Rectified AC/DC power supply troubleshooting

[physics_dude]

Alright, I recently had the need to build a rectified AC/DC power supply. Yes, I mean rectified, not regulated. It won't be used to power sensitive projects.

I used Adafruit's current and volt meters to read out said values.

The problem I have is that when the current meter had it's (larger) sense leads plugged in to the circuit, and more specifically the positive of the two sense leads, a short circuit is produced. The current meter reads this short circuit as well. The bug is caused by the shared ground needed by the volt meter.

Attached below is a schematic of this power supply. The circuit relies on 0 - 60v AC input from a variac + step-down transformer. This splits off to a DPDT switch and bridge rectifier. After the rectifier comes some filtering and out it goes to the two other poles of the DPDT switch. The two common poles of the switch output to the output terminals on the case.
The current meter is only to sense current when the DC output is switched on, therefore, the sense leads tap out between the rectifier and switch. The volt meter also senses directly from the rectifier and it's filtering; it works just fine on both AC and DC.
Both meters are powered by an external 5v source which unfortunately shares the same ground as the rectifier.

Does anyone have any solutions to remedy this bug without losing features?


Edit: Updated schematic here: http://forums.adafruit.com/viewtopic.php?f=8&t=35267&p=175543#p175543

[john444]

Hi 'Dude,

The problem I have is that when the current meter had it's (larger) sense leads plugged in to the circuit, and more specifically the positive of the two sense leads, a short circuit is produced.

You are correct. If the positive side of the power-supply is connected to the A-meter, the A-meters will short current to gnd.

The bug is caused by the shared ground needed by the volt meter.

It is not a "bug". It is clearly documented "The black sensing wire is tied to the black power wire internally,". The bold text is taken from the product description http://www.adafruit.com/products/574.

Does anyone have any solutions to remedy this bug without losing features?

"To do this, you'll need to make sure that you are low-side sensing. This is because the ground power wire is connected to the black current sense wire." Again, emphasis is not mine but from the product description.
This means that instead of trying to insert the A-meter sense leads in the (+) side of the power-supply, series the A-meter in the (-) [or GND] side of the power-supply.

Good Luck, John

[physics_dude]

Wow, I'm surprised I didn't catch that...

I switched the current meter's shunt leads to the negative side of the circuit and it all works swimmingly now.

Attached is a picture of the final project if you are curious.

Thanks John

[john444]

Dude,

Looks great! Nicely implemented.

Thanks for sharing the photo,
John

[htassell]

Hi John, nice work!

I just thought I should highlight to others the importance of having the mains isolation (step down in your case) transformer in your power supply.

Under no circumtances should anyone use an unisolated variac as a power supply to power anything other than mains rated devices using mains approved connections and enclosures.

A Variac is an adjustable Auto-transformer, so there is NO isolation from the mains.

This may seem OK as the neutral side of the mains should be at a very similar potential to the earth. However, from time to time people inadvertently transpose the Active & Neutral lines in a power outlet. In that case, what you think is the Neutral at the output of your Variac has the full mains voltage, 110V, with respect to earth. To make things worse, if you have set the variable tap for say 12V, you will have 110V-12V = 98V on that tap, so no matter which lead you touch, you can be electrocuted. Furthermore, in the scenario of even when Active & Neutral are normal, but you have lost your Neutral connection, all the transformer's connections except the earth lead connected to the body of the variac can electrocute you.

I would advise only playing with using variacs as low voltage power supplies if you are absolutely certain you know what you are doing, as physic.dude clearly does in this design.

[physics_dude]

In response to htassell,
Ah yes, I am quite aware of need for mains isolation. Getting a shock isn't the only thing you have to worry about either. Every soldering iron I know of is non-isolated at the tip. This will often cause an unsightly spark upon touching a live non-isolated circuit. You very well might also destroy the circuit you are working on or even the soldering iron itself.

In addition to the step done transformer, I also have two GFCI breakers on the line, just in case. You can see the one on the variac's lead in the upper right corner of the picture. The other is on the wall in which the bench's power strip is plugged into.

[physics_dude]

Here is the updated schematic for those interested in building their own.