So, if I get a boost module, then that pumps 27.2v trickle charge into my 24v circuit when attached.
Therefore, I need it to somehow prevent that. Relay? Coil supplied from the charger that disconnects the battery from the PCB when the charger is switched on?
Trouble is..... no room left on my PCB now... boo. Must be possible using discreet components.
It's easier to use a DC-DC converter to step it back down to 24V (or 18V, whatever) if you are worried about that. Don't switch the battery into and out of circuit; just leave it permanently connected, floating at 27.2V. That way it is always ready to take over the load instantly whenever you get a power failure.
Personally, though, I'm puzzled by this whole business. I find it hard to believe that the circuit was ever built in the way you describe (or at least, never tested and working). Nobody uses an 18V supply to charge a 24V battery. I also would think that a circuit board rated at 24V will almost certainly be capable of working at 27.2V.
There's more to this than meets the eye. Something isn't right. Is it possible the system never had batteries originally? And some inept person has wired them in without knowing what they're doing?
Needs 24v output. The batteries are locked away in a metal enclosure - 36 screws to get them out, and I am trying to use the existing enclosure if I can. It's designed for abuse.
I am so nearly there.... code is mostly written and the (Very large) PCB has come on well.
I will breadboard the switching/Mosfets etc to make sure they switch as I intend before committing to the PCB (not that they cost much these days anyway).
Just need to get this charging malarkey sorted. Relay seems the obvious/easy route, but that will be a hefty relay that may not fit in the box.
These are very expensive, commercially available stage pieces of equipment. The batteries are locked away inside an aircraft aluminium enclosure and not far short of bomb proof.
They have a tube of desiccant inside to control the moisture, and usually operate wirelessly over about 1km.
All I have is one of these modules that is broken. No way to control it or fix it.
But, they are 24v DC output and the charger is 18v DC. You can buy huge charging cabinets for these things that have arrays of 18v chargers built into them.
Why you would make life difficult using a charger lower than the battery... I don't know, but they did.
I could alter my wiring so the whole module is live, but in 'sleep' mode while being charged I suppose.
Got it sorted. Thanks for the advice.
It can constantly charge at 27.3v, and I have added a line to the processor to tell it that it's being charged. That puts the unit 'to sleep' while charging.
Using a booster module - bumps the voltage up to a max of 28v from the 18v. Battery spec sheet recommends 27.3v charge voltage for 12x cells (2.275v x 12).
That doesn't make sense, if the board is designed to work from a 12 cell lead acid battery then it should be OK with any voltage you can reasonably expect from a 12 cell lead acid battery, so at least 27V3. As I and others have commented, something isn't right here.
I don't think 'trickle charge' is a useful term here. You apply 2V2 (or 2V3) per cell to a lead acid battery and it takes whatever current it needs to charge up and remain charged. This method only applies to lead acid batteries and is referred to as float charging or floating the battery across the supply. It is one of the reasons I favour lead acid batteries, they are so easy to deal with.
Sorry Perry. You know you are fighting a losing battle talking to me!
My 5v regulators were max 26v input (because I had them - very old stock). I have changed those to regulators that can accept>35v, so now it is not an issue.
I have ordered one of those boost modules (dirt cheap anyway) and made space to add it on my PCB.
I will have the battery connected directly to the PCB (via a fuse and a diode for basic reverse connection protection). The charger will go through the booster up to 27.3v. That is connected to the battery via a diode as well. Might have to adjust the output up slightly to allow for the drop across the diode (that stops the battery back-feeding the boost module).
I cannot stress enough that the fuse needs to be physically close to the battery, putting it on the board is no use because it does not then protect against a short in the wires from the battery. This is about preventing a fire from starting if there is a short.
For reverse connection protection use a polarised connector.
Do not use a diode, there is no need and it will reduce the voltage by about 0V6.
Hmm OK. Can't really see where the heck I can get the fuse nearer the battery. The batteries lie in a compartment that is exactly the right size. I could possibly get an inline one on the + lead.... but it will be tight.
Why no diode? Will the boost module be OK with it's output linked to the +24v of the battery ?
Thanks for everyone's advice - especially Perry's.
I am breadboarding the whole PCB first to ensure it all actually works, so got a little wait for various components I am missing.