Hi all,

I'm using an op-amp and have set it to be unity gain in order to act as a buffer for a signal. However when i observe the output in the real world, it doesn't work as expected. The output is damped very much and doesn't work the same with my microcap12 simulation software. I have attached a simulation of the op-amp, the circuit was replicated in hardware exactly the same. The input is a 0-5V PWM signal with a duty cycle of 50% and frequency of 7.14MHz.

The used op amp is the ADA4851-1.

 

I'm using an op-amp and have set it to be unity gain in order to act as a buffer for a signal.

Meet the real world. There aren't any opamps that will operate the way you want.

The output is damped very much

I assume you're referring to the slope of the rising and falling edges.

doesn't work the same with my microcap12 simulation software

Bad simulator and a user too inexperienced to know it.

 

have attached a simulation of the op-amp

Don't see that.

 

You have a 7MHz squarewave, and you choose to buffer it with £2 op-amp which does it badly, or a 30p logic gate which does it well, and you choose the op-amp?

 

The input is a 0-5V PWM signal with a duty cycle of 50% and frequency of 7.14MHz.

Did not mention what the supply voltage is but for a 5 volt supply the input signal should be limited to 2.8 volts with that opamp
according to the spec sheet.
If the supply is 5 volts the input signal could be cut in half to 2.5 volts and set the amp gain to 2.
Something like this for simulation.

 

The Voltage supply is 12V and can deliver as much current as is needed. I ran the simulation you suggested and it seems to give the same results. The result I get when observing with the oscilloscope is a dampened signal with slow rise and fall times, it swings between 2 and 3 volts.

I have knowledge of Op-Amps and am aware that they definitely do not operate as per the simulation. I did find an Op-Amp that is well within the limits of my requirements. But the output is considerably worse compared to what it should be in practice.

 

Opamps do not normally work in the RF ranges. What you're looking for may require a device with a HIGH gain bandwidth product and a VERY fast slew rate. Something like an OPA847. Like everything RF, layout is critical. I don't recommend a traditional breadboard, but rather Manhattan style dead-bug construction.

 

I have a printed circuit board designed where the op-amp is used. I’ll give the OPA847 a shot as I can just switch them. Probably need an op-amp that is a lot more qualified for my application.

 

I have a printed circuit board designed where the op-amp is used

That ADA4851 op amp is a RR device which should have sufficient bandwidth and output slew-rate to operate properly as your sim shows, so your board configuration may be the problem.
Do you have a 10nF to 100nF decoupling capacitor directly from the op amps plus supply pin to a ground plane (you need a ground plane if you don't have one)?

 

Yes I do have decoupling caps placed closely to the supply pin.

 

Yes I do have decoupling caps placed closely to the supply pin.

Should be fine but why the 1K resistor from the output to the negative input?
Normally just shorted.

 

Yes I do have decoupling caps placed closely to the supply pin.

Does the board have a ground plane?

 

Yes 2 layered pcb with shared ground plane

 

Can you show a screen shot of the actual circuit response on the oscilloscope?

Are you using a 10:1 probe on the oscilloscope?

 

Yes I am using 10:1 probes, at the moment I cannot send you a picture of the output. I will have to do that tomorrow.

 

Yes I am using 10:1 probes, at the moment I cannot send you a picture of the output. I will have to do that tomorrow.

Is the probe ground clip connected close to the op amp?

 

A bit, it is connected to the ground pin that shares the same ground plane. On the PCB.

 

If you go with OPA847 to provide unity-gain look at figure 1.
Wideband, Ultra Low-Noise, Voltage-Feedback Operational Amplifier with Shutdown datasheet (Rev. E)

You can look at the Eval board documents to configure supply components and layout and recommendations on measurement probes.
At least try to get close to these on a descent universal pcb.

 

I’ll give the OPA847 a shot as I can just switch them.

The OPA847 doesn't do unity gain in non-inverting configuration.

 

The OPA847 doesn't do unity gain in non-inverting configuration.

Where did you read that?