DS1307 on fire

[brid0030]

So I was attempting my first use of a DS1307 Real Time Clock, and had the most miserable failure imaginable. I was using the surface mount module, and I had made a custom circuit board for it. I want to interface the device with a PIC16F874A. I got everything set up and plugged in 5 volts, and a few seconds later the DS1307 started smoking. I figured it must be some sort of basic hardware error, so I got out the multimeter and started checking for shorts. But I couldn't find anything wrong. VCC was getting 5V, the ground checked out, no short across the crystal, etc... So what could make a DS1307 explode? One thing that I could find no information on was whether the SQW/OUT pin required a connection (I just left it floating). So I tied a 10K pullup to SQW and the thing still heated up (of course, the IC could have been permanently fried at that point).

So I figure one of two things is happening. 1) the IC was damaged to begin with or 2) I am ignorant of something fundamental about using the device.

Here are some details that will be of interest to all you DS1307 experts: I'm not using a battery; Vbat is attached directly to GND. I have 10K pullup resistors on SCL and SDA. When I powered the board up, I did not have the PIC installed, so SCL and SDA were attached to nothing but the pullups.

So am I missing something, or do I have a bad IC? I've got another factory-fresh DS1307 that I can install, but I'd like to hear some advice before I potentially trash another IC. Thanks for any help.

Search terms: DS1307, RTC, overheating, burned up

[sayzer]

...Here are some details that will be of interest to all you DS1307 experts: I'm not using a battery; Vbat is attached directly to GND. I have 10K pullup resistors on SCL and SDA. When I powered the board up, I did not have the PIC installed, so SCL and SDA were attached to nothing but the pullups....


I am no expert but you have your answer in your post already!

You can not short VBatt!

And also, you must have a battery, otherwise the chip will give you strange data.


--------------------------

[Ron Marcus]

It does say on the data sheet that Vbatt gets shorted to ground if there is no backup battery. Next idea?

[Melanie]

I use the DS1307 extensively (Gawd knows why because today it's seriously overpriced for what it is - time to look around for something better and cheaper) without problems... but I ALWAYS have a Battery (CR2032)... it seems pointless having Timekeeper hardware if you allow it to lose its settings each time you power-off. I'll confirm that it's kinda hit or miss if it'll work without a Battery and you leave VBAT floating. If you really don't need a Battery, then a couple of Resistors as a potential divider across Vss and Vdd to give you about 3v... a quote from the Datasheet...

"VBAT – Battery input for any standard 3 volt lithium cell or other energy source. Battery voltage must be held between 2.0 and 3.5 volts for proper operation. The nominal write protect trip point voltage at which access to the real time clock and user RAM is denied is set by the internal circuitry as 1.25 x VBAT nominal. A Lithium battery with 48 mAhr. or greater will back up the DS1307/DS1308 for more than 10 years in the absence of power at 25 degrees C."

[brid0030]

Thanks for the suggestions.

I was aware of the funky performance when Vbat is floating, but I thought that shorting it to ground (as per the spec sheet) would prevent that. In either case, would this cause the IC to burn up?

I really do not want a battery (the device will be off for long periods and runs on a battery already). But I suppose I should send some current to Vbat one way or another.

Any other ideas as to what could have caused the meltdown?

[mister_e]

soldered in the wrong way is my best bet OR simply you revert Vcc and GND on your PCB layout. There's no chance that VBat short to GND is the explosion cause.

SQW/OUT could be left floating.

[brid0030]

[QUOTE=mister_e]soldered in the wrong way is my best bet OR simply you revert Vcc and GND on your PCB layout. There's no chance that VBat short to GND is the explosion cause.

Those were the first two things I checked. I must have a bum IC. I pulled it from another circuit board (which was never given power), so maybe there was too much soldering going on. Thanks for the bit about SQW!

I'll post the results of a second attempt in the next few.