opamp applications using the lt6203

[robismyname]

I put together a lt6203 opamp with a low pass filter on the non inverting input. My circuit is attached.

I connected a function generator on the non inverting input of the opamp and an o-scope on the opamp output (channel 1). Channel 2 of the oscope is just connected to the function generator to read the output signal going into the opamp.

It appears that my opamp is working. I say this because the o-scope is telling me the output signal of the opamp (channel 1) is much larger than the signal of the function generator (on channel 2). see png's attached.

It appears that my filter is working. I say this because when I increase the frequency of the function generator I noticed the output signal of the op-amp (channel 1 of the o-scope) gets smaller and smaller as I increase the frequency up and past 5MHz........see png's attached

Frequency O-scope-Channel 1 (Output of Opamp)
500kHz 10.4V
1MHz 10.6V
1.5MHz 7V
2MHz 4.8V
2.5MHz 3.7V
3MHz 2.3V
3.5MHz 1.48V
4MHz 1.08
4.5MHz 680mV
5MHz 460mV

There are some things I noticed that seem to make me a bit skeptical about my test setup and if I am interpreting the data correctly. They are:

1) My signal amplitude on the function generator doesn't match the reading from the oscilloscope. When I generate 5Vpp with the function generator why am I only reading a portion of the signal amplitude on the oscilloscope?

2) Why is the amplitude increasing when I increase the frequency?

Amplitude Freq Amplitude measured on O-scope

5Vpp 500kHz 132mV
5Vpp 1000kHz 250mV
5Vpp 1500kHz 372mV
5Vpp 2000kHz 494mV
5Vpp 2500kHz 600mV
5Vpp 3000kHz 748mV
5Vpp 3500kHz 828mV
5Vpp 4000kHz 928mV
5Vpp 4500kHz 1.03V
5Vpp 5000kHz 1.12V (not enough room to attach)

3) Why is the output waveform of the opamp trying to look like a square wave. Shouldn't it resemble the signal made by the function generator? They looked similar in my multisim simulation.

4) What would be the ramifications if I removed the .1uF coupling capacitor out of the circuit?

 

[LvW]

Hi robismyname,

the noninverting opamp input cannot draw any input bias current.
Thus, you have to remove the coupling capacitor - or you must provide a dc path to ground.

 

[robismyname]

Hi robismyname,

the noninverting opamp input cannot draw any input bias current.
Thus, you have to remove the coupling capacitor - or you must provide a dc path to ground.

So all of this I noticed was caused by the coupling cap?

Providing a dc path starting from where?

 

[LvW]

So all of this I noticed was caused by the coupling cap?


Yes, I think so - since the rest of the circuit looks fine.

Providing a dc path starting from where?

If you need a coupling cap there must be a high valued resistor (app. 100k..500k) from the +terminal to ground (parallel to C). But watch the dc voltage created across this resistor.

 

[FvM]

As the input is most likely ground referenced (e.g. a function generator), why did you provide an AC coupling at all?

 

[robismyname]

As the input is most likely ground referenced (e.g. a function generator), why did you provide an AC coupling at all?

Long story but I was told to do so by a superior. He also told me to replace -5V with ground. Perhaps I misunderstood what he was suggesting I do?

 

[LvW]

As the input is most likely ground referenced (e.g. a function generator), why did you provide an AC coupling at all?

Long story but I was told to do so by a superior. He also told me to replace -5V with ground. Perhaps I misunderstood what he was suggesting I do?

That´s not a good engineering behaviour. When somebody requires a specific detail in circuit design he should be able to give the reason. You should ask: why a coupling capacitor? And of course: How many uF?