Good morning afternoon night or evening depending where in the world you are.

I am having a little trouble with my small project being as it's been a while sonce I have actually built anything from scratch.

Firstly a bit about me.

I studied electronics at high school for 4 years so am reasonably handy although that was a while ago. I then studied as a electrician and now I am a telecomunications technician.

Project
I am trying to rebuild a emergency light circuit for my local sports club who I am heavily involved with. We had a very obsolete faulty 6v system which I am converting.

I am using automotive led flood lights as the club is large halls approx 5 lights at 18w each 12-24v dc

I have a transformer out of a broken ups which I have used and have made my rectification circuit.

I am planning on using 12v alarm batteries prob 4-5 in parallel but my transformer is pumping out a nice 19.2 vdc

What would be the best way to safely and constantly drop the 19v down to about 13-13.5 vdc to keep the batteries charged nicely?

I have already setup my relay etc to stop the batteries back feeding the while building when power is cut.

Just struggling to work out a good way to drop the voltage

Thanks in advance

 

What current can your transformer handle - does it have a rating?

Are your "alarm" batteries nicad or sealed lead-acid (SLA)? Between the rectified DC from the transformer and your battery, you need something that can control the charge on the battery and make sure that the charge current stays within specifications of both the transformer and the battery. That something - the charge controller - depends on the chemistry of the battery. SLA is probably easiest, since you can charge to a constant voltage.

 

What current can your transformer handle - does it have a rating?

Are your "alarm" batteries nicad or sealed lead-acid (SLA)? Between the rectified DC from the transformer and your battery, you need something that can control the charge on the battery and make sure that the charge current stays within specifications of both the transformer and the battery. That something - the charge controller - depends on the chemistry of the battery. SLA is probably easiest, since you can charge to a constant voltage.

Haven't actually brought any batteries yet so either is a option
Was thinking of using sealed lead acid as they are cheap for me

and for transformer rating I am unsure but this is what's written on it

11460KB583AL-G-220v/11221114071

It's a multi coil 196v , 225v, 252v, primary side to 15.5v center tap, 29v secondary side. It's a big beast and is out of a large UPS so should easily have the capacity for what I am after .

 

You might get some ideas by reading my thread here.

 

Thanks for the link, If i use your exact circuit will work? have lost all faith in myself when it comes to circuits as its been 10 years since i have actually build a electronic circuit.

 

use a simple LM317, set to 13.8v like this circuit

 

and a heatsink on the lm317

assumption - that transformer probably got more ampage than what the regulator can handle.

 

Is it safe to just use a lm137. Thingy? Due to me wanting to hook up 3 7.0ah batts in parallel and due to the capacity of the transformer. Won't the heat be way too high.

I also will need a perfect 12v supply to switch a relay but have a MOSFET for that and will just limit current with a resistor

 

You are going to power 5 x 18 W LED lamps that's 90 Watts. Just a quick calculation shows a drain of approx. 7.5 Amps from the battery. What is your expected emergency operation time? The LM 317 circuits will do fine, but are limited to <1.5A, therefore, I would rather go for the LM350 (3A device) or the LM338 (5A device) with similar circuitry as the LM317.


If you have more than one power outage in one evening, you might need the extra amperage capacity, otherwise the LM 317 circuits will do if you have more than 24 hours to recharge the battery.
If you do want to use the 317, here is a circuit attached below.

Seeing that you have not yet bought batteries, rather buy one (1) deep cycle battery of adequate Amp/h than opting to go for smaller ones in parallel. Batteries in parallel is not as reliable as the equivalent bigger one.

 

Power outages are very rare I think we have had 1 in the last 2 years and batteries only need to last 10mins . Idea behind using multiple batteries was if a battery failed the system would still work but for a shorter time.

I am apart of a indoor shooting club so all the lights are for is to make the firearms safe and get people outside.

I also have a second hand lm317 I found on the dead ups circuits and a few heatsinks. I might muck around tomorrow night and see what I get .

