consumption enough to last at least 6 month from CR2032? I do plan to fit everything under a rotating knob.
So what alternatives do I have?
Use the stepper as a generator to power the gizmo, as well as the encoder. Charge up a supercap, for example.
sends data to arduino via nRF24l01
Can't do that with CR2032 alone, as too much current is required during transmit.
Try to design/build a working prototype first, learn from it, and then use the gained knowledge to try to design a better device (in your case low power).
Leo..
Hmm.. interesting. I found a project with Whisper Node based on atmega328p with RFM69HW and DHT11 sending data every 60s and running for over a year on CR2032. But first I maybe try Wawa's suggestion.
Hi,
What is the application?
Thanks.. Tom.. 
To wirelessly send 0-255 number to arduino nano/mega which will use received number to control pwm dimmer http://www.inmojo.com/store/krida-electronics/item/pwm-2a-ac-light-dimmer-led-50hz-60hz/ in future I might go for Krida 8 channel dimmer, but it isn't pwm compatible.
I'm not sure I need mcu on incremental encoder side. I think it would be fine to directly send a and b signals wirelessly, but I don't know if it is possible without atmega/arduino in between op-amp and nRF24l01.
I don't believe that claim. Please post a link.
There are a lot of calculations in that article, but no actual demonstration that the project can run for a year on a CR2032 cell. I still don't believe it.
The problem is that the transmit function requires large currents, if only for short periods of time, but that severely degrades the battery capacity. Here, the author claims to use the highest power transmit mode of the RFM69HW, which seems pretty silly:
Battery and power supply voltage reading and transmitting:
0.6ms * 9mA + (1.6ms * 2) * 55mA =
5.4mAms + 176mAms = 181.4mAms*
But why bother? A single, much cheaper AA cell can power a remote sensor for years at a time.
You might be right, I'm no expert in electronics. I would consider going for single AAA battery, but then I would need some sort of regulator/converter to get 3V?
Use two.
Agreed, the entire project can run on 2 AAs, with no energy wasting step up converter. A bare bones ATmega328 and an NRF24L01 will both run on as little as 1.9V.
To make the "bare bones" Arduino, I suggest to get a 3.3V Arduino Pro Mini and remove the regulator and power LED. Nick Gammon's site has lots of great ideas and information.
LED displays are very power hungry.
That's what capacitors are for. Couple hundred uF will do it just fine. I did something like this a few years ago for a school project, and it transmitted just fine from a CR2032 once we added the big decoupling capacitor. We also clocked the ATTiny84 we were using down to 4 MHz to get the full battery range.
You can, but what you really want is a comparator, not an op amp. You can do a lot better than 20 uA per amplifier that way. Also, don't forget to take into account the power consumption of the resistor network around the device. That might end up wasting more power than the chip itself.
Yes, caps are a short term solution. But the point in the contention is the project author's claim that the gizmo in the linked article would run for a year on a CR2032.
It would not, especially considering the other obvious errors in the article, which I won't bother to list.
This topic was automatically closed 120 days after the last reply. New replies are no longer allowed.