switching reg question

[pnguin]

This is really a general electronics question. I am making a small adjustable power supply from a wall wart 12VDC, 500mA. Hooked it up to the breadboard with an Dimension engineering SWADJ3A regulator and a small (220 ohm resistor) load. Voltage stays steady as set at 5V but amps start out at 4.75A and decline slowly about .01A per second and will not hold steady. I am wondering why the amps don't stay steady. Any help would be greatly appreciated.

[Breaktru]

This is really a general electronics question. I am making a small adjustable power supply from a wall wart 12VDC, 500mA. Hooked it up to the breadboard with an Dimension engineering SWADJ3A regulator and a small (220 ohm resistor) load. Voltage stays steady as set at 5V but amps start out at 4.75A and decline slowly about .01A per second and will not hold steady. I am wondering why the amps don't stay steady. Any help would be greatly appreciated.

Your 12v supply is not sufficient to feed the SWADJ3A
Put a volt meter across V1 and watch it drop under load.
I don't know if it's a typo or not but your battery symbol is backwards.

[CraigHB]

Assuming a perfect switching regulator (100% efficient), you'd only be able to get 6W from that 12V wall wart so at 4V output, max current would be 1.5A.  So yea, that wall wart is not going to have the amperage you need to drive an atomizer.  However, according to your schematic, you're only trying to drive a 220 ohm resistor.  It should work without issue.

4.75A would not be possible anywhere in that circuit.  That would be way beyond the capability of your wall wart.  You're measuring something that is definitely not current.

[pnguin]

Here is the diagram with the ammeter on it. Had it on the 10 amp manual range. It read 4.75. Don't know if the voltage source is a typo or not, I just placed a DC voltage source from the component library on the diagram, I have not figured out how to assign an amp rating to the power source for sims yet. I am still learning the Multisim software. The data sheet for the reg is here: http://www.dimensionengineering.com/datasheets/DE-SWADJ3.pdf
Thanks guys.

[pnguin]

I must be doing something wrong with the meter as the data sheet shows max current 3A.

[Breaktru]

I must be doing something wrong with the meter as the data sheet shows max current 3A.

Don't know what kind of meter that is but typically, you would put an ammeter in series with the the circuit to read amperage.

[CraigHB]

As Breaktru mentioned, you normally you put an ammeter in series with your load to measure current. 

There are several routine ways to obtain current readings.  Aside from putting an ammeter in-line with your circuit, you can use a clamp-on type of ammeter that measures current by clamping a sensor on to a wire.  It uses inductance to sense current.  It's common for automotive and industrial, but not electronics.  Another method is to measure current with a shunt where you take a voltage reading across a known resistance then divide to find current.  This is more common for electronics.

I use a shunt routinely.  It works better than an in-line ammeter for low currents using a high resistance and for high currents using a low resistance.   It's because voltmeters are typically much more accurate than in-line ammeters when working at the upper and lower limits of the ammeter's range.

What you're showing is a current measurement using a shunt less the load that should be in series with it.  So, that would have to a be a voltmeter you're using.  If it's reading 4.75V, then current would be 21.6mA assuming that resistor is exactly 220 ohms.  That would be a reasonable value.

If that is in fact an ammeter you're connected as shown, what you're seeing is the maximum instantaneous current the regulator and wall wart can deliver until the wall wart and/or converter recognizes an overload and shuts down.  In-line ammeters have very low resistance, usually in the tens of milliohms so you can consider one just a straight wire far all intents and purposes.

[pnguin]

Sorry, total brain fart. I was measuring voltage across that point and just changed over to ammeter (even put the leads in the socket for 10A on the meter) without moving the leads from the position across the resistor. With the leads in series I get 22.72mA with Va at 5V. Which is exactly what Ohms law says it should be. I will blame it on the flu, I should go back to bed before I electrocute myself. Thanks again guys. Sorry for wasting your time.

