Maker Pro
Maker Pro

common mode rejection vs common mode voltage

J

Jamie Morken

Jan 1, 1970
0
Hi,

I was looking at the datasheet of the HCPL-2201, and it specifies
a minimum CMR value of 10kV/us at a common mode voltage of 300Volts.

When the common mode voltage is less, ie. 50volts, the CMR goes down
to something around 1kV/us.

I have a fet driver circuit that requires a minimum of 10kV/us CMR
for the opto, for a floating fet gate, would this opto work properly
under these conditions? I am not sure how to know what the common
mode voltage would be to know if I will get 1kV/us or 10kV/us CMR
or something in between.

HCPL-2201 datasheet:
http://search.digikey.com/scripts/DkSearch/dksus.dll?Detail?name=516-1649-5-ND

cheers,
Jamie
 
L

legg

Jan 1, 1970
0
Hi,

I was looking at the datasheet of the HCPL-2201, and it specifies
a minimum CMR value of 10kV/us at a common mode voltage of 300Volts.

When the common mode voltage is less, ie. 50volts, the CMR goes down
to something around 1kV/us.

I have a fet driver circuit that requires a minimum of 10kV/us CMR
for the opto, for a floating fet gate, would this opto work properly
under these conditions? I am not sure how to know what the common
mode voltage would be to know if I will get 1kV/us or 10kV/us CMR
or something in between.

HCPL-2201 datasheet:
http://search.digikey.com/scripts/DkSearch/dksus.dll?Detail?name=516-1649-5-ND

You're misreading the spec.

On page 12:

HCPL2211/2 has the higher CMRR
HCPL2201 has the lower CMRR.

For a specific part number, the CMRR varies, at a specific forward
current, depending upon whether the CM is applied when output logic is
high or low.

In one state, the applied opto current is effectively nulled by CM
current.
In the other logic state, all apparent opto reciever current is due to
dvdt effects alone.

RL
 
T

Terry Given

Jan 1, 1970
0
legg said:
You're misreading the spec.

On page 12:

HCPL2211/2 has the higher CMRR
HCPL2201 has the lower CMRR.

For a specific part number, the CMRR varies, at a specific forward
current, depending upon whether the CM is applied when output logic is
high or low.

In one state, the applied opto current is effectively nulled by CM
current.
In the other logic state, all apparent opto reciever current is due to
dvdt effects alone.

RL

I have somewhere an app note that describes how to improve the CMR of
optos. basically DONT drive the led open-ended, as when the LED is off,
the impedance preventing any current flow is very high, so dV/dt can
capacitively couple current thru to the LED side which can then flow
thru the LED turning it on..... a pair of gates driving the LED in
push-pull is a good solution; the other is a resistor to Vcc to turn it
on, and shorting the LED to turn it off. hungrier this way though.

ISTR some of the FOD series optos have fast enough prop. delays Jamie....

Cheers
Terry
 
J

Jamie Morken

Jan 1, 1970
0
Terry said:
I have somewhere an app note that describes how to improve the CMR of
optos. basically DONT drive the led open-ended, as when the LED is off,
the impedance preventing any current flow is very high, so dV/dt can
capacitively couple current thru to the LED side which can then flow
thru the LED turning it on..... a pair of gates driving the LED in
push-pull is a good solution; the other is a resistor to Vcc to turn it
on, and shorting the LED to turn it off. hungrier this way though.

I think this is the way I have it set up (resistor to VCC and transistor
to gnd):
"http://rocketresearch.nekrom.com/new/floatingfetdriver/floating fet driver.jpg"

the two wires on the left lead to a transformer secondary (floating fet
driver supply)
ISTR some of the FOD series optos have fast enough prop. delays Jamie....

Yes I've seen some of these, they are mostly 5V supply though, and would
require a small 5V regulator for each floating fet driver, nicer to have
a opto that can take up to 20V :)

cheers,
Jamie
 
J

Jamie Morken

Jan 1, 1970
0
Terry said:
I have somewhere an app note that describes how to improve the CMR of
optos. basically DONT drive the led open-ended, as when the LED is off,
the impedance preventing any current flow is very high, so dV/dt can
capacitively couple current thru to the LED side which can then flow
thru the LED turning it on..... a pair of gates driving the LED in
push-pull is a good solution; the other is a resistor to Vcc to turn it
on, and shorting the LED to turn it off. hungrier this way though.

ISTR some of the FOD series optos have fast enough prop. delays Jamie....

Here's the recommended LED drive for ultra high CMR:

"http://rocketresearch.nekrom.com/new/floatingfetdriver/high CMR optodrive (HCPL-3120).jpg"

cheers,
Jamie
 
Top