Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronic Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Register Log in

Operational amplifiers for RF

Status
Not open for further replies.

Ricewind

Newbie level 6
Joined
Mar 14, 2011
Messages
13
Helped
0
Reputation
0
Reaction score
0
Trophy points
1,281
Activity points
1,412
Hi,

I intend to work adding different sin waves at RF badwith (10-100 MHz) but I am not very familiar on how to select an OA for this task.

I see on the internet OAs with 100 MHz bandwith and slew rate of 750 V/us... but I guess these parameters have to be taken as absolute maximuns and thus the OA not working properly for such frequencies.

Can anybody help me on how to select properly an OA for this task?

Thank you in advance.
 

FvM

Super Moderator
Staff member
Joined
Jan 22, 2008
Messages
47,959
Helped
14,145
Reputation
28,549
Reaction score
12,831
Trophy points
1,393
Location
Bochum, Germany
Activity points
278,225
I guess these parameters have to be taken as absolute maximuns and thus the OA not working properly for such frequencies
Usual OP circuits are based on the assumption of infinite respectively sufficient high loop gain. This assumption most likely won't be true for RF OP circuits operated near the bandwidth limit. The consequence is to calculate the circuit's real transfer function based on the OP data. Non-linear, e.g. slew rate limitations should be considered as well.

You can get OPs with GBW (gain bandwidth product) of several GHz, so you are basically able to design OP circuits that work in a 100 MHZ range. You can refer to the reference circuits suggested in OP data sheets to get a first impression. It should be clearly noted, that suitable circuit layout and particularly supply bypassing are substantial for RF OP circuits. You should consider continuous ground planes a design standard.

Current feedback (CFB) OPs are an interesting variant for RF circuits. You can e.g. refer to TI THS3001 as a work horse of high speed OP circuits or it's above GHz successors.
 

Ricewind

Newbie level 6
Joined
Mar 14, 2011
Messages
13
Helped
0
Reputation
0
Reaction score
0
Trophy points
1,281
Activity points
1,412
The only problem with the current-feedback is that I need to give different gain values to the signals I want to add. It is not enough just to amplify the difference or addition of them, but also of them having different gain values.

Is there any way to do this with a current-feedback?
 

FvM

Super Moderator
Staff member
Joined
Jan 22, 2008
Messages
47,959
Helped
14,145
Reputation
28,549
Reaction score
12,831
Trophy points
1,393
Location
Bochum, Germany
Activity points
278,225
Basically yes. I don't see any restriction for summing amplifiers, it's easy to figure it out yourself. There are only a few circuits, that can't be implemented well with CFB OP, e.g. non-inverting integrators and some filter types.
 

jiripolivka

Advanced Member level 5
Joined
May 21, 2010
Messages
2,832
Helped
831
Reputation
1,660
Reaction score
802
Trophy points
1,393
Location
Santa Barbara
Activity points
22,012
Using opamps for 10-100 MHz band is now possible but to my opinion quite complex.
I would rather recommend using low-cost and easy to handle "gain block" MMICs. There is a wide selection of such integrated circuits, for ~DC up to > 10 GHz.
I have used many over this frequency range, and can recommend them.
Usually they have inputs and outputs matched to 50 Ohms, require a single DC power supply, from +3 to +6..9 V, current according to a maximum output power. Stability is not a problem. To add two or more signals , you can use a power splitter (e.g. from Mini-Circuits) with a low-loss summing, or, resistive combiners if you can allow 3-6 dB loss.
If you need connectorized devices, see e.g. ZFL-500; if you can handle some soldering, see RAM ans MAR series.
There are many other "gain-block" manufacturers: I used Agilent, RF Micro-Devices, Maxim and Analog Devices models. For "your" band, 10-100 MHz, MAX 4286 is my favorite, as well as MSA0408 by Agilent.
All are stable and much friendlier than opamps. Try some!
 

FvM

Super Moderator
Staff member
Joined
Jan 22, 2008
Messages
47,959
Helped
14,145
Reputation
28,549
Reaction score
12,831
Trophy points
1,393
Location
Bochum, Germany
Activity points
278,225
I would rather recommend using low-cost and easy to handle "gain block" MMICs.
I completely agree for classical RF applications. OPs would be mainly suggested for wideband (including DC capable) or pulse circuits, also high impedance buffers and similar specialities.
 

mtwieg

Advanced Member level 5
Joined
Jan 20, 2011
Messages
3,501
Helped
1,246
Reputation
2,498
Reaction score
1,219
Trophy points
1,393
Activity points
26,256
I would steer clear of using "high bandwidth" voltage feedback op amps, as it's often very difficult to get them to behave. I have had better luck with current feedback OA, though they take some experience to get working correctly.

Also using MMIC chips is quite easy and I recommend playing with them if you never have before. As long as you don't need certain things (like high input impedance or very low output impedance, or complex topologies), they are very useful.

For doing a variable gain amp, I would consider the option of using a fixed high gain amp, and a variable attenuator.

Also, for variable gain, there are lots of monolithic VGAs which are quite easy to use. Look up the AD8331, for example.
 

Status
Not open for further replies.

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Top