PDA

View Full Version : Utilizing a 3.3v LDO and a 3.7v Lipo



Javalin
05-09-2008, 01:10 AM
Good afternoon all from the sunny UK!

I am making a project which will have solely 3.3v requirements (propeller, Gps, etc)·- I'd like to run it on a ~1000ma/h single cell Lipo.

My problem is the dropout on the LDO Regulator (~<0.5v) and hence only being able to use half the capacity on the battery before it drops the LDO regulators·output.

The only three solutions I have come up with are as follows:
1)· Use a two cell battery - don't want to increase size and weight really
2)· Use a bigger battery - ditto disadvantage as above
3)· Use a DC-DC step up jobby to 5v - best option so far - and sacrifice some battery in losses.

Anybody any other ideas?· If (3) - any small, easy to use, SMT DC-DC step up's to suggest?

Many thanks all,

James

Bean
05-09-2008, 01:16 AM
James,
Depending on what current you are using you might be able to use a shunt regulator. These are basically a resistor and a zener diode.
The trick is calculating the resistor value depending on your current draw, and the getting a diode that can handle the power.

Bean.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
Did you know that 111,111,111 multiplied by 111,111,111 equals 12345678987654321 ?

www.iElectronicDesigns.com (http://www.iElectronicDesigns.com)

·

TomS
05-09-2008, 02:43 AM
James,

Look at Linear Technology's LTC3405A.· It would be a lot more efficient than a LDO or shunt reguator.

Tom

Erik Friesen
05-09-2008, 08:18 PM
I'm not sure about li-po vs li-ion but mouser has some nice li-ions that are 3.7v, but in reality they start at 4.2 volts and 3.7v is at the lower percentile of the total capacity with 3.5 volts being about the bottom.· There are some Ldos that have a .2 volt drop if you keep your consumption around 100ma or less.· You could run your gps strait off the battery.

Javalin
05-12-2008, 05:35 PM
>but mouser has some nice li-ions that are 3.7v, but in reality they start at 4.2 volts
Yeah the range is 4.1/4.2 down to about 3v per cell is the minimum you should use - so i generally switch off at 3.2 or so. With the LDO and a ~~.3v droppout it becomes prohibative.

>You could run your gps strait off the battery.
the gps would be fine - but expects a very clean signal and obv. the propeller wants 3.3v to - so a reg of some form is required.

Thanks for the ideas guys - i'll check into it....

James

Ale
05-13-2008, 02:29 AM
I think that the LiPO are a bit safer than the Li ion batteries, in any case I'd recommend you use a battery charger controller designed for these kind of batteries. The size, weight and power are really great, but I find them a bit dangerous without proper charging methods especially in multi-cell that is not this case. Remember that these cells discharge themselves quite fast.
A DC-DC converter, is a good idea for maximum efficiency, there are some in a SOT-223 package (smaller than SO-8) that only need a couple of resistors and capacitors and a small inductor. They can give you good regulated 3.3 V in a wide range of input voltages.

Javalin
05-13-2008, 02:42 AM
Hi Ale,

Yes I am using proper charge controllers, etc.....

yes I've been looking and a good DC-DC or maybe a good switching-step-down might be the way forward....

james

Phil Pilgrim (PhiPi)
05-13-2008, 03:46 AM
A half volt scarcely qualifies as "low dropout" these days. There are linear regulators with MOS pass transistors that spec dropouts in the tens of mV range. Try Googling ultra-low-dropout or go to www.micrel.com (http://www.micrel.com) to see if you can find some better specs. Also, there's no reason you have to run the Prop at exactly 3.3V. The datasheet specs the Vdd range as 2.7V - 3.6V, so a 3.0V regulator like the Micrel MIC5301-3.0YD5 (40mV dropout @ 150mA) may fit your requirements nicely.

-Phil

Paul Baker
05-13-2008, 07:00 AM
A common solution nowadays for supplying a voltage that can be above or below the supplied voltage is using a combo regulator. The battery is fed through a step-up switching regulator, it is designed to provide a high enough voltage so that a linear regulator will never enter drop-out for the entire battery cycle (for this design probably 4-5V), then a standard linear regulator is used to obtain the actual voltage out (3.3V). There are even a few manufacturers now that are providing parts that do all of this in a single chip.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Paul Baker (mailto:pbaker@parallax.com)
Propeller Applications Engineer
[/url][url=http://www.parallax.com] (http://www.parallax.com)
Parallax, Inc. (http://www.parallax.com)

Ale
05-13-2008, 12:41 PM
I'm using one of these in one of my battery powered calculators, but for 2 NiMH cells, can be used for LiPo also, very small:

focus.ti.com/docs/prod/folders/print/tps61070.html (http://focus.ti.com/docs/prod/folders/print/tps61070.html)

Have fun.

Javalin
05-13-2008, 03:27 PM
Thanks Paul - i didn't know there was dual chips arround - i'd come to the conclusion of a step-up then a ldo reg to do the job.

