sending data over the dc power line

does anyone know if this circuit going to work??

Dear Manchested
Hi
What do you want to do exactly ? if you tell me a bit more explanation , perhaps i can help you .
Best Wishes
Goldsmith

Hi goldsmith,

I am trying to send 0/1 signal from the transmitter to the receiver over the DC power line, which is mean for the power source for my circuitry, to control some applications.


Thank you.

Manchested said:

Hi goldsmith,

I am trying to send 0/1 signal from the transmitter to the receiver over the DC power line, which is mean for the power source for my circuitry, to control some applications.
Thank you.

Click to expand...

Using capacitors to couple your data on-to/off-of the DC line is the correct implementation. Be careful that you can't simply send static 1's or 0's. The capacitor will only couple AC signals onto the power line, so you may need to encode your data such that you use the rising/falling edges to signify what your data is (rising edge = "1", falling edge = "0").

Also, be sure to adequately filter the DC power pins of the adjoining circuitry. By sending data across your DC bus, you're basically adding HUGE transient signals to your nice clean DC line, which could make your other devices act really strange. I'd at least add a small series inductor and large-valued shunt capacitor to your DC input lines. That will filter the AC data signals off of the power pins for the neighboring devices.

If you want to send static (steady) one or zero signals, you can send one frequency for a one and no signal for a zero. Alternately you can send two different frequencies for a one and zero. You just detect the frequency(s) with a bandpass filter(s) to determine whether you have sent a one or zero.

As enjunear noted you will likely want to isolate the AC signal from the DC with inductors and capacitors.

Thank for the replies. I have also seen another method which uses an magnetic ring to couple the signal to the power line (inductive couple?) How should I decide which method to apply? In addition, I have come across some IC for power line communication,e.g ST7537, will I be able to achieve the communication without using those chips, perhaps just with microcontrollers?

Dear Manchested
Hi
it is simple , but to obtain your desired aim , you should increase the frequency as high as possible . thus there is no problem , and with a simple coupling you can do it . ( i did such a that thing for AC power line )
Best Wishes
Goldsmith

Thank you goldensmith,
Also, can you answer to my questions on post #6?

How should I decide which method to apply?

Click to expand...

Hi again
there are some different parameters for each designer , for example , price / quality / size / precision / THD / ... etc .

and about magnetic way , i think it is good , because of good isolation but it will be a bit more expensive than the other methods.
Best Lucks
Goldsmith

Manchested said:

Thank for the replies. I have also seen another method which uses an magnetic ring to couple the signal to the power line (inductive couple?) How should I decide which method to apply? In addition, I have come across some IC for power line communication,e.g ST7537, will I be able to achieve the communication without using those chips, perhaps just with microcontrollers?

Click to expand...

Capacitive coupling would be the most direct, but will take some development and trials to get the signaling method well-defined and working well. The pre-packaged chips would likely be more expensive, but may reduce your development/tweaking time. The magnetic method would be similar to the capacitive coupling, but potentially more work that just using two caps.

You'll have to compare and contrast the design options and figure out which one best fits your needs/constraints/budget/time, etc.

I view capacitor failure as a matter of "when?". For safety
you might want your isolation by magnetic means, using a
high frequency transformer suited to your carrier frequency
and using blocking caps to keep the 50/60Hz from burning
the little transformer up. If the cap does let go, you will
hurt a winding on the outside of your isolation boundary
and not the whole system.

Otherwise you will want a very over-rated capacitor
(compared to mains nominal peak-ground voltage) to
withstand the ugly voltages that sometimes happen.

dick_freebird said:

I view capacitor failure as a matter of "when?". For safety
you might want your isolation by magnetic means, using a
high frequency transformer suited to your carrier frequency
and using blocking caps to keep the 50/60Hz from burning
the little transformer up. If the cap does let go, you will
hurt a winding on the outside of your isolation boundary
and not the whole system.

Otherwise you will want a very over-rated capacitor
(compared to mains nominal peak-ground voltage) to
withstand the ugly voltages that sometimes happen.

Click to expand...

The original post was about trying to put AC data onto a DC power line. I'm not sure when the conversation migrated to AC mains :shock:. In the case of putting data onto AC lines, I very much agree with the "not if, but when" sentiments, and would also suggest the magnetic coupling approach. In the original DC case, I'd be fine with the OP using a simple cap due to the large margin you have from most DC voltages on a PCB, to the rated Vmax on most ceramic caps.