Battery power circuit
fluxin
Posts: 3
I want to battery power my next little project and was wondering if I could get some help with the power circuit. It will be a simple little project that will read info off an SD card and sequence 8-10 leds using the info. I was thinking of maybe using a CR123 or perhaps even some circle lithiums but I'm not sure how much power reading the SD card takes. Thanks for any help!
Comments
A quick web search shows active power drains in the range of 40 to about 80mA. Inactive current drain is on the order of a few hundreds of microAmps.
Post Edited (Mike Green) : 2/6/2008 4:46:30 PM GMT
Depending on the way you charge them, you can limit the voltage to 4V (without that Lithium batteries are sometimes charged upto 4.2V; which has little advantage.. only a few % longer run time). I have a Schottky diode between the battery and the propeller circuit, reducing the voltage downto to 3.7 and 3.1 This uses 80% of the battery capacity (end avoids deep discharge). But even a very full battery should not do much harm..
The drawback is a varying voltage, not good when needing references as with delta-sigma modulators.
The current that can be drawn is dangeriously high of course!
3V button cells do not work well; voltage very soon drops below 3v and the Propeller generally resets.
Post Edited (deSilva) : 2/6/2008 8:25:36 PM GMT
--Chuck
You are very optimistic
The AP130 is a (very) low drop down regulator as good as anyone. It needs 4V input for 3.3V.
But you must be aware that the NOMINAL voltage printed on rechargeable or non rechargeable batteries has little to do with what you see at their output. It all depends on the LOAD. It's Ohm's law over and over...
The second issue is the current you want to draw. Lithium based photo batteries are not build for that! The inner resistance (also a NOMINAL value) is considered 100 Ohms for most of them (and for button cells as well). So you cannot expect to draw anything exceeding 40 or 50 mA out of them. Lithum batteries however have a very specific impulse characteristic: For a few hundred milli seconds you can draw A LOT more current, but it then breaks down immediately!
However there is good news. There is now a new generation of Lithium (Poly) based RECHARGEABLES with CR123, CR2032, CR2450 form factor. Note that terms as CR123 are mainly FORM FACTORs, not necesarily designating a specific product!
Their nominal output is closer to 3.6V than to 3V due to the different technology used, and they are sourcing a much higher current than the non-rechargeable photo batteries.
Of course they only hold 1/2 to 1/3 of the capacity of comparable non- rechargeables, as with all rechargeables.
And of course they are prohibitively expensive (around $10 per Wh)
Post Edited (deSilva) : 2/7/2008 7:36:33 AM GMT
It uses Pulse-Frequency-Modulation (PFM) and has therefore an a real high efficiency! The input voltage ranges from 4V to 11V
and the output current can be up to 225mA.
Saluti Joerg
Secure Digital and MMC cards run on 3.3V, not 5V, for reading or writing.
-tom
I've used low-current (LC) LEDs for quite a while now and have never experienced even one failure. I cite the little Velleman kits that use LC LEDs and run off a 9V battery for months, if not longer. I've got a Velleman kit that has been running for more than five years off a DC supply, and not one low-current LED has failed. Hmmm...
I'm pretty sure Velleman is using Kingbright LC LEDs.
Kingbright has 3mm and 5mm, 2mA Ifwd at 20mcd Red diffused 30 degree view LEDs that cost around $0.10 USD in unit-qty (per DigiKey, maybe cheaper elsewhere). For some reason the 3mm LEds are harder to find, but cost seems to be the similar/same, 3mm vs. 5mm.
Attached to this post is a SPICE model I generated for the Kingbright 5mm LC LED from the datasheet. The Kingbright part numer is WP7113LSRD, the DigiKey stock number is 754-1267-ND. The model was generated using the datasheet curves (visually scraping the data), but it is fairly-close in practice (let me know otherwise). The model includes Vfwd vs. Ifwd and Vrev vs. Irev. No AC parameters or Recovery Time parms (these latter two are not needed typically in human-eye/LED applications, and the base data for them is not provided in the Kingbright datasheet).
Recognize: IMHO good visibility (high mcd rating typ. 20mcd+) at low Ifwd (<=2mA) in individual LED form-factor, and at low cost (<=$0.1 USD in unit qty) is today really only availbe with Red LEDs. Other colors suffer in brightness significantly, e.g. 1-5mcd
Regards, David
nothig against LC LEDs! They are most probably very fine products when used according to spec. That does not always happen with me.....
I once used 220 Ohme resistors out of habit for a nice row of LC LEDs, and they went " piff, paff, pofff" I am sure it's not their fault...
Furthermore there is very little choice with LC LEDs...
I am absolutely happy with my approach of having a huge choice of high brightness LEDs available, all working extremely well with 2mA.. And if I have the power, they will work with 20 mA also well. And will NOT go "piff, paff, poff" even with 40 mA.. at least not immediately...
I could even control that by software: much power -> duty cycle up; low power -> duty cycle down. Very cool
Hmm.... so when you say LC LEDs die easily, it means you're using them at higher currents, perhaps more than 2mA. If so, I can understand why they might die prematurely. I still think everyone should have a bunch of these low-current LEDs around, just for debugging on a pin across a lot of hardware technologies. Of-course that will change with Propeller-II (as it already has with current PLD/CPLD/FGPA parts), which will have 1.2 and/or 1.8V native interface voltages.
Back on-topic, I think if you treat these LC LEDs right, they'll last as long as any other LED. Please correct me if I'm wrong.
Have to turn-in now, I'm GMT+7...
Best Regards, David
Have fun
Edit: TPS61090/1/2 have up to 1.5 A of output current
Post Edited (Ale) : 2/7/2008 8:39:32 PM GMT