Reading a High Voltage

[Jas]

Hi

I need to input a high voltage (> 500V) into an ADC of a microcontroller.
The microcontroller runs at 5V. To do this I was thinking of using
resistors to divide the voltage down (potential dividers). Is this a good
way of doing it? How do I calculate the power requirements of the resistors
if they are being used in this way?

Thanks.

 

[John Popelish]

Jas wrote:

Hi

I need to input a high voltage (> 500V) into an ADC of a microcontroller.
The microcontroller runs at 5V. To do this I was thinking of using
resistors to divide the voltage down (potential dividers). Is this a good
way of doing it? How do I calculate the power requirements of the resistors
if they are being used in this way?

Thanks.

Assuming the negative side of the 500 volts is common with the
negative side of the PIC supply, a potential divider should work. But
not only do you have to deal with the power wasted by these resistors
(V^2/R) you also have to worry about the voltage rating of the
resistor and leakage paths on the board. The PIC is pretty accurate
with any input resistance less than about 10k, so I would start with a
10k to ground and 5.1 meg to the 500 volts. Than keeps the heat down
to 500^2/ 5Meg=0.05 watts. Keep in mind that unless you also have a
precision reference for your pic to compare to this divided voltage,
the measurement will be no more accurate than the PIC supply voltage.
--
John Popelish

 

[CFoley1064]

Subject: Re: Reading a High Voltage
From: John Popelish jpopelish@rica.net
Date: 7/28/2004 4:12 PM Central Daylight Time
Message-id: <410816BF.38ADF6AD@rica.net

Jas wrote:

Hi

I need to input a high voltage (> 500V) into an ADC of a microcontroller.
The microcontroller runs at 5V. To do this I was thinking of using
resistors to divide the voltage down (potential dividers). Is this a good
way of doing it? How do I calculate the power requirements of the
resistors
if they are being used in this way?

Thanks.

Assuming the negative side of the 500 volts is common with the
negative side of the PIC supply, a potential divider should work. But
not only do you have to deal with the power wasted by these resistors
(V^2/R) you also have to worry about the voltage rating of the
resistor and leakage paths on the board. The PIC is pretty accurate
with any input resistance less than about 10k, so I would start with a
10k to ground and 5.1 meg to the 500 volts. Than keeps the heat down
to 500^2/ 5Meg=0.05 watts. Keep in mind that unless you also have a
precision reference for your pic to compare to this divided voltage,
the measurement will be no more accurate than the PIC supply voltage.
--
John Popelish

Hi, Jas. Mr. Popelish has this exactly right. I'd like to add one small piece
of advice, if I could. If it's possible your 0-500V input voltage is going to
change quickly, you also have to take into account the capacitive divider, too.
All resistors have an inherent capacitance in parallel with their resistance.
It isn't much (low pF level), but in the event of rapidly changing input
voltage, it could result in a nanosecond excursion at the input pin of a couple
hundred volts.

If I had to use a VDVR to measure high voltage with a PIC, I'd use a 470 pF cap
to GND in parallel with the 10K resistor, which should swamp out the capacitive
VDVR. It will also slow down your response time, but will give you some
improvement in input noise.

High Voltage VDVR
Voltage Input PIC Input
___
o---|___|--o----o--------o
R1 | |
.-. |
| | ---
R2| | ---
'-' | C1
| |
=== ===
GND GND

created by Andy´s ASCII-Circuit v1.24.140803 Beta www.tech-chat.de

Good luck
Chris

 

[Dr Engelbert Buxbaum]

CFoley1064 wrote:

Hi, Jas. Mr. Popelish has this exactly right. I'd like to add one small piece
of advice, if I could. If it's possible your 0-500V input voltage is going to
change quickly, you also have to take into account the capacitive divider, too.
All resistors have an inherent capacitance in parallel with their resistance.
It isn't much (low pF level), but in the event of rapidly changing input
voltage, it could result in a nanosecond excursion at the input pin of a couple
hundred volts.

If I had to use a VDVR to measure high voltage with a PIC, I'd use a 470 pF cap
to GND in parallel with the 10K resistor, which should swamp out the capacitive
VDVR. It will also slow down your response time, but will give you some
improvement in input noise.

High Voltage VDVR
Voltage Input PIC Input
___
o---|___|--o----o--------o
R1 | |
.-. |
| | ---
R2| | ---
'-' | C1
| |
=== ===
GND GND

And possibly replace R1 with several smaller resistors in series, to
prevent high voltage from leaking into the PIC. I would additionally
place protection diodes from the PIC input to ground and positive
supply, just in case.
o +Ub
|
_|__
/\ \ ZD5V6
/ \
----
|
___ ____ ____ ____ |
o---|___|--|____|--|____|--|____|--o----o-----o--o PIC input
| | |
4 x 1.275 M .-. | _|__
| | --- /\
R2| | --- / \ DUS
'-' | C1 ----
| | |
=== === ===
GND GND

This should limit the input voltage of the PIC to -0.7 to +5.6 V,
probably survivable.