Hi,

I need help designing a circuit that is intended to drive a 30Vpp 10MHz AC output. I am attempting to use a basic non-inverting opamp setup with a closed-loop gain of 6 (can be easily adjusted) to amplify an input from a function generator. With my setup, the gain drops off at frequencies above 10kHz. This is nowhere near the frequency I need. At frequencies > 1MHz, there is no signal being read at all.

As I'm not too familiar with opamps, I'm not sure if I am missing something in the specifications of the opamp, or need to have something specific in my circuit setup? I tried to pick an opamp with a high Gain Bandwith product. Are there any opamps that can be recommended, or if there is a better (simple) way to produce such an output without an opamp?

The opamp I am using is the LT1365CN (link to data sheet).

Thanks for any help.

 

You are looking for 942 volts per microsecond of slew rate. I found one that might work for $66.75 but you will have to wait for it to arrive because none are in stock at www.mouser.com
http://www.mouser.com/Semiconductor...zaZ1yzmm0v&Rl=6j73mZgjdhptZ1yvcyfxZ1yv9hj3SGT

 

Wendy's Index
[URL='http://forum.allaboutcircuits.com/showthread.php?t=11847']High Speed Op Amp Query[/URL]

 

You should be getting an output at 10MHz with that op amp although you won't get quite a gain of 6 at 10MHz (about 1.2dB down) due to the marginal gain-bandwidth of that op amp.
What feedback resistor values are you using? They should be ≤2kΩ.

Here's the simulation of an Lt1259 higher gain-bandwidth op amp.
It's a current-feedback type which is less sensitive to stray input capacitances at that high frequency.

If you want to change the gain, vary R3, not R4. R4 must be used even for a non-inverting gain of 1 for a current-feedback amp.

Be sure you have a good ground with the 0.1uF ceramic caps directly from each power pin to ground with short leads.

 

Hi,

I need help designing a circuit that is intended to drive a 30Vpp 10MHz AC output. I am attempting to use a basic non-inverting opamp setup with a closed-loop gain of 6 (can be easily adjusted) to amplify an input from a function generator. With my setup, the gain drops off at frequencies above 10kHz. This is nowhere near the frequency I need. At frequencies > 1MHz, there is no signal being read at all.

As I'm not too familiar with opamps, I'm not sure if I am missing something in the specifications of the opamp, or need to have something specific in my circuit setup? I tried to pick an opamp with a high Gain Bandwith product. Are there any opamps that can be recommended, or if there is a better (simple) way to produce such an output without an opamp?

The opamp I am using is the LT1365CN (link to data sheet).

Thanks for any help.

It's a little weak for what your trying to do, but it should do better than what you have described. Post your schematic.
Edit:
Missed the previous post.

 

You should be getting an output at 10MHz with that op amp although you won't get quite a gain of 6 at 10MHz (about 1.2dB down) due to the marginal gain-bandwidth of that op amp.
What feedback resistor values are you using? They should be ≤2kΩ.

Here's the simulation of an Lt1259 higher gain-bandwidth op amp.
It's a current-feedback type which is less sensitive to stray input capacitances at that high frequency.

If you want to change the gain, vary R3, not R4. R4 must be used even for a non-inverting gain of 1 for a current-feedback amp.

Be sure you have a good ground with the 0.1uF ceramic caps directly from each power pin to ground with short leads.

View attachment 118647

Thanks for your help. Could you explain why I should be using resistances <2k? I used higher resistances. As for the gain, I could definitely lower the gain and increase the input voltage, would that produce better results?

It's a little weak for what your trying to do, but it should do better than what you have described. Post your schematic.
Edit:
Missed the previous post.

I essentially have the exact same schematic as the post above, except with resistors of greater resistance.

You are looking for 942 volts per microsecond of slew rate. I found one that might work for $66.75 but you will have to wait for it to arrive because none are in stock at www.mouser.com
http://www.mouser.com/Semiconductor...zaZ1yzmm0v&Rl=6j73mZgjdhptZ1yvcyfxZ1yv9hj3SGT

Thanks, I will look into this if I get a new opamp.

 

Could you explain why I should be using resistances <2k? I used higher resistances.

I suspected as much, considering the low frequency response your circuit had.
At that high frequency any small stray capacitances will have a significant effect on the frequency response.
For example 10pF stray capacitance at a node (not an unusual value) with a 10kΩ resistor will have a -3dB rolloff of 1.6MHz.

I found one that might work for $66.75...
Thanks, I will look into this if I get a new opamp.

The one I used in my simulation in post#4, which has better specs, is only a few US$.

 

The one I used in my simulation in post#4, which has better specs, is only a few US$.

I didn't even suspect that was the only chip that could do the job. Just, it was the only chip at Mouser that could do the job and I was too lazy to search all available vendors until we found out if this was going to be a one post member.

 

I suspected as much, considering the low frequency response your circuit had.
At that high frequency any small stray capacitances will have a significant effect on the frequency response.
For example 10pF stray capacitance at a node (not an unusual value) with a 10kΩ resistor will have a -3dB rolloff of 1.6MHz.
The one I used in my simulation in post#4, which has better specs, is only a few US$.

I see, that makes sense. I'll test the circuit with lower resistances then and see how that works out.

In theory my opamp should work at high frequencies, just that there might better options (like the one you mentioned), right?

I didn't even suspect that was the only chip that could do the job. Just, it was the only chip at Mouser that could do the job and I was too lazy to search all available vendors until we found out if this was going to be a one post member.

Makes sense, no one likes having their time wasted.

 

In theory my opamp should work at high frequencies, just that there might better options (like the one you mentioned), right?

Yes, it should work, but as I mentioned, the response will be down by about 1.2dB at 10MHz for a gain of 6.

 

Hello,

The attached table will contain opamps with a GBW greater than 50 Mhz.
(just remove the .txt and open it in MS excel or Open Office calc).

Bertus

 

Hi,

I made the recommended modifications to the circuit (<2k resistances) and tested it, and I'm happy to say that it works much better - at 10Mhz I am getting >25V with a gain of 2. Thanks so much for the help!

The only thing i noticed is that the opamp heats up very quickly, is this normal of high-speed opamps?

 

The only thing i noticed is that the opamp heats up very quickly, is this normal of high-speed opamps?

Probably.
Generating a high frequency at such a voltage requires a fair amount of power to charge and discharge all the stray capacitance.
Does the op amp get warm with no signal?

 

Probably.
Generating a high frequency at such a voltage requires a fair amount of power to charge and discharge all the stray capacitance.
Does the op amp get warm with no signal?

No, leaving it powered with no input signal does not cause it to heat up.

 

Can we see your schematic?

ak