Archived Forum

20 mA is the most DC that can be taken from the AC supply, otherwise the AC signal will be disturbed. 

If you need more DC current than 20 mA, you would need to provide a separate supply.  The emonTx V3 has a (micro?) USB port where 5V DC power can be supplied.

But for my project the USB supply is not a solution.

I need supply all system with the AC-AC adapter,  (microcontroller) (GSM modem) (relays) etc

If i use a better AC-AC transformer (with more current), and a full bridge rectifier, it is possible to make it?

My big problem is the ground connections. lock this schematic, i can has the grounds connected this way?

I think no, but i don't know how to solve this problem.

Vertix: There is no need to post the same comment 6 times, and with an attachment one more time. If you looked carefully when you posted, you should have seen that your comment was queued for moderation.

You cannot use a bridge rectifier and a single transformer - you need to consider very carefully what the reference voltage for the ac measurement input signal will be if you do. I suggest that you should construct the essential part of the circuit in a simulator and see what you get out. It will not be the sine wave that you expect. If you cannot see why, redraw the circuit and then tell me what you notice about the diode connected between 3 & 4.

You have two solutions.
The first is to use a separate power supply either for your modem only, or for both the emonTx and the modem, and retain the standard ac adapter (or an equivalent transformer) for the voltage measurement.
The second is to use a transformer with two isolated secondary windings, one for measurement and the other for power.

Sorry was inadvertently.

I made this circuit and made a simulation.

I has used a half rectifier (1 diode), and in oscilloscope is my waves.

Yellow : 9V AC

Blue: 9VDC (have ripple)

Purple: 5VDC (have ripple, becausse 9DC nos stable)

Green: To ADC sampling (0/3.3V)

When request more current i have a ripple in 9VDC after the rectifier, and consequently a ripple after the 5VDC regulator.

I dont understand the real problem about the emontx current limitation, is because can overheat the 3V3 regulator and can be destroyed?

Or the current increase can make a undesired change in the sine wave (9AC,sampling)?
Or the real problem is because the 9VAC transformer dont have suficient power?

If i use a more power transformer, 12V or 24V AC the problem can be solved?

Can you explain me the real problem about this current limitation?

The emonTx V3 power supply was designed to provide only enough current to operate the emonTx fitted with an RFM12B radio module and two DS18B20 temperature sensors, using the standard ac adapter, and at the same time to provide, for the voltage monitor, an acceptable waveform having minimal disturbance. In that configuration, it achieves that aim. There was simply no requirement for more current.

If you wish to redesign a power supply capable of providing 2 A, then you are free to do so. Unfortunately, you have not made a good start. Your simulation has omitted the source impedance of the transformer, and when you include that, as you must, you will find that the current pulses as the reservoir capacitor charges will very likely have a significant effect on the fidelity of the voltage sample that is used for the power calculations, thus giving erroneous results. You must include all these parameters if you are to achieve a satisfactory design.

My advice is still that you should, for a 2A peak load, use a separate supply.

What about Using a Battery for the burst - like an online UPS.

http://www.adafruit.com/products/259 then you can use the USB as power supply and deliver 2A in a burst.'

regards

Christian