I just have a brief question here. If I wanted to construct an active bandpass filter using an opamp, what op amp parameters would be the most important? (e.g slew rate, bandwidth, input resistance etc..). Same question for a gain amplifier.

Cheers,

 

It depends on the frequency, the desired amplitude in and out and the source impedance. Once you define the spec you can search for an op amp to do the job.

 

Slew rate and gain-bandwidth product are the ones that bite you. Input impedance is always on the mind of a designer, so you rarely make mistakes in that department.

 

It depends on the frequency, the desired amplitude in and out and the source impedance. Once you define the spec you can search for an op amp to do the job.

This is the schematic of the channel that we have to build. The input is limited to under 0.65V due to the diodes. The input comes from a NI-DAQ unit which has an output resistance of 50 ohms. The active bandpass filter has a lower cutoff frequency of 282Hz and a high cut-off of 6046Hz.

This circuit is basically replicating a communications 'channel'. I'm going to have to send a modulated signal from the NI-DAQ down this channel, where it will be read and demodulated.

In this context what parameters would be the most important?

 

What is the modulated signal? Are the two diodes there for protection, or are they intended as a clipper to assure that the input to the first opamp is a squarewave? Single power supply or +/-?

With a gain of only two, AC coupling, and a hi freq cutoff of 6 KHz, the amplifiers can be normal parts. You don't need extra low noise, extra tight DC performance, or extra wide bandwidth.

I recommend the LM741.

ak

 

With your signal limited to under 1V, I'd suggest an opamp with low input offset voltage.

 

I recommend the LM741. ak

and I recommend he doesn't use it with a 0V to 5V supply voltage.

 

Thanks for the helpful advice guys.

 

I'm being picky, but that's not an active filter in the usual sense of the word (with feedback to generate a second or higher order response). It's simply a first order RC bandpass filter with an op amp buffer.

Also I would not recommend a 741 op amp as it's an ancient device with very poor characteristics for any but the most uncritical of uses.

 

Never use a 741 for anything. if you must use an older op amp, at least use something like an OP275.

 

Catfeesh, I apologize to you. I threw in the 741 comment as a humorous poke at some of the frequent responders. #12 came through as I knew he would. "Kids" these days, no sense of history.

But I stand by my original analysis. Based on your schematic, the amplifiers must have bipolar (+/-) supplies. Also, the first stage has a gain of only two so any input offset error appearing at the output is very small; and completely irrelevant because the next stage is AC coupled and removes all DC components, whether errors or input. The peak signal out of the first amp is 1.5V and out of the second amp is less, so a 741 with +/-5V supplies will not clip anything. Also, the signal bandwidth is so small that there is plenty of gain margin, feedback margin, etc. True, the 741 output stage crossover distortion isn't great, but with so much negative feedback and a low operating freq, I don't see it as an issue in this application.

It is absolutely true that just about every opamp designed after the 741 is "better" in one or more (usually many more) parameters. But that doesn't mean that they automatically improve the performance of all circuits. Also, I've been in this forum for less than a year and in that short time I've been surprised by the number of threads from people who do not have access to contemporary parts and are limited to parts from the 60's or early 70's at best. All humor aside, based on what we know about this specific application the 741 will work just fine.

ak