[Breaktru]

Sorry, total brain fart. I was measuring voltage across that point and just changed over to ammeter (even put the leads in the socket for 10A on the meter) without moving the leads from the position across the resistor. With the leads in series I get 22.72mA with Va at 5V. Which is exactly what Ohms law says it should be. I will blame it on the flu, I should go back to bed before I electrocute myself. Thanks again guys. Sorry for wasting your time.

That's okay pnguin, you're not wasting my time. Glad you got it straight now.
Feel better friend

[CraigHB]

Hehe, no problem, I can't tell you how many fuses I've blown in my DMM from attempting a voltage measurement when my meter is still set on milliamps.  I always have to keep reminding myself to change the connection and setting when going back and forth between the voltmeter and ammeter.  It's easy to do, leave the meter set up for current then forget when you go to take a voltage reading. 

[Mimms]

How did ya make out pnguin? Did ya get it working?

[pnguin]

The range is not wide enough for a power supply so I am going with the LM317 linear reg. Got a large heat sink from the north bridge on an old computer to heat sink it and a fan for the enclosure. Running off 110 AC so no worries about battery life. I will let you all know when I get it together. Range should be 1V to 22V. Might add a separate booster to get 16V to 30V on a second circuit.

[Breaktru]

I use the LM317 for my micro-controller power supply.
Note: it only puts out 1.5 amps

[pnguin]

The 317 is all I can get locally. I may just order the LM 138 or 338 from mouser. They go to 5 amps. The ATX puts out 17 amps on the combined 12V rails, but I have not found an adjustable reg that will go past 8 amps and those are $45. I guess that is why a high powered lab supply is so expensive.

[CraigHB]

You do know that your regulator does not have to meet the maximum current output of your supply, right?  The regulator only needs to support the maximum load and and 5 Amps would be more than enough.  3 Amps is probably adequate.  That's 18W at 6V. 

One issue you will likely run into using a 12V supply with a standard regulator is heat, a lot of it.  The regulator will have to dump a lot of power.  For example, if you vape at 5V with a 2.5 Ohm atomizer, that's 2A the supply has to put out.  Since the supply is 12V, that's 24 watts.  10W is dumped by the atomizer so 14W must be dumped by the regulator.  That's a lot and the regulator may shut down due to thermal limiting even if the regulator is rated for higher currents.

To step down voltage that much with the currents required for an atomizer, it's best to use a switching regulator.  Switching regulators are efficient and do not have to burn off power to step down voltage.  Instead of 2A out of the power supply, you'll have less than 1A.   Barring losses of around 10%, input power is always the same as output power for a switching regulator.  This is the big difference from a standard regulator where it's input and output current that are always the same.

[pnguin]

Yup, I ended up with a 3 amp switching regulator. Since it is going to regulate the variable voltage side of a bench top power supply for all kinds of circuit testing. I am building the control circuit into a well ventilated plastic case with its own cooling fan and a large heat sink from the north bridge of the same computer I got the ATX from. Should be plenty of airflow to dissipate the heat. It will probably be a few weeks until I find time to get it all put together. I will post some pics when done. Thanks for all the help.

[CraigHB]

I used a switching regulator (OKI-78SR-5) to build some high output USB cig lighter adapters for my car.  I gave them a good work out on the bench to make sure they would be reliable.  Even at 2A output, the adapter barely gets warm to the touch.

As an example, if you're vaping at 18W (3A @ 6V), losses in the switching regulator are going to be around 10% which is less than 2W.  That's spread out over several of the regulator's components.  That might be under a half Watt for the switch, under a half Watt for the inductor, and under a Watt for the rectifier .  That's really not that much.  You're probably using way more cooling that you need.  I bet it would work fine with no cooling at all, not even a heat sink.

[pnguin]

True. I had planned all that cooling capacity for a linear regulator. Will probably ditch the fan and just use a heatsink for sustained high voltage runs.