# How change circuit to get 15VDC@1A rather than [email protected]? [closed]

I have designed a power supply circuit which gives output as [email protected]... but I am in need of output as 15VDC@1A...

How to increase the current value alone? I have gone through many search in internet but it gives me the answer as "current increases as voltage decreases"... I need 15V@1A as constant output.... what is the way to achieve?!

In the image, instead of LM7812, I used LM7815 to get 15VDC...

• 9 Watts in, 15 Watts out? By magic of course. Or maybe the overunity people have a snakeoil solution for you. Commented Feb 5, 2015 at 13:08
• You need a transformer designed for atleast 18V 1.5A or better. Commented Feb 5, 2015 at 13:28
• As mentioned, if your transformer is 12V at 1A... how do you expect to get 15V at 1A? (Power in) = (power out) * (efficiency percentage). In this case, you want more power out than is put in, that's not going to happen. Currently, even with your changes, it's not optimal, as your regulator is having to do a boost mode conversion to up the 12V given to 15V, but of course, at the expense of some amperage. Refer to WhatRoughBeast says, keeping in mind that you should always use components with a bit of headroom over their max rating. Commented Feb 5, 2015 at 14:07
• POwer Out always <= Power in. | Vin x Iin >= Vout x Iout ALWAYS Commented Feb 5, 2015 at 14:25
• Since he doesn't mind abusing the transformer, his circuit may well work - sort of. Peak voltage from a rectified 12 VAC signal is about 17 volts, so you might hope for 16. Then it decays during the rest of the power half-cycle until the next peak. As I mentioned, I expect the regulator output is nowhere near constant. Transformer ratings aren't like power supply ratings - you can often get more than rated output. Of course, at some point the transformer starts smoking, but that's called a learning experience. Hopefully not a Darwin Award, but that's possible, too. Commented Feb 5, 2015 at 15:16

The question is, what don't you need to change.

1) Running the transformer at maximum current (1 A average) is doable, but not a great idea. So go to a 2 A unit. And the transformer voltage is way too low. If you have a scope, look at your output with a 0.6 A load, and I'll bet you see dips in the output which occur at a 120 Hz rate. 18 volts is a much better choice.

2) 1N4007s are, in fact, rated for 1 A. That doesn't mean it's a good idea. Get beefier diodes.

3) Your 1000 uF capacitor is marginal at 0.6 amps, and probably inadequate at 1 amp. Go with at least 2200 uF, and 4700 uF is even better. Plus, if you boost your transformer voltage, at no load you'll be right near 25 volts. Better go to a 35 volt capacitor at least.

4) With a decent input voltage and 1 amp output, your 7815 is going to dissipate more heat, and without a proper heat sink it will fry. Get a heat sink.

5) Eh, you can keep the 1 uF capacitor.

• Op is using a lm7815, 15V output, not 12v. Commented Feb 5, 2015 at 14:04
• Sorry, missed that. What I get for depending on the schematic. And it's not the first time. Edited. Commented Feb 5, 2015 at 15:02

You need the transformer to be rated at a minimum of 1.6A RMS, preferably 2A, to get 1A of DC.

A 1000uF capacitor is woefully inadequate as a filter. The p-p ripple at 100Hz will be of the order of 10 volts (won't actually be that much because the regulator will drop out). More like 10,000uF is called for.

Your input voltage is insufficient for a 15V output. You need enough to account for mains variations + ripple + diode drop + regulator dropout. 16VAC output (transformer RMS rating at full load) might be okay if you don't allow for much mains variation (-10%) and have a big filter cap.

The regulator will get hot with 1A of current. At 10% high mains and a 16V transformer you might have 10W to get rid of. That calls for a big heat sink or a fan. A switching regulator would likely be a better choice.

The 1N4007s are adequate for this job but keep the leads short and reasonably heat sunk.