Help on schematic

[TheUnnamedNewbie]

So, we built up this schematic on a PCB for a project. The idea is to have a UPS-style supply for a automatic shuttle prototype on small scale. Theoretical max load is 8A peak, 4A average.

When testing, as soon as the batteries were connected, the LT1512 decided to destroy itself. The issue was clearly the current in the Vsn pin (since half of the pin decided to vaporise).

What caused this? is there something obvious I'm doing wrong here?

EDIT: I would like to add that since we revisited the datasheet, we placed a 7812 with the neccisary caps to power the Vin of the charging IC. All the other parts (the 22uF cap and the inductor on that same line as Vin) are still at 24V. Is this the issue? this is the first time I ever used a charge pump.

[PA0PBZ]

EDIT: I would like to add that since we revisited the datasheet, we placed a 7812 with the neccisary caps to power the Vin of the charging IC.

Is that because of this line in the ds: *Maximum Input Voltage = 40V – VBAT ?

I think it refers to not only Vin but also Vsw (connected via the 33uH inductor). I'm also not sure if you are actually allowed to give Vin a lower voltage than Vsw.
Maybe trying to charge a 24V battery with a 24 volt supply (for 48 volt total) is just not the thing for this chip. You could try to charge a 12V battery by changing your resistor divider just to see if the rest of your circuit is ok.

[TheUnnamedNewbie]

Yes, that was because of that line. I think we should try to run the rest of the charging circuit of of 12V aswell, since it's a charge pump, it shouldn't matter.

[ptricks]

The LT chip you used isn't the best choice for what you want to accomplish, it  is complicating what could be a very simple design as it sounds like you have enough power to charge the battery.   Most UPS I have taken apart seem to go the comparator or opamp route, often using the cheap lm324.

[TheUnnamedNewbie]

The LT chip you used isn't the best choice for what you want to accomplish, it  is complicating what could be a very simple design as it sounds like you have enough power to charge the battery.   Most UPS I have taken apart seem to go the comparator or opamp route, often using the cheap lm324.

We went for the charging IC route because we didn't have the knowledge to design our own CV/CC charger, and a lab assistant here told us we should just look into a prebuilt IC. However, turns out that charging IC's in a DIP package that can power 24V batteries are hard to find.

Since this has to work for 8 hours tomorrow at a demonstration and then never work again, we don't really have time to redesign and build an entire new schematic - so I just need to try and get what we have to work.

[Zero999]

I got a sore neck trying to read your schematic so I've flipped it and increased the contrast.

There's no V_SN pin, don't you mean V_SW? I agree, it looks like you shouldn't make V_SW higher than V_IN