200 Volt DC read from PIC

[secretagent]

Hi, I'm trying to read 200 volts on a pic. How can I do this without a transformer? I want to keep the DC power and just read it... 0-200 needs to be 0-5 volts... And I want the ADC to pick this up... If I put the 200 volts to the VREF of the chip would it burn it? Ive seen 12 volt on it.... And is there any way to bring 200 volts dc to 5 volts dc without a transformer?

 

[Sceadwian]

A resistor divider will do just fine for basics, but you better have some good protection circuitry in there for transients or you'll smoke the MCU fast.

 

[secretagent]

The power im using is from a coil gun and there is a massive amount of current from large amounts of massive capacitors...

How would i lower it without resistance to ground?

 

[Gayan Soyza]

Try this one.Use Trim pot for R2, better use a 5K Trim.

 

[secretagent]

Is that AC in??? Im using DC in... If its not AC in do I put ground on the bottom line?

 

[Gayan Soyza]

That is DC in

 

[Mosaic]

Hmm.. a coil gun produce a massive back emf in the coil due to projectile movement after the initial pulse current. Your voltage divider should be designed to limit the current based on a potential KiloVolt level back emf pulse. Then you need a robust Transorb to clamp the voltage spike to the PICs supply voltage level.

 

[secretagent]

Mosaic I didnt understand one word of that !!!! But a voltage devider wouldnt work because it would put a load on the capcitors while its trying to charge.........

 

[Nigel Goodwin]

Mosaic I didnt understand one word of that !!!! But a voltage devider wouldnt work because it would put a load on the capcitors while its trying to charge.........

The load is miniscule, and the least of your worries.

 

[پروژه های الکترو]

Somewhere I saw that for measuring high dc voltages, somebody have used an optocoupler.
I think that's a good idea because it isolates the low voltage part from high voltage part.
But notice that biasing resistors must be chosen carefully to drive transistor in linear(not saturated) region.
Also it must be biased as a common-collector amplifier, not commonly used common-emitter amp.

 

[Nigel Goodwin]

Somewhere I saw that for measuring high dc voltages, somebody have used an optocoupler.
I think that's a good idea because it isolates the low voltage part from high voltage part.
But notice that biasing resistors must be chosen carefully to drive transistor in linear(not saturated) region.
Also it must be biased as a common-collector amplifier, not commonly used common-emitter amp.

Sorry, but a pretty useless idea - opto-couplers aren't linear and are no good for that purpose.

What you do is have the measuring done at the live side, and then transfer digital data over the opto-coupler.

 

[پروژه های الکترو]

Sorry, but a pretty useless idea - opto-couplers aren't linear and are no good for that purpose.

What you do is have the measuring done at the live side, and then transfer digital data over the opto-coupler.

I know that optos are non-linear but it is the art of electronic designer to use proper parts for linear measurement.

PN: What's your idea about **broken link removed** and **broken link removed**?

 

[alec_t]

Isolation is a good idea, but the opto diode may need more current than a simple voltage divider

 

[پروژه های الکترو]

I have simulated a common optocoupler in LTspice.
Note that this is just a simulation test and in reality some other parts may be added(for example a protection diode and ...).
But I think for an input of 0-200V the output is highly linear.
Also input current is less than 250uA.

**broken link removed**

**broken link removed**

 

[Diver300]

Opto-isolators are very non-linear and temperature dependent in real life.

The "linear opto isolators" are opto-isolators with a whole lot of extra electronics to make them linear. They can digitise the signal, or convert to a time, or they can have two identical opto-isolators in one IC, so that a known current's effect can be compared with an unknown current's effect.

 

[secretagent]

Somewhere I saw that for measuring high dc voltages, somebody have used an optocoupler.
I think that's a good idea because it isolates the low voltage part from high voltage part.
But notice that biasing resistors must be chosen carefully to drive transistor in linear(not saturated) region.
Also it must be biased as a common-collector amplifier, not commonly used common-emitter amp.

Yes, Optocouplers work great I just need to find one with an led that can support 200 volts!

 

[ronsimpson]

Yes, Optocouplers work great I just need to find one with an led that can support 200 volts!

The resistor is what makes any opto support 200 volts.

 

[languer]

Read the datasheets of the optocouplers carefully. Most are not truly meant to represent analog signals. The fact that the 4n25 simulates properly in LTSpice is one thing; but in real life you may be fighting all sorts of device parameter drifts for many reasons. My experience is totally biased 'cause we've used such devices almost exclusively to isolate digital signals from one system to another, or to detect the presence of high voltages; but not measure them. For linear measurement of high voltage signals with isolation, we've used transformers or isolated ADCs (e.g. https://www.analog.com/en/interface/digital-isolators/ad7400/products/product.html). As for optocouplers, never tried this one (https://www.avagotech.com/pages/en/...ic_high_linearity_analog_optocoupler/hcnr201/) but we've looked at using it a few times now.