Small 5V supply from AA batteries

[spacedementia87]

So I built a thing!

It runs off USB 5v at the moment but I want to make it more portable and run of AA batteries.

I have done a bit of research and seen that I need a switch mode boost supply.

I have found a couple of build tutorials, would these be ok and efficient enough?

http://www.instructables.com/id/Adjustable-Voltage-Step-up-07-55V-to-27-55V/?ALLSTEPS

https://learn.adafruit.com/minty-boost/overview

Sent from my ONE A2003 using Tapatalk

[Psi]

impossible to say without knowing how much current your "thing" draws.

[krolu]

You don't have to use SMPS in boost configuration. It all depends on how many batteries you want to use. If you join 4 AA batteries in series you will get 6V total when they full so no need for boost. You could use buck/boost controller if you want to use 4 AA and get them completely dead before your project will be out of power.

[jpanhalt]

A simple boost reulator might work for you.  The NCP1402 gives a compact device.  Sparkfun sells a pre-made version for about $6.00 (https://www.sparkfun.com/products/10968 ).   Others are on ebay and banggood.

Of course, as already mentioned, you need to know your required supply current.  The SF booster can provide 200 mA, but at that maximum, your AA batteries won't last more than a few hours. 

John

[Zero999]

You don't have to use SMPS in boost configuration. It all depends on how many batteries you want to use. If you join 4 AA batteries in series you will get 6V total when they full so no need for boost. You could use buck/boost controller if you want to use 4 AA and get them completely dead before your project will be out of power.

For how long?

The batteries won't stay at 6V for long. Even with fresh batteries, at heavy loads the voltage will drop to below 1.25V per cell immediately.

Rather than a boost converter, use six AAs and a buck converter, for maximum battery life.

[BradC]

The batteries won't stay at 6V for long. Even with fresh batteries, at heavy loads the voltage will drop to below 1.25V per cell immediately.

Batterizer to the rescue!

[spacedementia87]

impossible to say without knowing how much current your "thing" draws.

It is a small analogue clock using smd LEDs.

I haven't tested but it can't be more than 100mA

Will test it soon.

I guess I am asking a more theoretical question.

Most efficient way to provide 5V without wasting power in the power supply circuit.

Sent from my ONE A2003 using Tapatalk

[alanb]

Redesign to work within the voltage range of the chosen batteries. There is probably no need for a 5 volt supply.

[jeroen79]

What kind of "thing" did you build?
Must it have exactly 5V?
Or could it also run on 4,5V (3cells) or 6V (4 cells)?

[spacedementia87]

Redesign to work within the voltage range of the chosen batteries. There is probably no need for a 5 volt supply.

What if there is a need for 5V?

Sent from my ONE A2003 using Tapatalk

[Zero999]

Please post the schematic for the clock.

[alanb]

Redesign to work within the voltage range of the chosen batteries. There is probably no need for a 5 volt supply.

What if there is a need for 5V?

Sent from my ONE A2003 using Tapatalk

If it has to be 5V then you will need to have a regulator, either buck or boost depending on the supply voltage. In your earlier message you mentioned that you are using LEDs. This is what made me think that you could possibly redesign for the available voltage. Without seeing the circuit it's not possible to  be more specific.

[spacedementia87]

Redesign to work within the voltage range of the chosen batteries. There is probably no need for a 5 volt supply.

What if there is a need for 5V?

Sent from my ONE A2003 using Tapatalk

If it has to be 5V then you will need to have a regulator, either buck or boost depending on the supply voltage. In your earlier message you mentioned that you are using LEDs. This is what made me think that you could possibly redesign for the available voltage. Without seeing the circuit it's not possible to  be more specific.

The IC requires 5V, min 4.5v which is why I want 5V.

But I also want to learn how to build an efficient circuit to provide 5V from a lower voltage.

I am currently looking at the max1674

Sent from my ONE A2003 using Tapatalk

[spacedementia87]

Please post the schematic for the clock.

The IC has a min input voltage of 4.5V

I am looking at how to make a circuit to efficiently provide 5v from AA batteries.

I might use it for other applications like an arduino.

It is me learning and in the future I might want to build 3.3v or other.

Have you ever used the max1674?

Sent from my ONE A2003 using Tapatalk

[mariush]

I've made in the past a boost regulator using LT1307 which can take in anything between 1v and around 10v and outputs a higher voltage at a maximum of 100mA : http://cds.linear.com/docs/en/datasheet/1307fa.pdf

The datasheet is very good, explains in detail the purpose of all components and how should be chosen and even shows how you would wire everything (see figure 12) and example circuits, it's a piece of cake to make it yourself.

It's expensive compared to other regulators (being Linear part) but as it works at high frequency it can work with ceramic capacitors and small inductors and it's also available in DIP package. LT1308 is the beefier brother, which if i remember correctly can do up to 1A. 

[alank2]

I've got one of these:

http://www.proto-advantage.com/store/product_info.php?products_id=2120002

[Zero999]

Please post the schematic for the clock.

The IC has a min input voltage of 4.5V

Which IC? Do you have the part number or data sheet?

[mikeselectricstuff]

Could you just use a usb battery pack?

[spacedementia87]

Please post the schematic for the clock.

The IC has a min input voltage of 4.5V

Which IC? Do you have the part number or data sheet?

It has an NE555 timer

"Input supply voltage, 4.5 V to 16 V. (SE555 maximum is 18 V)"

But anyway, I would like to learn how to create an efficient 5v battery supply.

