Low noise amplifier design with -10dB gain?

[Ruritania]

Hi, gentlemen,

Does anyone here has the experience of designing LNAs with negative gain like -10 ~ -20dB and keeping the input impedance matched at the same time?

OR do you have any suggestions or references?

Thanks in advance.
Ruri

 

[FvM]

I don't see a reason, why a passive attenuator shouldn't fit the requirements. There's no use for a LNA with negative gain, to my opinion. VGA applications may of course include negative gain in it's gain range, but they have a positive gain LNA in front, usually.

 

[vfone]

So, you want to design a Low Noise Attenuator. I’ll think about it.

 

[RealAEL]

designing LNAs with negative gain like -10 ~ -20dB

If an amplifier gives 'gain' in the range -10 to -20dB then it's noise figure is similarly 10 to 20dB. So is that a Low Noise Amplifier?

A filter or even an attenuator would be much better.

 

[khouly]

this is nonsense , the LNA should give gain and add minimum noise.

ur talking about attenuator , not amplifier

Khouly

 

[GVVIN]

Ruritania explain your requirement, match what to what, Do you need attenuation only??

 

[Ruritania]

Thank you very much for your replies, guys. I'm sorry that I did not elaborate my requirements.

That's absolutely right that LNA should have gain and low noise generally, but for some cases, for example, with large signals input, the LNA should be able to switch low gain mode for linearity consideration. In short, with very large signal input, the LNA needs to be an attenuator, instead of amplifier, say -10dB gain or even less.

So, what I wanted is to use the same circuit with high gain mode, and keeping the input matching stuff unchanged, swith the LNA to -10dB attenuator.

Hope my description is clear enough for you this time, and thanks.
Ruri

 

[madengr]

Well you can have a cryogenic attenuator that attenuates but produces little noise. These are used in noise figure measurements and could certainly have use in fields such as radiometry where you want to increase your dynamic range without increasing noise.

 

[FvM]

Assuming, that cryogenic cooling is not available for the said application, either an attenuator operating at room temperature or a LNA bypass switch seem to be the only option. High performance instruments as spectrum analyzers are mostly using RF relays, sometimes switching diodes or MOS swtches for reduced requirements.

 

[radiohead]

If I get the picture right, I think you summed it up right. The LNA is only useful in case the received signal is small enough so that the amplifier operates in linear mode. The other options are:

1. The signal is very big. This can happen if receiver and transmitter in a wireless system for example is very high. In this case the received signal is very big, the signal to noise ratio very high and you indeed prefer to attenuate first. But you don't mind, the signal is good enough as such.
2. A disturbing signal is very big. Then it becomes trickier because it can create intermodulation on the (small) signal you're trying to receiver. That's why first of all you need a good RF rejection filter. Secondly for in-band signals that in fact handled by the standardization of the communication system. The clever guys of all those nifty IEEE committees sit together and try to find all different scenarios of the channel and hence receive strength. The receiver should be able to handle all scenarios.

Perhaps it's a wild guess, but are you working on multistandard receivers? Because that's where the fun starts: No RF rejection filter, and big unwanted out-of-band spurs can exist. The only conclusion to make then is use a linear LNA consuming a lot of power which can handle big blocking signals.

I hope this might help a little.

 

[yusun_98]

y can use a simple LNA with variable output impedence distribute voltage.

Added after 43 seconds:

that's true and i realize it.

 

[disisku_22]

hi
you can try out LNA with by pass switch, which as the gain of -6dB when its in by pass mode. check the avagotech web site for LNA with by pass

thanks
kumar

 

[kspalla]

In my view, the topology will be
LNA cascade with a range attenulator.
At higher input powers, the attetnulator will start.
We need not worry about the NF at this case as the signal is too strong when compared to the NF.

If you have further question please ask.

 

[BigBoss]

VGA=Variable Gain Amplifier will fit your requirements with an AGC loop.Otherwise, the amplifier can not understand how much signal has arrived its input and it can not see how much intermodulation will be produced.
LNA+AGC system should be considered.

 

[kspalla]

Correcting my own response above,
To my understand, range attenuator is AGC.

We need not worry about the NF at this case as the signal is too strong when compared to the NF.

Here first NF is noise figure where as second NF is noise floor.