I am kinda new to electronics design and I am having some trouble with the Analog Devices LT319AN chip. I am trying to do simple 250KHz PWM which worked on a lesser LM393 chip in this exact setup. I know I am probably missing something simple due to the lack of some basic knowledge, any help would be much appreciated! Anyway I have added a schematic diagram and oscilloscope pictures so you can see what I am dealing with.

enter image description here enter image description here enter image description here


Forgot the Schematic!

enter image description here

Edit #2:

Thank you @DKNguyen! Per your input I added 10k OHM resistors between +5vDC and ground to bias the input's and this is the output am now getting:


Thank you for your help!


Add ceramic decoupling capacitors across the power pins of the comparator.

Input bias currents need a DC path to flow. You can't just leave them DC floating like that.

You can normally handle this by sticking resistors from each input to gnd. But in your case, you should stick each input in the middle of a resistor divider instead to DC bias in the middle of the input common mode range the because your input is bipolar but the comparator is not (it only has a unipolar supply). But this should have been an issue with the LM393 too.


simulate this circuit – Schematic created using CircuitLab

| improve this answer | |
  • \$\begingroup\$ Is it possible to DC Bias the input with my frequency generator and just add the resistor to ground for testing? And do I do the same with my ramp input? Would this work for testing so I can use my frequency generator to bias the ac input in the middle of the ramp signal? \$\endgroup\$ – Boeggs Jan 10 at 1:39
  • \$\begingroup\$ @Boeggs Not as long as you have those DC block caps. They block DC after all, including from your frequency generator. \$\endgroup\$ – DKNguyen Jan 10 at 1:40
  • \$\begingroup\$ Ah that was a lightbulb xD I forgot DC can't pass through a capacitor like that. \$\endgroup\$ – Boeggs Jan 10 at 1:41
  • \$\begingroup\$ Thank's buddy I added a video of what the output looks like now. Thank you for the awesome explanation! \$\endgroup\$ – Boeggs Jan 10 at 1:59
  • \$\begingroup\$ @Boeggs Note your comparator does not have rail to rail input. I don't think you have enough input range with your large input amplitude and low voltage supply so that can just bias it lower than 50% to avoid clipping. You're input signal is getting clipped from the comparator's perspective. \$\endgroup\$ – DKNguyen Jan 10 at 2:02

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.