can anyone tell me why the waveform come out like this way not dc voltage

Thread Starter

Benengineer

Joined Feb 6, 2016
133
I have an original circuits as follows:
View attachment 100838
Figure 1 original circuit

View attachment 100839
Figure 2 12V supply circuit
The circuit in Figure 2 provide 12 V to LM258D amp. Input DIMI_L and DIMI_N are from flyback converter output as follows.
View attachment 100845
Figure 3 flyback circuit provides power input for the circuit in Figure 2

When I measured the pin 7 output at its original circuit in Figure 1, its output is dc voltage around 10.9V.
After I modified the circuits, which Q1 was replaced with 47K resistor and I put diode between pin 7 and the base of Q3. The output was as follows.

View attachment 100842
Figure 4 Modified circuit

Here is the modified circuit output, which is not my expectation.
View attachment 100844

Why does its output perform like this way, not dc voltage output? How does this happen? I don't know if this saturated and the top was cut since this is very big output gain. LM258D Datasheet is attached.

Thank you for your help
 
Last edited:

GopherT

Joined Nov 23, 2012
8,009
So, each valley is about 66 microSeconds apart (at 15kHz). That means, according to my calibrated eye, the valley is about 25 microsecond wide. That means 13 microSeconds going down and 13 microseconds going up.

The datasheet says the slew rate is 0.3 (min) to 0.5 (typical) volts per microsecond on this LOW POWER op amp. So, let's take 13 microseconds @ 0.4 Volts per microsecond, you get about 5.2 volts down and 5.2 volts back up in another 13 microseconds. Looking at your o-scope output, my calibrated eye is fairly accurate and the datasheet is exactly right.

You need a slower Pwm frequency or a faster op amp. Note that low power devices are usually slow speed devices.
 

AnalogKid

Joined Aug 1, 2013
10,987
Why did you replace an emitter-follower pull up device with a 47K resistor? This is a very large increase in output impedance, and introduces a phase shift and low pass filter into the second opamp's feedback loop. Also, what is the function of Q2 in this circuit?

ak
 

Thread Starter

Benengineer

Joined Feb 6, 2016
133
So, each valley is about 66 microSeconds apart (at 15kHz). That means, according to my calibrated eye, the valley is about 25 microsecond wide. That means 13 microSeconds going down and 13 microseconds going up.

The datasheet says the slew rate is 0.3 (min) to 0.5 (typical) volts per microsecond on this LOW POWER op amp. So, let's take 13 microseconds @ 0.4 Volts per microsecond, you get about 5.2 volts down and 5.2 volts back up in another 13 microseconds. Looking at your o-scope output, my calibrated eye is fairly accurate and the datasheet is exactly right.

You need a slower Pwm frequency or a faster op amp. Note that low power devices are usually slow speed devices.
Sir GopherT,
Can you have a look at my post? Based on your suggestion, I have slowed PWM. It makes noise much smaller, but the ripple is still there. Please let me know what next step to kill the ripple. Thank you.
 
Top