Sent from my ONE A2003 using Tapatalk

[tszaboo]

You use a buck-boost or a boost converter. Buck boost if your input voltage voltage is below or above 5V, boost if it is below. I suggest the Texas Instruments simple switcher family. That is like the complexity of a linear regulator + an extra inductor. If you want to learn, learn how to select the correct one. We will help.
You can also use a pre built module, sold on ebay.
This is an example:
http://www.ebay.com/itm/DC-DC-Step-Up-Boost-Power-Supply-2V-5V-to-5V-2A-Stable-Fixed-Output-/331949748204?hash=item4d49bf2fec:g:HDYAAOSw6n5Xs6Y2

I will mention, this: The NE555 is not going to be high precision. You will be told about this by a dozen other forum members. Just a heads up.

[sleemanj]

Ideally, efficiency speaking, series up batteries to sufficiently higher than you need which will still be higher than you need when "flat", then use a buck converter to go down to 5v.  8 AA cells in series will be 12v full 6.4v empty.

Thats ideally speaking, because buck is a bit more efficient than boost.  But the practicality is that efficiency of a converter depends on so many things that it really isn't going to matter what you choose, boost, or buck (or sepic, or cuk or....), you are going to get somewhere in the area of 75 - 95% efficiency.

So the answer is less to do with efficiency, and more whatever is most convenient for the capacity you require - how many cells do you want to fit in to your space and what cell holders are available to fit that space, is most likely to dictate your choice. 

[Zero999]

Please post the schematic for the clock.

The IC has a min input voltage of 4.5V

Which IC? Do you have the part number or data sheet?

It has an NE555 timer

"Input supply voltage, 4.5 V to 16 V. (SE555 maximum is 18 V)"

But anyway, I would like to learn how to create an efficient 5v battery supply.

Then use six AA cells in series.

Or you could replace the NE555 with an ICM7555 or TS555, which will work at lower voltages and use less power.

As stated above, the 555 timer has a poor tolerance and is not very stable enough to keep accurate time.

[spacedementia87]

You use a buck-boost or a boost converter. Buck boost if your input voltage voltage is below or above 5V, boost if it is below. I suggest the Texas Instruments simple switcher family. That is like the complexity of a linear regulator + an extra inductor. If you want to learn, learn how to select the correct one. We will help.
You can also use a pre built module, sold on ebay.
This is an example:
http://www.ebay.com/itm/DC-DC-Step-Up-Boost-Power-Supply-2V-5V-to-5V-2A-Stable-Fixed-Output-/331949748204?hash=item4d49bf2fec:g:HDYAAOSw6n5Xs6Y2

I will mention, this: The NE555 is not going to be high precision. You will be told about this by a dozen other forum members. Just a heads up.

Thanks for the tips, I will start to look at that range and see what is available.

Ideally, efficiency speaking, series up batteries to sufficiently higher than you need which will still be higher than you need when "flat", then use a buck converter to go down to 5v.  8 AA cells in series will be 12v full 6.4v empty.

Thats ideally speaking, because buck is a bit more efficient than boost.  But the practicality is that efficiency of a converter depends on so many things that it really isn't going to matter what you choose, boost, or buck (or sepic, or cuk or....), you are going to get somewhere in the area of 75 - 95% efficiency.

So the answer is less to do with efficiency, and more whatever is most convenient for the capacity you require - how many cells do you want to fit in to your space and what cell holders are available to fit that space, is most likely to dictate your choice.

Thanks for the tips. I will look into the possibilities.

Please post the schematic for the clock.

The IC has a min input voltage of 4.5V

Which IC? Do you have the part number or data sheet?

It has an NE555 timer

"Input supply voltage, 4.5 V to 16 V. (SE555 maximum is 18 V)"

But anyway, I would like to learn how to create an efficient 5v battery supply.

Then use six AA cells in series.

Or you could replace the NE555 with an ICM7555 or TS555, which will work at lower voltages and use less power.

As stated above, the 555 timer has a poor tolerance and is not very stable enough to keep accurate time.

But surely 6AA in series the internal resistance is starting to get pretty high. I was always taught that linking up cells in series is a pretty inefficient way of getting higher voltages.

They also won't give a level voltage. What if I have another project with a tighter working voltage range?

Sent from my ONE A2003 using Tapatalk

[Zero999]

But surely 6AA in series the internal resistance is starting to get pretty high. I was always taught that linking up cells in series is a pretty inefficient way of getting higher voltages.

They also won't give a level voltage. What if I have another project with a tighter working voltage range?

The internal resistance does rise but so does the voltage so the power and energy available still increases.

Yes, it's true that the voltage won't be regulated but a 555 timer can work at much higher voltages so that's not a problem. If you have another project with tighter voltage requirements, then perhaps you could built a voltage regulator into said project, so it no longer needs exactly 5V?

[spacedementia87]

But surely 6AA in series the internal resistance is starting to get pretty high. I was always taught that linking up cells in series is a pretty inefficient way of getting higher voltages.

They also won't give a level voltage. What if I have another project with a tighter working voltage range?

If you have another project with tighter voltage requirements, then perhaps you could built a voltage regulator into said project, so it no longer needs exactly 5V?

Ok, but hypothetically what would be an efficient way of doing that. A 7805 needs a higher input voltage and burns up excess energy as heat as it is a linear regulator.

Are the ti range of switching regulators a good choice?

Sent from my ONE A2003 using Tapatalk