i am using lpc2378 arm processor. In lpc2378 Vbat pin I am connected 3.3V supply through bat60 and 3V lithium battery connected through IN4007. RTC working fine, in power down (OFF) mode VBat pin consuming 13 to 14mA and RTC updating fine. But battery will down with in half hour. In my application per day minimum 1 to 2 hour power cut will happen. Pl gives the suggestion what type of battery need to connect?
Thanks ravikiran
ry
rto
Are you sure the Vbat pin is drawing 13 mA? That is a *huge* current for an RTC. Maybe you should consider a car battery?
Rick
If it is a non rechargeable battery you need 2 diodes. An they should be switching diodes not rectifier diodes.
Huh?
For UL you *do* need a current-limiting resistor, but... why two diodes? and when is a diode not a diode?
Steve
--
When is a diode ideal? All diodes have leakage current. You want to make sure that there is a minimum of leakage into a primary cell. Leakage current into a non-rechargeable battery can cause very bad things to happen.
What exactly is the purpose of the current limiting resistor to meet UL approval? Is this to prevent the battery from exploding or to prevent the battery from dumping too much current into a short?
Rick
It's an older and now obsolete document but it describes the matter on page 10:
Have you read the erata sheet - ES_LPC2378_5.pdf. One item is titled 'Increased power consumption on Vbat when Vbat is powered before the 3.3 V supply used by rest of device'. It sounds like you have to put the battery in while the device is powered!
Peter
This document is about lithium-ion batteries, but does this also apply to lithium (coin cell?) batteries?
That you'd have to duke out with the manufacturer. Consumer grade versions often have a high source resistance but some industrial versions do not. I don't think the latter are sold in regular stores.
Huh? That sounds like a major booboo by the manufacturer. Batteries for backup, realtime or data retention purposes are always in there, so they'll always be there before mains power comes on.
Yes, thats why its an erratum. However, it looks like the problem is when the battery is first added. They say that this is due to inrush current powering the NV RAM array. But if we are talking 14mA, I'd say its latch-up.
I tried to find the spec for the battery current when powered off but could only find 45uA, which is horrid. I designed an RTC chip in. We need the battery to last >5 years. I'm using a 2377 not a 2378 but I suspect they're bond out options on the same die.
Peter
A latch-up? Wow. But yeah, a jump from tens of microamps to 14mA sure sounds like it. That is much more serious than the usual erratum. IMHO they are supposed to test for such problems before releasing the chip. It would make me a bit uncomfortable, not knowing what other issues might surface. At least they should fix it and come out with a new rev level version.
45uA is quite horrid.
Ah, I see. And yes, the reverse leakage is a factor, but I'd be looking for low reverse-leakage rather than switching diodes.
To prevent explosion.
Steve
--
one so Vcc gets to the RTC, the second so VBat gets to the RTC. So the Battery only powers the RTC and not the rest, At Vcc does not charge the battery ( Unless it is rechargeable)
D'oh! Of course. Silly me. (I've usually done this stuff slightly differently, but hey...)
But I was more interested in the switching diodes vs rectifier diodes thing. I'd use a Schottky (with low reverse leakage) - is that what you mean?
Steve
--
Have something to add? Share your thoughts — no account required.
Ask the community — no account required