Thanks heaps guys please keep giving me ideas and info as your knowledge far exceeds mine on the matter.

I am also a summing the lm317 is a 1.5a voltage regultor

 

Neither the circuit posted by Dodgy or Johann incorporates current-limiting to protect the regulator (or the battery) that I describe in my link.

In light of your low-impedance transformer (bigger than it needs to be), I think the current-limiting is an essential part of your project. The method of current limiting shown in my link can be merged with the higher-current regulators...

 

,,,
I am also a summing the lm317 is a 1.5a voltage regultor

Practically, no. It is hard to mount the LM317 on a big enough heatsink to get 1.5A out of it. It has a built-in overtemp shutdown circuit in it, and if it self-heats, the practical maximum current depends on the voltage drop across it. The higher the transformer voltage, the lower current that the LM317 will pass...

 

I think the current-limiting is an essential part of your project.

+1 Many DIY-ers get away with using the LM317 without current limiting because the IC will shut itself down when it overheats. They seem fairly robust and may survive many such events. But it's not something you want to design in, on purpose, especially when long life reliability is a critical part of the application.

You need to decide what charging rate you want to support. At 1A, the LM317 on a big heatsink is an option. If you need 10A, say to quick-charge a big depleted battery, that calls for something completely different. Only you can decide on the specifications for your project.

 

+1 Many DIY-ers get away with using the LM317 without current limiting because the IC will shut itself down when it overheats. They seem fairly robust and may survive many such events. But it's not something you want to design in, on purpose, especially when long life reliability is a critical part of the application...

I have had mixed results with this approach (without secondary current limiting). I have lost LM317s when connecting a charger to a mostly depleted battery. I tried increasing the area of the heat sink, and that seemed to make it worse.

My conclusion is that to rely on the LM317's built-in thermal current limiting, you cant put it on a large heatsink. Increasing the thermal mass seems to allow the LM317 to blow itself up before the tab heats up and it begins limiting its own current.

On the other hand, if you use it without a heatsink, the maximum charging current to the battery is only a few hundred mA because it goes into thermal current limit early...

 

Thanks for all the input I guess I will go shopping tomorrow for some components and try build the complete circuit I need.

Do what batteries would you guys recomend ? How many ah?

 

On the other hand, if you use it without a heatsink, the maximum charging current to the battery is only a few hundred mA because it goes into thermal current limit early...

learn something knew every day

 

Thanks for all the input I guess I will go shopping tomorrow for some components and try build the complete circuit I need.

Do what batteries would you guys recomend ? How many ah?

Restate your load requirements again?

Figure that if discharging a 12V SLA to no less than 11.2V (highly recommended), you are using about 80% of its capacity, so upsize it a bit...

You might do well to use several smallish SLAs, each with its own LM317 charger...

 

Wouldn't an LM1084, work better for this charger?


http://www.mouser.com/ProductDetail...=sGAEpiMZZMsGz1a6aV8DcCERHZHPu4lvcWtzk6SyWLc=

 

Wouldn't an LM1084, work better for this charger?...

Looking at the data sheet, they show a battery charger with add-on current limiting (Rs) in Fig 27. That suggests that hooking this regulator directly to a battery (without secondary current limiting) might have the same issues as I ran into with the LM317?

The circuit in Fig 27 is not as good as the one with the current sense resistor and NPN transistor I linked to earlier. Putting Rs inside the control loop makes a poor battery charger that will take a very long time to charge a battery.

The LM1084 with a 0.22Ω sense resistor and the NPN circuit might work...

 

I do not have an LTSpice model for the LM1084, so I just used the LM317 set up with two kinds of external current limiting; one using the transistor, the other just Rs. I chose the external components to get a peak current of ~1.1A in both cases. Note how long it both circuits take to charge a 5F capacitor (simulated battery) starting from 11V...

Note that the circuit with the transistor is a much better constant-current source than the one with Rs inside the feedback loop...