Searching for a module

hi 
well optocouplers are all most linear .... but that is not enough . I am working on a motor control circuit , that 's what for  i need the hardware. I am trying to implement direct torque control on 3 phase permanent magnet synchronous machine using a dsPIC microcontroller. Now i use a small development board on 24V supply from Microchip which is not exactly built for that .. and a small motor. I am planning to implement this with a bigger 380V motor anyway so it's time to bother with the hardware problems .
Direct torque control needs minimum 2 phase currents and the phase voltages or at least 2 line voltages to calculate machine state based on motor modell. Knowing machine state the control algorithm can generate the correct switching of the inverter for the next control cycle. Also speed is obtained based on a model reference adaptive system (no encoder , tahometer...)
Some papers describe a method which calculates phase voltages of the motor  based on the DC bus voltage and the actual switching pattern of the inverter so they claim that direct torque control can work only measuring the voltage on the DC bus and 2 phase currents.. but that works only on paper (and computer simulations) that i can tell for shore. In theory that could work with a 3 phase symmetrical load. But the motor does not behave like that even if the 3 windings are the same because of the magnetic flux of the rotor which interacts with the windings ....
To make it short , because the control is based on a lot of math an almost linear measuring circuit will not do the job .
Somehow i will figure out that circuit in the end .. For now i am dealing with a lack of speed control in the speed control loop even with the little motor (of course the computer simulation runs perfectly, practice says no no no.. )
PS very detailed and explained  picture 

Oh and i bought a raspbery pi . I want to make a server with it. 
And the wireless doorbell is not ok to open my gate . I think something like a CB radio for cars would be more suitable .

Hi Ivan,
I think it would be. The input voltage range is -24 to +24 (48V) so the 1.4V drop is nothing, just 2.9%.

That which I built worked find for a few years, then the black insulating tape I bound them up with began to slide. My "permanent" solution was to boil two bits of pasta tube, stuff the LDR in one end of the big pasta, use the smaller pasta tube as an adapter to hold the LED.It dried and shrank beautifully. Then black felt-tip stopped any external light.

I read about that pasta trick about 20 years ago. Cannot remember the source.

BR Harry - EA/SM0VPO

PS - had a nice day up the mountain. 1½ contacts with my 5000 mW @ 477m ASL:

Hi,
to be exact, the light intensity of LEDs is linearly dependent on the passing current. The resistor should have such value, so that the major part of voltage drop is on the resistor, not on the LED. I expect it should be bigger than 1Kohm.

VBR Ivan

I know this is old, but I once built a nice passive RF power meter using torch lamps and an old selenium photo-voltaic cell. No diodes, wide band, but not very linear.

Now, to your problem, why not use an LED with a resies resistor, shining on a photo-resistor (orp12, ETC)?.

If the LED is fed with a suitable resistor, then the light intensity will vary almost linear with voltage. The resistance of the LDR can be used in a resistive voltage divider to give you any range you like. Feed the LDR with a constant current to improve linearity.

As a matter if interest, "opto-couplers) already do this, but they are not so linear.

BR Harry - EA/SM0VPO

Ruud wrote:https://www.ebay.com/itm/Analog-Devices-5B41-06-Isolated-Wideband-Volt-Input-/142952126777

Hi,
thanks,
buying used modules is my last option . It does not bother me that is a used part, i generally avoid obsolete components .

Hi,
i'm in search of a module that can provide galvanic isolation between input and output , to measure dc voltage . Input is in range -24 V +24 V and output is 0  - 5 V . I found 5B41-06 from analog devices which i could use with a voltage divider on it's input . Unfortunately when i try to purchase it says it's obsolete. Voltage to frequency converter is not an option, need something with voltage output .
Any idea would help