9V battery pack as power source

[Pjdd]

One unit of my current project is a wireless input device, although it will not be hand-held. It will spend most of its time in standby, drawing a few microamps except during brief button presses. Apart from tests, the button presses will occur anywhere from 2 to 20 times a day. It will be used for several days in a row and then kept in storage for a few months in between uses.

The circuit needs a regulated 5V supply and draws 90-100mA during those brief button presses. I'm thinking of using a 9V PP3 pack as the power source. Even though these packs are rather wimpy, I'm inclined to think they'll be up to the task, even Carbon-Zinc types.

I considered AA cells but a box for the unit with a compartment for the cells with a neat, easily removable cover is more of a problem than with a 9V pack.

This is the kind of situation that needs judgment based on intuition and practical experience rather than precise calculation. I'll really appreciate your informed opinions on the matter.

 

[KlausST]

Hi,

Amperes x hours = AmpereHours.
A x h = Ah.
mA x h = mAh

A VARTA 9V "High Energy" pack has about 550mAh.

***
Example:
A day has 24 hours. Let´s say standby current is 50uA, then it takes 24h x 50uA = 1.2mAh per day in standby.
If 20 button press x 10s x 100mA = 20 x 10/3600h x 100mA = 5.6mAh per day active.
in total: 1.2mAh + 5.6mAh = 6.8mAh per day.
550mAh / 6.8mA per day = about 81 days.
81 days is the expected lifetime.

Klaus

 

[Mrunal Ahirrao]

Hi,
which 5V regulator you will be using? linear regulators are known to draw current in quiescent state. So use a regulator with very low quiescent current.

 

[Pjdd]

Thanks for your interest, KlausST, but things don't always work out in practice according to calculations based on basic principles. Real-world battery life is a complex issue. 100mAh load is on the high side for a PP3. Internal resistance, recovery during rest periods, acceptable end-of-life voltage on load, etc. all play a part.

The "brief" button press I mentioned will normally last not more than a second or so each time, but the relatively high load could cause the terminal voltage of an ageing battery to drop below a usable level even if the battery's still perfectly OK with a lighter load. And I want to avoid committing myself to a specific product model. All of this is why I asked for "informed opinions" based on practical experience.

Hi,
which 5V regulator you will be using? linear regulators are known to draw current in quiescent state. So use a regulator with very low quiescent current.

Yes, I intend to do that, or make a low Iq regulator from discrete components.

 

[KlausST]

Hi,

what do you expect?

You don´t trust the standard calculations.
But don´t give information so we can calculate in detail.

It seems you don´t like values like mA, mAh, seconds and so on.

You should know that our answers can only be as good as your given informations.

****
From my experience: (I try to avoid values like mA and so on)
I developed a measurement device powered by a 9V PP3 battery.
When I sell the device it is equipped with a high quality branded battery. The users will change them with unknown batteries.
Some of the measurement devices work for more than three years with the same battery.
Some users tell the battery is empty after two months. When I check the suspend and active current consumption (both values are tested and archived before selling) there is no considerable change.
In the measurement devices there is a active_time_counter. I check it and often I see users thay claim the battery is low too early use low quality batteries but have high active_time.

Klaus

 

[Audioguru]

The mAh rating for some 9V batteries is when its voltage has dropped to almost half (4.8V). Then your "5V" regulator will not work and maybe your circuit will not work.

 

[Pjdd]

Hi,

what do you expect?

You don´t trust the standard calculations.

Hey, I meant no offence. But no, I don't trust the standard calculations. I'm as familiar with those as any other experienced designer but, depending on the circumstances, they don't always produce accurate results and may even be way off when applied to certain situations. I already gave some possible reasons in general terms for such errors. To give some examples, in real life, 1mA for one minute every hour will give a much higher total mAh capacity than, say, 50mA for half an hour every hour. A 1A load will not not give even a quarter of what a 550mAh capacity implies. I admit that these are extreme examples, but they should serve to illustrate that calculations blindly based on elementary principles don't always hold out in the real world.

But don´t give information so we can calculate in detail.

What other information should I have given? I did give details about the load and expected usage pattern. The one piece of info I omitted was the duration of each button press because there's no precise value to give, but the term "brief" should be indicative.

It seems you don´t like values like mA, mAh, seconds and so on.

You should know that our answers can only be as good as your given informations.

I clearly indicated in my opening post that I was not looking for a numerically precise answer. What I asked for was informed opinions on whether a pp3 battery is a practical solution.

My various hand-held digital instruments run for at least a couple of years of regular use on cheap Chinese 9V batteries. My analog MM draws current - a small one - from the 9V battery only when measuring high resistances and lasts for pretty much the shelf life. These observations are somewhat indicative but the load currents and duty cycles are quite different from that of my project.

On the other hand, a TV remote control may see a more similar usage pattern. But the ones I have use AA or AAA cells which have different characteristics from a 9V battery. Besides, just as important as the mAh capacity is how much the terminal voltage drops on load due to increased internal resistance as the battery ages. This is why I mentioned that the circuit needs a regulated 5V supply. Even an LDO will need at least about 5.5V input.

 

[FvM]

"Standard calculations" involve a bit more than a single mAh number. I would refer to a datasheet curve like below.

At worst case the capacity with short 100 mA pulses will be higher than for continuous 100 mA discharge so that the datasheet curve can be considered as a conservative estimation.

 

[Pjdd]

The mAh rating for some 9V batteries is when its voltage has dropped to almost half (4.8V). Then your "5V" regulator will not work and maybe your circuit will not work.

True, and the actual mAh capacity varies widely with different loads and usage patterns. That's why I didn't ask for numerical values. There are simply too many variables that cannot be precisely defined.

Someone who has used 9V batteries with a load and duty cycle comparable to my design will have a good idea of how suitable or unsuitable they are for such a purpose. That someone may have been frustrated with their short life or be pleased with how long they last.

 

[BradtheRad]

Button presses are sudden and unexpected, even after some months in storage. Yet you want immediate reliable response. I think you need to put a storage capacitor on the supply pins of the device only. The capacitor can provide instant full power. Then you can get by with a less powerful battery. Charge the capacitor quickly after a button press. The capacitor does not need to charge instantly. If you choose the proper values, then you can combine Ampere draws in a manner that also cooperates with battery capacity.

Alkaline has a lifetime of years. Does not self-discharge.

9V batteries are made to be convenient, all-purpose. It's up to our ingenuity to adapt our device to it. However now that 5V devices are popular, wouldn't it be a great million-dollar idea, if I could put the right number and type of cells together, and market a 5-5.5 V battery? I haven't seen them in the supermarket, but I think the world needs it.

 

[c_mitra]

Alkaline has a lifetime of years. Does not self-discharge.

Of course one example proves nothing, but I have used one Varta, high energy, +23% more power, made in Germany Alkaline battery in the wall clock (it uses one AA cell only) and it happily lasted one year (+/- couple of months) - about the same life as any other ordinary carbon-zinc cell.

I am also using the same Varta AAA cells (two) in my spygmomanometer (citizen make) but I do not see much greater life (compared to their price). I admit that my usage pattern of the blood pressure machine is erratic.