Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Damaging Op-Amp

Status
Not open for further replies.

drkidd22

Member
I have the below circuit that does't work reliably. Basically the PWM duty is varied to do a few tests, but once in a while the U47 opamp goes bad, I raplace it and then it will work for maybe 3 months before going bad again. I think it has to do with the input limitation, the rail is at +5Vdc, but occationally I see the input to pin 5 of U47 go to about 5.5Vdv peak, according to datasheet input voltage is -0.3 to +0.3 from the rail. could this 5.5Vdc peak be causing the damage?

R174 is simply used as a load pulled to 5Vdc, i though about adding a series resistor to see if that helps, but wanted to see what other ideas on what's causing the issues is and other fix options you guys see.
PWMCntrl.PNG
 
At power-up, C140 is initially discharged. If Q53 is off, pin 5 could initially be pulled up to well above 5V, albeit briefly, as C140 charges.
 
The data sheet: from inputs to supplies there are 10mA diodes. (brake above 10mA)
We don't know how fast the 24V comes up or how fast it falls at power off. Either way the 0.47uF cap charges and discharges via the protection diodes in the amp. The internal diode in the MOSFET is not the best and will limit the voltage to about 1V. So it is of little use. I think if you take a screw driver and short out the 24V it might kill the amp.

I would cut pin-5 open and insert 2.4k resistance. (2k to 10k) This will limit the stress to under 10mA.
 
I just double checked this, C140 is actually not present. Below image shows pin 5 at power up and second image is running on of the tests.
I will add the series resistor as it is the easiest, if it doesn't help maybe a clamp?

scope_72.png


scope_71.png
 
What is the supply voltage for the op-amp? Is it +5V_ISO? or another 5V supply? If different that is a problem.

R120, 200 ohms on the output of the amp, ….. high current on the output of the amp.

I do not understand what this amp is doing. I think you are trying to measure the average voltage ….. but......
 
It is a separate power supply for the opamp rail, both the +5V_ISO and 5V for the op-amp are derived from the 24VISO.
I agree that the 200Ohm resistor adds high current.
Basically just trying to measure current load on R174 at different duty cycles.
 
Now I am thinking the two 5V supplies go up/ down slightly differently. There is a 500 ohm 2W resistor that connect the two through the amp input. If you had 5V and 4V for a mS = trouble. Add 1K to the input.
 
For best protection, put a 10k ohm resistor in series with pin 5, as suggested, along with a small Schottky diode (e.g. BAT54) from pin 5 to the op amp plus supply voltage.
 
I've added the series resistor (R256), but it looks like it just made things a little worse. Now I'm thinking I should remove the +5V_ISO from R174 and connect it to +5V_REF. Below is updated schematic with a little more of the circuit. R120 only sees ~10mA, so load on U47 is not huge, I couldn't find on data sheet what's the max load, only short circuit is specified as 15mA.


admeas.PNG
 
This is what happens at power up. Yellow is +5V_ISO and green is +5V_REF.
+5V_REF is 'always' on.

scope_74.png


Below is power off.

scope_75.png
 
Last edited:
Any time +5V_ISO higher (150mV) than +5V_REF there could be 10mA in R174 15 ohms 2W. I think there is about 1V difference at one point.
1578068354762.png

I think you should add 5.7k to 10k. It is possible for +5V_REF to not come up at all. (manufacturing error) That will kill the amp!
1578068744121.png
 
It's not shown, but is anything with significant inductance connected to the FET drain?
 
I've added the series resistor (R256), but it looks like it just made things a little worse.
It should be directly at the op amp input.
 
Output of OPA4344 rail to rail op-amp +5.0 volts. Drop across IL300 coupler, 1.25V typical, drop across R120 = 3.75 volts. Current demand = 3.75/100 = 37.5 mA. The OPA4344 is rated for a max output current of 15 mA.
 
Last edited:
Output of OPA4344 rail to rail op-amp +5.0 volts. Drop across IL300 coupler, 1.25V typical, drop across R120 = 3.75 volts. Current demand = 3.75/100 = 37.5 mA. The OPA4344 is rated for a max output current of 15 mA.

It can go as high as 5V, but I've only been measuring 9mA on R120, depends on the feedback and R152 what goes through R120.
I added the series resistor of 8K, will run like this for a while see how it goes.
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top