Powering a low power transmitter issues & voltage levels

Thread Starter

seanstevens

Joined Sep 22, 2009
251
I am playing around with design of a ISM remote control and I have a dilemma on the power supply section of it, I would be very interested in your suggestions and views.
Referring to the very simplified diagram, relevant sections are as follows:
Battery is a LI-ion 18650 3V7 3A
MCU is a low power/voltage PIC operates down to 2V
TX module works at 3V7 max @ 125mA

The battery fully charged can be as high as 4V2 this is the main issue as the TX module max Vin is 3V7. I know the simplest way is to put a diode in series and drop a little across that, but I will be wasting power as the design is to have the battery last for a couple of years at least before a recharge is required. I don’t particularly need rechargeable battery but the 18650 is the only battery that fits the bill in terms of V/I, and physically being easy to remove from the unit to charge. I can’t put a socket for recharging as the IP rating would be an issue and don’t want to do wireless charging as it would complicate it unnecessarily.
At the moment I have Q1 representing a low voltage 3A high side switch that can be switched via the MCU, I could put 2 of these one with a diode in series to drop the excess and one without. The MCU is connected directly to the battery and R1/R2 feed the A/D of the PIC for voltage measurement once the battery V has dropped to 3V7 then I can switch the path to the switch without diode. Battery V is also monitored so that once it drops to 3V the switch would not be operated so that to avoid battery damage as battery would need to be charged. Operating cycle would be once a week or something like that so that battery would last for a very long time. As I couldn’t find a non-rechargeable battery that would give me 3V7 & around 2-3A I ended up with 18650.
No very efficient way, also the logic high out of the PIC would be above VCC of the TX modules. Would appreciate any thoughts on this.
 

Attachments

GopherT

Joined Nov 23, 2012
8,009
Starting at 4.2 v, a lithium ion battery will drop about 7mV per day through self-discharge. That means you'll hit 2.5v in about 235 days. A lithium rechargeable cell is a bad idea for this application if you need several years of standby time. Get a lithium metal non-rechargeable instead.

Look at figure 5 in this link, and text just before the figure. http://batteryuniversity.com/learn/article/elevating_self_discharge

Energizer Ultimate Lithium is a good choice (20-years standby).
 

GopherT

Joined Nov 23, 2012
8,009
Also, a traditional N-Channel mosfet like the IRF 530 needs a 10V gate-Source voltage. That is, the gate must be 10v higher than Source pin to achieve the specified Rds per the datasheet. Note that threashold voltage is when current first starts to flow (high resistance).

Still related, getting a gate voltage to 10v above source on a high-side switch means you need to know what the source voltage will be and add 10v to that.
 

Thread Starter

seanstevens

Joined Sep 22, 2009
251
Thanks for your input, shame I thought they were more stable than that. I have looked at Lithium batteries, Googling brings up a few if these are the right ones (see below). They all have a limited low current output, I need about 125-130mA for every use, the links below seem to have a continuous current well below my needs and a possible burst of just about my need but for only 100mS per minute. Unless I am looking at the wrong thing...

http://docs-europe.electrocomponents.com/webdocs/0709/0900766b8070972c.pdf
http://docs-europe.electrocomponents.com/webdocs/157f/0900766b8157fbd1.pdf
http://docs-europe.electrocomponents.com/webdocs/00cc/0900766b800cc0c8.pdf
http://expo.made-in-china.com/produ...thium-Thionyl-Chloride-Battery-743928145.html (65mA continuous, 130mA burst [100ms])

Do you know of any other batteries, their 3.6V is perfects though.

For the high switch I am playing around with this one (an actual high side SW with charge pump), using 3.85V straight from battery to switch it on, it only drops a few mv so it seems to be pretty good.
http://www.ti.com/lit/ds/symlink/tps27081a.pdf

Ah, if a battery can do 1-2 years that would be OK seeing it is causing problems.
 
Last edited by a moderator:

GopherT

Joined Nov 23, 2012
8,009
To make this simple, use a PNP bjt as a high side switch.

If your transmitter is on for only a burst every so often, it is not really wasting too much power. Just use a diode drop or two as needed.

Most important is making sure your microcontroler inputs are set to inputs before it sleeps between measurements to keep current as low as possible between events.

Also, the Energizer Ultimate Lithium AA size batteries (1.5V each) can output 1000 mA continuously and higher bursts.
I would use a 3-cell holder and keep it simple with diode drops.
 

Thread Starter

seanstevens

Joined Sep 22, 2009
251
Thank you GopherT for your suggestions. I will give it a try with 3 EUL AAs and see how it goes.
I use their AAA for my short range walkie talkies when I go skiing, they are great especially at low temperatures.
 

Thread Starter

seanstevens

Joined Sep 22, 2009
251
Just before this thread is closed, what are your thoughts on the CR123A Lithium batteries for this project. They are 3V and up to 1500mA and have good burst capabilities, I am thinking of 2 in series uses smaller space the AA or even AAA Lithium?
 

BobaMosfet

Joined Jul 1, 2009
2,110
Just before this thread is closed, what are your thoughts on the CR123A Lithium batteries for this project. They are 3V and up to 1500mA and have good burst capabilities, I am thinking of 2 in series uses smaller space the AA or even AAA Lithium?
Look for a datasheet from the manufacturer. All battery makers have datasheets for every battery they make, and these are a wealth of information. Normally, as with any datasheet, understand what you are reading-- oems write datasheets to put their component in the best light possible, in a competitive market. Therefore, the values they show as 'test' are usually the true actual values you want to go with, not the maximums.
 

GopherT

Joined Nov 23, 2012
8,009
Look for a datasheet from the manufacturer. All battery makers have datasheets for every battery they make, and these are a wealth of information. Normally, as with any datasheet, understand what you are reading-- oems write datasheets to put their component in the best light possible, in a competitive market. Therefore, the values they show as 'test' are usually the true actual values you want to go with, not the maximums.
They go about 60 hours at 100 mA. Not bad. Spec'ed at 200 and 500 mA too but MUCH less lifetime. Surge capacity will likely meet his needs.
 

Thread Starter

seanstevens

Joined Sep 22, 2009
251
Yes, I have been going through the CR123A data sheet and test data from several manufacturers, they look good. Worth a shot.
Thank you both.
 

sghioto

Joined Dec 31, 2017
5,381
Just before this thread is closed, what are your thoughts on the CR123A Lithium batteries for this project. They are 3V and up to 1500mA and have good burst capabilities, I am thinking of 2 in series uses smaller space the AA or even AAA Lithium?
Why use two CR123's? Use just one, then no voltage problems exist.
Steve G
 

Thread Starter

seanstevens

Joined Sep 22, 2009
251
Yes looking at the datasheet of the CR123A cant see why 1 wouldn't be enough. I will start with one and see how that goes for my app calculations show well over a year and as you say no voltage issues.
 
Top