0
\$\begingroup\$

I have a 10 kHz PWM (0V (LOW) to 3V3 (HIGH)) which I wish to level shift as following; when PWM is high (3V3) the output should be 0V while when PWM is low (0V) the output should be -12V. Ideally this has to be done using discreate components (NPN/PNP transistors) while providing some sort of isolation (optical) between the 3V3 and -12V rails. In terms of current capabilities, the PWM pin can source/sink 8 to 12 mA max while I need not more than 200 mA for the -12V rail. Another requirement worth mentioning are the transient requirements, looking to have a clean 0 to -12V PWM signal.

So far, I came up with this circuit, but it is not behaving as I needed. Any recommendations/tips are much apricated. ThanksCCTenter image description here

\$\endgroup\$
3
  • 1
    \$\begingroup\$ but it is not behaving as I needed - please explain why. Why do you need isolation when both sides of the opto are shown (in your diagram) connected to a common 0 volts/GND? \$\endgroup\$ Commented May 3, 2021 at 11:49
  • \$\begingroup\$ @Andyaka sorry I am new to this, how can I post and other photo of the simulation result? \$\endgroup\$ Commented May 3, 2021 at 11:52
  • \$\begingroup\$ @andyaka isolation is not critical for this, but since the PWM is comming from an FPGA, and switching a -12V rail, prevents issues, in hardware they are two differnet gnds \$\endgroup\$ Commented May 3, 2021 at 11:56

1 Answer 1

Reset to default
0
\$\begingroup\$

Maybe you just need to use a decent high speed isolator with a buffered output like this: -

enter image description here

Available from virtually everyone including RS components. Of course, if you want to do it on the "el cheapo" you'll need a much lower load impedance than 10 kΩ and the addition of a buffer driver to give you 12 volt output compatibility. The RS part costs a couple of quid. I wouldn't waste my time trying to design something that gets anywhere near to the above ACPL-4800 performance.

\$\endgroup\$
8
  • \$\begingroup\$ Alternatively you could use a digital isolator, SPI/I2C and some DAC placed on the secondary side. This assuming the PWM comes from a MCU and not some other source. \$\endgroup\$ Commented May 3, 2021 at 12:05
  • \$\begingroup\$ @andyaka I don't have that avail, I have VOH1016, do you think that this will do the job? \$\endgroup\$ Commented May 3, 2021 at 12:13
  • \$\begingroup\$ @Lundin, as i prev. mentioned, isolation is not the most importnat factor, what I mostly need is the correct output, a 0 to -12V pwm signal \$\endgroup\$ Commented May 3, 2021 at 12:14
  • \$\begingroup\$ @SJ28 yes, that device also appears suitable and is very similar to the one I posted in my answer. Just try it and see. \$\endgroup\$ Commented May 3, 2021 at 12:18
  • \$\begingroup\$ @andyaka, correct me if I'm wrong, I think that both devices can't handel -ve volatges ? \$\endgroup\$ Commented May 3, 2021 at 12:24

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.