Cheers all,

James

Javalin
05-13-2008, 08:47 PM
Ah - got it. Thanks Phil.

Found a 3v LDO reg with 400ma capacity and 0.2v dropout. Discovered that I can run all the board off 3v (gps, prop, adc, eeprom and micro SD)

Cool.

James

realtime
10-22-2011, 05:33 PM
What do you guys think of the
TPS62291DRVTWould it be suitable for 3.7 Lipo to regulate down to 3.3 with an evil current draw of 700-800mA ?
I wonder if the IC can deliver that kind of draw if the Lipo is at 3.3,3.4,3.5,3.6 ...V

Phil Pilgrim (PhiPi)
10-22-2011, 06:00 PM
realtime,

Welcome to the forum!

I believe Figure 22 in the datasheet answers that question for you. Of course, it also depends on the internal resistance of the LiPo battery as it's being drawn down.

-Phil

realtime
11-07-2011, 09:10 PM
Actually Fig 22 does not tell the whole story. I wanted to make sure that it would deliver about 750mA @ min. 3.3V - max. 4.2V. Any insights?


Best
RS

plainsteve
11-07-2011, 09:25 PM
Look up one figure?

Phil Pilgrim (PhiPi)
11-07-2011, 09:57 PM
Just so we're clear on the reference, this is the figure I was referring to:

http://forums.parallax.com/attachment.php?attachmentid=86665&d=1320706590

The schematic labels indicate pretty unambiguously that it will deliver 3.3V @ 1A from a 3.3V to 6V input.

-Phil

Gareth
11-07-2011, 09:57 PM
Here is a solution i have used ........tested using just a single 1.2 Volt.


http://youtu.be/wLoc_dFSMYg

Its basically a step up voltage circuit (bought cheap off e-bay)
Input Voltage: 0.9 - 5V

Output Voltage: 5.0 - 5.2V (I have attached a 3.3V regulator to the output stage .......)

Rated Output Current: 1000mA



Gareth

Kevin McCullough
11-07-2011, 11:19 PM
The device is guaranteed to deliver "up to" 1A output current. But that is hugely dependent on (limited by) ambient temperature, and the power dissipation of your device to the copper on the board. For example it can't do 1A at 80 degrees C. Also, it is a function of the input voltage too because that determines how long the high- and low-side switches are on (high-side has larger on-resistance). Also, as the (internal) junction temperature increases, the on resistance of the high-side and low-side drivers also increases a little. Have a look at Figures 17 and 18.

Does your design absolutely require 1A continuous output current? If so, I would strongly suggest stepping up to a different/higher-current device, or circuit with external switching MOSFETs. If you could get away with something a little less like 500mA, this device could be a good choice.

However... it's important to note this device is only a Buck converter - it can only step down. What you would really want for this kind of application is a Buck-boost converter. I've used these (ADP2503 and ADP2504) in previous designs with really great success. http://search.digikey.com/us/en/products/ADP2504ACPZ-3.3-R7/ADP2504ACPZ-3.3-R7TR-ND/1972290. Really, that device is just about the perfect solution for mid- to low-power battery applications. It takes in around 2.3V - 5.5V input, and delivers clean 3.3V constant output - it's almost like magic. However, just like the other device, even though it says "up to" 1000mA, you still have to de-rate it at higher temperatures due to power dissipation.

Toby Seckshund
11-09-2011, 09:09 AM
The thing that gets me is that a lot of these regulators quote the dropout voltages but fail to point out that a lot of them have a huge spike in quiesant current at the point of low input voltage. One reg I was looking at quoted a 100mA quiesant when the input/output voltage was less than a volt or two. The Buck/Boost chips, that seem to be suitable, come in QFN in varying degrees of nano-technology sizes which isn't so good for the remaining eyesight.

I often salvage the cells out of old laptop batteries (usually it is only one or two of the cells that are faulty) and usually tread the easy route of 8.4 Volts (2 cells) reg'd down to 5 Volts, and then reg'd down again to 3.3 Volts. None of which is stylish, efficient or small, ... but it's very free.

I have run a Prop straight from a single cell for a while with no failure, and often with a diode to drop off 0.7 Volt giving 3.5 - 3 Volts range. That is closer to the official figures and has that diode as a "Protection Feature".

Obviously these non official voltages are on a self risk basis.

realtime
11-10-2011, 07:14 PM
@Kevin Your exactly hit the nail on the head. I was using a 3V3, 800mA LDO Voltage Regulator first, till I realized that the ability to deliver the needed Current decreses with the decrease of V(in). Especially with a low voltage treshold of Vin 3V3 min. - 4V2 max. the LDO can't really handle this situation.

I was tempted to use the Buck IC I proposed, because I would switch off the device at approx. 3.3V anyways.

But Kevin..... the IC you found is like tha magic wand I was dreaming of.....Fig. 4 from http://www.analog.com/static/imported-files/data_sheets/ADP2503_ADP2504.pdf looks awsome. I hope ther are no other caveeats ....and its also cheaper than the TI solution .....

So long ...
T