Choosing an mcu for battery cell powered motor project

Thread Starter

salvus

Joined Apr 2, 2020
49
Hi, I am looking into which MCU to choose to control two small coreless motors that will be powered by a battery with low voltage. I came across the ATtiny43U but it is seems to be a bit on the pricey side (€1.30 for 5000 according to the website). Does anyone know of anything that might be suitable at lower cost? What I want to do is, have the motors turn on and off when a sound of a certain frequency is detected.

many thanks for your time and help :)
 

MrSalts

Joined Apr 2, 2020
2,767
So, before we all start chaising wild geese (a wild goose?), what is your desired price target if €1.30 is too much? How many will you be buying (should we assume the 5000 pieces you quoted?). Be sure to include shipping charges. Also, do you have a way to program the MCU? Mount the MCU to a PCB?
 

Thread Starter

salvus

Joined Apr 2, 2020
49
For controlling a couple of toy motors I would expect something below 50 cents. Does that seem a bit unrealistic? Seen as they are tiny, does the shipping really add that much if you buy in the thousands? Regarding the programming, a one time cost of a few thousand $ would be acceptable. Do you think it is achievable?
 

Thread Starter

salvus

Joined Apr 2, 2020
49
Perhaps it’s possible just to use a super cheap MCU that uses higher voltage and have a separate boost converter chip. I saw this tutorial that talks about using a 555 timer as a boost converter:

https://theorycircuit.com/boost-converter-circuit-555/

I think they are only a few cents. Do you think that could be used to step up the voltage to power an MCU? Also, in the video for the ATtiny43U (
) it is mentioned that a motor can be powered directly from the ICs pins. If we don’t need to be so stringent on the voltage, are there any lower cost MCU that would also allow for this? Is that quite common?

thanks again for your time and help
 
Last edited:

MrSalts

Joined Apr 2, 2020
2,767
For controlling a couple of toy motors I would expect something below 50 cents. Does that seem a bit unrealistic? Seen as they are tiny, does the shipping really add that much if you buy in the thousands? Regarding the programming, a one time cost of a few thousand $ would be acceptable. Do you think it is achievable?
In the past year, there doesn't seem to be any 50¢ micrcontrollers - you've probably heard of the chip shortage. Now is a challenging time to design anything because part inventory changes before you can get a design manufactured.
 

MrChips

Joined Oct 2, 2009
30,806
For controlling a couple of toy motors I would expect something below 50 cents. Does that seem a bit unrealistic? Seen as they are tiny, does the shipping really add that much if you buy in the thousands? Regarding the programming, a one time cost of a few thousand $ would be acceptable. Do you think it is achievable?
What difference does it make between a $0.50 chip and a $2.00 chip when shipping charges will be $8.00 and you will spend 100 hours of your time learning how to program the chip?

Here you go:
PIC16HV616 going for US$0.365 for quantity of one.
https://www.newark.com/microchip/pic16hv616-i-p/microcontroller-mcu-8-bit-pic16/dp/74K8595
 

MrSalts

Joined Apr 2, 2020
2,767
What difference does it make between a $0.50 chip and a $2.00 chip when shipping charges will be $8.00 and you will spend 100 hours of your time learning how to program the chip?

Here you go:
PIC16HV616 going for US$0.365 for quantity of one.
https://www.newark.com/microchip/pic16hv616-i-p/microcontroller-mcu-8-bit-pic16/dp/74K8595
It makes a difference if he is making 5000 pieces of a toy - it's the shipping charges that make no difference at that volume. Does that make sense?


€1.30 for 5000 according to the website
does the shipping really add that much if you buy in the thousands?
 

MrSalts

Joined Apr 2, 2020
2,767
If he is planning on making 5000 pieces then he's come to the wrong place for advice.
Why is that? Have you never made 5000 pieces? Everyone has to start somewhere for guidance. Watch and see, he'll get some direction here from someone.
 

MrChips

Joined Oct 2, 2009
30,806
Why is that? Have you never made 5000 pieces? Everyone has to start somewhere for guidance. Watch and see, he'll get some direction here from someone.
1) He doesn't know how to select an MCU.
2) He likely doesn't know how to program an MCU.
3) He doesn't realize how much it costs to develop an MCU application for commercial use.
4) He doesn't know what it takes to manufacture 5000 pieces of anything.
 

MrSalts

Joined Apr 2, 2020
2,767
Perhaps it’s possible just to use a super cheap MCU that uses higher voltage and have a separate boost converter chip. I saw this tutorial that talks about using a 555 timer as a boost converter:

https://theorycircuit.com/boost-converter-circuit-555/

I think they are only a few cents. Do you think that could be used to step up the voltage to power an MCU? Also, in the video for the ATtiny43U (
) it is mentioned that a motor can be powered directly from the ICs pins. If we don’t need to be so stringent on the voltage, are there any lower cost MCU that would also allow for this? Is that quite common?

thanks again for your time and help
If you want really low cost, Chinese-made stuff, then look for Holtek - you can get a an 8-bit MCU with a 12-bit ADC for 16¢ at the volume you are looking at. The question becomes, what kind of sound will be input that the microcontroller has to find the specific frequencies? Like a whistle or flute? Or an orchestra? Also, how will you hide the noise of the motor from the input microphone to avoid interference?

if you'll have a wide array of sounds and looking for volume of a certain frequency, you'll need some audio signal conditioning as hardware, software or a DSP chip. If it is a flute or whistle or timing fork, then you'lljust need a microphone and possibly an amplifier (depending on distance of sound from microphone and type of microphone).
 

MrSalts

Joined Apr 2, 2020
2,767
1) He doesn't know how to select an MCU.
2) He likely doesn't know how to program an MCU.
3) He doesn't realize how much it costs to develop an MCU application for commercial use.
4) He doesn't know what it takes to manufacture 5000 pieces of anything.
And, someone from the site could offer to help him, the questions you listed arrive here every week except the 5000 piece issue but that is no different from ordering one assemble PCB from a ship in China. PCBway.com will love a project like this, probably under $3000 including etched boards, assembly and microcontroller programming. I can point him to a prototype company and a programmer as well. Do you have any suggestions?
 

Thread Starter

salvus

Joined Apr 2, 2020
49
I don’t have much experience with MCUs. I have dabbled a bit in the past. Definitely no expert or regular user. Fairly competent programer. If needed, 100s of hours to learn and develop it would be totally acceptable. The sound will be a tone of a set frequency. Ideally, I do want it at a frequency beyond typical human hearing. Do you feel that would be unrealistic for such a low cost device without a DSP chip? I had in mind that it could run a FFT. Regarding using a boost converter and a lower cost chip, do you think that is feasible?
 
Last edited:
Top