Minimum amplifier input power

Hello,

Am trying to use a general purpose wideband amplifier in the S Band. This a 50 Ohms matched amplifier. When I give an input power of about 0 dBm or even -12 dBm, it amplifies with a gain of 8-9 dB. However, just for testing if I input a power lower than -30 dBm, there is no gain. In fact, there is a loss. Now, this is an HBT based amplifier.

My question is, what is the minimum threshold for an amplifier to amplify a signal? Does such a threshold exist? Is there any way to calculate that? When you bias an amplifier aren't you bringing it into the required region of operation? So it should amplify any signal once it is in the correct region since any amplifier amplifies noise as well. The noise is going to be lower than -30 dBm in any case. What is the difference that input power level makes?

Please advise.

Thanks.

For class A amplifiers, Yes. Class C amplifiers aren't biased like that - the transistor's quiescent current is zero.

Hello godfreyl...thank you for the reply.

I am using am amplifier IC, not a discrete transistor. The datasheet does not mention the class of operation. It only mentions the input voltage recommended at 5V and the current drawn would be about 80 mA. In this case then, can I safely assume that it is operating in Class C, given that it does not work below -30 dBm i.e. it turns off below -30 dBm RF input.

Thanks.

I would expect much lower idling current for class C. What chip is it? Can you give a link to the datasheet?

Hi godfreyl..

This is the link to the device.

Thanks.

Sorry, the link doesn't work. Navigation on that site is weird - seems to be cookie based.

Anyway, can you give the part number, or a direct link to the PDF?

Hello godfreyl... It is GVA-62+ from minicircuits.

Thanks.

NOTE: The device is a MMIC rather than a power block module
any biasing that may be done is internal to the device

try this one GVA-62+

you havent mentioned at what freq you are using the device ?
Also any minimum input threshhold would be specific to a given device
This input Vs output power levels are often graphed in datasheets

Dave

There is a limit to every thing and every operation. surely the amplifying ic can not work on the required low level as because it is not designed for low operations. from graphs in data sheets you can extract all such info

@ mw_rookie:
It's a mystery to me. Are you sure there wasn't a mistake in the measurement?

I didn't see any graph in the datasheet. Do you have a better datasheet or a graph you can show us?

I've used similar MMIC amps from minicircuits, and they're all biased class A, so you should not see any loss in gain as input power diminishes. I think it must be a flaw in your measurement, or the circuit construction. The fact that you're only getting 8-9dB (it's rated for ~15dB) normally also suggests something is wrong with how you're using it. You'll have to give more details.

Did you test the amplifier separately?

is there any component before the amplifier?

can you pls explain the chain of components?

I have designed X receivers and Exciters with MMIC from miniicircuts and i didnt have any issues related to low power level ,the MMIC amplifier

must provide a proper gain at lower power levels .I i didnt understand the concept of the amplfier providing only a gain of 8 to 9 dB since the

datasheet says that the Gain is 15 dB at S band frequencies and the compression point at 16 dB , you must have done sonething wrong in

the measurement !

I suggest you measure the input and output cable assemblies first before further measurement!

Also share the biasing method of the MMIC so that we can help you better!

@godfreyl
See the Attach Doc!

Thanks for the replies.

I have re-measured as per the suggestions given including the cable losses as well.

The information is as follows:

1. I am operating between 4-5 GHz. I have 100 pF as input and output blocking capacitors.

2. The supply bypass capacitors used are: 0.1 uF, 1000 pF, 100 pF.

3. The RF choke used is 10 nH with SRF at 5 GHz.

I have observed that changing the RF choke to 20 nH (with SRF at 5 GHz), the gain does not increase. I am unable to understand the same as it should be better than in the case of 10 nH. Also, the datasheet says that the gain is flat with a variation of only 0.7 dB upto 6 GHz. However, in my case, the variation almost 5 dB.

On board, I have a multiplier, filters as required apart from this component. The cable losses are also included.

Thanks.

Change the Coupling capacitor value,

Simulate the capacitor S parameter in ADS, see what you get for insertion loss.

you have to change the coupling capacitor value according to Xc =1/(2*pi*F).

the Xc is to be kept as low as possible.

that's right, 20nH has the same function as 10nH, and the gain should also be the same.

Thanks for the suggestions.

I have simulated using the S-Parameters mentioned on the website.

The simulated gain is in the region of 11-12 dB.

Also, will change the coupling capacitors on board and measure again. However, have a small doubt regarding the same. I am currently using 100 pF. That implies an impedance f about 0.3 ohms provided by the coupling capacitor. However, I have seen some app notes where the coupling capacitor impedance is in the range of 3-4 Ohms. Is there any specific reason for that? wouldn't a lesser impedance serve better? Also, I have also seen shunt capacitors added at the input side. I am not sure how they will help. Won't adding a shunt capacitor at the input reduce the amount of signal into the amplifier?

This is one of the datasheets I found with shunt capacitors:

http://www.analog.com/static/importe...ts/ADL5320.pdf

Capacitor at the location C3 is for tuning purposes!

it might help in the input impedance matching at the specified frequencies as mentioned.

but i recommend GALI- series amplifiers from mini circuits for your application they are 50 ohm matched broadband devices.

@shashy.br: Thanks for the clarification.

GVA-62 is also a broadband matched MMIC as per the datasheet.