Unregulated PS vs transformer + rectifier

[joecool85]

I have a "12vDC 500ma" wall power supply, but when I checked it, it comes out 18.2v, so I assume that means its unregulated. I'm building a small SLA battery charger and need 17-18vDC to come in, I was going to use a 12.6v 300ma transformer, then a bridge rectifier. Would the unregulated "12v" power supply work alright?

 

[Oznog]

It will probably "work" in that it will charge the battery at a useful rate and won't overheat the transformer. However with no voltage regulator it will overcharge the battery once the battery's fully charged. This will over time degrade the battery's capacity and ultimately shorten its life, potentially dramatically.

 

[joecool85]

I failed to mention a very important part here. Regardless of using a transformer and rectifier, or using adapter, I will be using a lm317 regulator to regulate to 13.5v

So it would be:
transformer->rectifier->regulator->battery
or
"12v" adapter->regulator->battery

 

[stevez]

You might want to estimate the current you need then put a load (bulb, power resistor, etc) on the unregulated power supply to simulate that load - measuring the voltage across the load. You may find that the unregulated supply voltage drops off rapidly. The concern is that if your unregulated supply voltage drops to some level, let's say 13 volts, you'll not have 12 volts after the 317.

 

[joecool85]

I think I'm just going to go with the transformer. I guess my question here is, will the transformer's voltage fluctate that much? Its rated for 12.6v and it really can't get below 12 or so or it won't be high enough (at least 17v) after rectification to give me enough headroom to get 13.5-14.0v out of the lm317.

 

[stevez]

You might actually have the same problem with the transformer as you might have with the unregulated supply.

The output of your transformer is dependent on the input voltage. If the input varies the output of the transformer will vary. If the incoming voltage is low then the output will be low.

If you rectify the AC output of the transformer you'll get pulsating DC. A capacitor filter, with no load, will charge to the peak voltage - that might be what you are seeing with the unregulated supply when you get 17 volts or so. With any load the voltage settles down closer to the RMS output voltage of the transformer.

I have to get going but if you don't get clearer explanations here, send me an IM with your email address and I'll see if I can be of some help. It's an interesting subject and not all that complicated.

 

[joecool85]

It can get higher than 12.6, so long as it doesn't go below 12.6, I guess thats what I was getting at.

Since the "12v" adapter is 12v at 500ma, but 18v no load, I'm assuming the 12.6 transformer will be 12.6 at 300ma (its rating) and higher unloaded. Its no problem for it to be higher though, since it will be going through a regulator anyway. So long as it stays below 25v I will be happy (thats what my caps are rated for lol).

 

[stevez]

Bear in mind that the rating is tied to an apparently unspecified input voltage. Someone knows what that input voltage is but we don't know. We might guess that it's 110 volts or 120 volts - or maybe 105 volts. What is actually runs at will be dependent on the line voltage at the time you use it.

I only bring this up because I don't know how close to the edge you might be. Unfortunately the easiest way to know what will or won't work is just to try it and see. Sure, you could do all sorts of tests and experiments but those take energy/resources too. I am lucky enough to have a variac and if I were doing this I'd put some load on the output (light bulb) then measure with varying input voltages to see where things come out. If it appeared that it was likely I'd drop below some threshold I'd abandon the use of that transformer or supply.

The advantage of thinking in thru and talking it thru is that you are more likely to spot problem areas before you've consumed resources. Of course the down side is that you feel like you aren't progressing toward a solution.

 

[joecool85]

Well, I need at least 17v after rectification so that the lm137 can regulate down to 14v. So...it really can't go below 12.2v or so.

 

[stevez]

Joe- unless the current is extremely low it sounds like you will not get what you need from the transformer.

At the nominal 12.6 vac (rms) you will get a peak voltage across a capacitor filter of 17.8 vdc with no voltage drop across the diodes. Since there is some drop I'd say that this "test" fails, so to speak. Even without voltage drop from the diodes - a relatively small amount of current will pull the capacitor down below 17 volts. They do make low dropout regulators but they still need some headroom to operate. Where the dc voltage in to the regulator will bottom out is not too tough to approximate but you are so close that approximations aren't of much value. It can be calculated but it's pretty tricky, relatively speaking.

Things you could do - voltage doubler if current requirements allow - or a switching supply. It may be better to just get a new transformer.

 

[joecool85]

If worst comes to worst I can always just use it at 13.5v which will be my main charge setting anyway. I dunno, its worth a shot. I can always swap out the transformer later for a higher voltage one.

 

[joecool85]

Looks like I'm going to go with a 20v 500ma transformer instead. A buck cheaper than the 12.6v 300ma from radioshack, and Jameco ships fast so its no biggie.

 

[neon]

if you are going to charge a battery the damn thing doesn't care about voltage one bit all it cares is current coming in to charge it. that means the voltage shell be higher then the battery fully charged volatge. next thing to consider is the current coming in if it is too much then the battery overheat and premature failure so the current shell be limited to a decent level what should be it depends type ampere hour and mostly heat dissipation. so now you may charge a 12v from any dc pulsating source. 12v are never 12 volts anyhow loke 9v is never 9 vand 1.5 is never 1.5v close but not exact.

 

[joecool85]

I know all that. 500ma should be fine to charge the 3.2ah battery. And the battery doesn't care about the voltage so long as its over its own, but it does affect charging time. Also, I've heard its good to "top off" a battery at 14.6v or so every so often to keep it working well.

 

[William At MyBlueRoom]

If memory serves you can multiply your AC input voltage by sqrt(2)

 

[audioguru]

If memory serves you can multiply your AC input voltage by sqrt(2)

Yes and no.
The square root of 2= 1.414. Then the full wave bridge rectifier is basically 2 diodes in series which drops the output voltage about 1.4V to 2.0V. The filter cap might be too small creating a ripple voltage that further reduces the output DC voltage.

 

[William At MyBlueRoom]

forgot about the voltage drops across the diodes. Seems my math is off tonight. Had a good chuckle at the 8051 555 15MHz project. seriously... Whats that all about?

 

[joecool85]

No idea, but I'm sure it has nothing to do with this.

Anyway, I plan on starting to build my SLA charger using a 20v 500ma transformer from jameco around the beginning of may. I have to finish up this semester before I can do much, lots of projects going on right now for my new media courses.