constant current load

[franticEB]

hi,
i have to mantain a constant sinusoidal current in RLOAD with simple circuit in figure below




but the current in RLOAD isn't sinusoidal because of the circuit itself.



How could i mantain a constant sinusoidal current in RLOAD? Which kind of circuit is more suitable for this application?

Thanks let me know.

 

[KlausST]

Hi,

You just need
* your sinusoidal signal source
* a series resistor
* your load.

Connect all in series. Finished.

No need for diodes, or transistors.

Klaus

 

[franticEB]

i need a constant current because my sinusoidal signal source amplitude could change and i have to mantain the same current in RLOAD.

 

[KlausST]

Hi,

how can a sinusoidal signal be constant?

The RMS of the sinusoidal can be constant. Including DC or excluding DC? Excluding any other frequecy (like overtones) too?
But to keep RMS constant you
* either need a constant input signal you can refer to
* or you need to measure the RMS of the input signal (or the current signal) and use an analog multiplier to regulate the magnitude of the sine.

For sure every design begins with specifications.
--> give us all your specifications and requirements at once. (please not piece by piece)


Klaus

 

[franticEB]

The rms current in RLOAD should be constant(for example 20mArms) for a input signal variable from 50 to 150Vrms and from 50hz to 20khz. The current flowing in RLOAD is mesaured by other board in order to detect the presence of the input. If that board is programmed for a 20mArms and that current becomes 30mArms because input raises, the board goes in error condition.

 

[KlausST]

Hi,

Maybe you don´t know this by now: But this really is a difficult job.

Tell us more about:
* The circuit will generate distortion. How much is allowed?
* How fast do you need to regulate?
* How precise do you need to regulate?

**
I currently don´t see a pure analog solution.
Depending on regulation speed and precision requirement....maybe you can switch different resistors to get the desired signal.

For example if you need 5% precision: then I recommend to use 2.5% steps or smaller.
For the dynamic of 50V ... 150V you need 46 logarithmic steps.

A simple - and not very percise - way to measure RMS current is to use an appropriate incandescent light bulb and regulate for constant brightness.

We definitely need more information.

Klaus

 

[erikl]

The rms current in RLOAD should be constant(for example 20mArms) for a input signal variable from 50 to 150Vrms and from 50hz to 20khz.

Can you get along with an overlaid DC current? Referred to this DC current, the relative current (or voltage) change caused by the 1:3 ac voltage change can be much smaller. Built into a transistor (nearly) constant current circuit, the current change by this a.m. voltage change related to the total current can be less than 1% :

​

Unfortunately, the sinusoidal output voltage or current wave form isn't quite symmetric.

If you need the ac voltage to be connected to GND, you can exchange it with the DC voltage source. But then the latter one must be floating of course.

If you don't want to handle high voltages, you can transform/divide down the ac voltage, and scale down the circuit voltages.

 

[crutschow]

Can you regulate to the average rectified value of the signal or does it need to be the RMS value?
Does the current through RLOAD need be be a low distortion version of the input voltage signal?

The current flowing in RLOAD is mesaured by other board in order to detect the presence of the input.

If the other board is just detecting the presence of the input, why does it have to be so precise and regulated?
Just use a simple circuit to detect the signal and output a 20mA current in response.

Your requirements seem excessive for what you are trying to do.