Continue to Site

Welcome to

Welcome to our site! is an international Electronics 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.


Not open for further replies.


Newbie level 6
Oct 21, 2007
Reaction score
Trophy points
Activity points
1ghz bw amplifiers with low noise

I want to design a LNA with spec. as following. Please help me. Thanks ahead.
freq: 3G~6G
gain: 20db
NF: <3db
power: as less as possible
Could anyone give me some design example?

Best wishes

low noise amplifier design project

Attached is block diagram of LNA I have design in C-band(3.4 to 4.2Ghz ) and measured results are meeting the specs mentioned in attached file you can extend its bandwidth by doing some matching.

Asim Ali
narrowband lna mosfet bjt

is this LNA is a CMOS , or u wnat it discrete level ?

check avago transistors they are working very good


two stage lna design match

yes, khouly is right. avago PHEMT transistors have googe NF performance, and enough software tools. But broadband(one octave) LNA design is a little difficult, you can imply the method of series feedback or parallel feedback.

low noise amplifier download ebook

yeah by this wide band it is possible , specially his NF performance not so critical , the pHemts of avago can provide u with .4 to .9 dB of NF at this range

so the feedback , can do a good job in this design


atf-35143 design

For your spec, I would recommend to use SiGE HBT (NPN in CE mode (2 transistors back to back)) in a simple and single ended structure.

atf-35143 phemt spice model parameter

yeah this can work , but u need to use 2 stages to get this spec's

but u didn't mention anything about the linearity of ur LNA


lna noise figure match

it's important to select the component bjt or MOSfet ,then get the S2p or spice model which can downloading for the net, then use ADS or other EDA tool to simulate, if you can cal use matlab then use ADS to simulate I think it will be better then pay some time on PCB layout ,which is the most important step in the all design process.

ephemt spice model

the PCB should be EM simulated to take all the parasiics into account , the via alone add about 0.2 nH :)


if lna amplifier

Salam Friends,

I have some questions about designing of low noise wide band amplifier at frequencies between 0.5GHz to 1.5GHz. <0.5dB Noise Figure, Gain is 25dB,s11<-10dB
I will explain a some sort of my project.

I was searched material for different manufacturers for LNA design, I have compared 35 transistors in which one of them showed Fmin is good, i.e.ATF 35143 pH EMT(reason is Fmin is flat compared to other E pHEMT as well which was linear)

first step that I need to match the LNA for minimum noise figure of the transistor,

As we know we can match the transistor in two ways, 1) Input matching 2)Noise matching.

I have started first with noise matching which has matched using lumped components(Lumped components are used cause of freq less than 2GHz) exactly at 1 GHz frequency(At single frequency), i.e. LNA noise figure nf(2) at the input side 0.12dB was obtained which is exactly to the Fmin of 1GHz frequency of ATF35143(Fmin=0.12dB)

Now I have chosen for broader frequency range which was not matched(0.5 to 1.5 GHz) except at 1GHz.

Now Noise matched values are(The values which has matched from Source to input side of the transistor, by adding Series L and Admittance Capacitance was matched at the in put side , L=23.95 nH, C=0.5188pF)

Further I have tuned the circuit by adding components, starting from the Source (Admittance Capacitance(0.6 pF)+Series Capacitor(6.8 pF)+Shunt Inductance(23.91 nH))+Old matched circuit of the above mentioned values but was also tuned are the value having (Series Inductance(11.975 nH)+Admittance Capacitance(C=0.67466 PF))

But this above tuned circuit was given me good NF at frequencies of 0.5,1,1.5GHz(0.289dB,0.260dB,0.260dB).

?•Here I have one question is, now my circuit is become large after tuning can I minimize this circuit while having better than this NF, cause of more components will add NF itself at input side.

After reaching this point now I have to look Stability, which has showing for this tuned circuit less than 1, S11 is bad.

Now the second step to influence stability and input impedance(or S11),doing Power matching or Input matching.
For Input matching now I have added feed back Inductance to the Source which will move Gamma Opt and S11 in smith chart and stability at higher frequencies goes up to >1.

S11 is improved -1 to -3 0r -5dB at broader frequency.

I have improved stability at less than 2 to 3GHz frequencies by adding RC feed back (drain to Gate), but has given me stability greater tan 1 at lower frequencies as well.(R=746 Ohms, C=50pF).

Cause of resistance and capacitance give me resonance, here my NF gone to worse from 0.259 dB to 0.8 dB, Gain also decresed.

Here my questions are How can I stabilize LNA without influence NF and S11(only just stability taking in to account from 1 GHZ to 20 GHz range and other parameters are same at wide band from 0.5 to 1.5 GHz has to be match).



I WANT TO MAKE TWO STAGE LNA BUT IF I WANT TO MAKE DIFFERENTIAL LNA HOW LONG WOULD BE TAKEN( I have short time to finish my design part of project In shaa ALLAH)




Added after 8 minutes:

Is it OK if I send my questions in the same topic of Low Noise Amplifier (HELP ME)or I have to make by myself new topic.(SORRY WITH OUT ASKING I HAVE SENT ALREADY MY MESSAGE)

Best Regards
Afzal Hussain Mohammad

low noise amplifier design pdf

about the diferential LNA , will it be on board ?

i have seen very littile designs which employ differential LNA on board , coz the transistors will not be matched pairs , or u will by a matched pair ?

what do u mean by minimize the circuit , do u want to minimize the components count in the matching networks ?

to minimize the components this mean u will use a single stage matching network , which is narrow band by nature .

about stabilization the transistor without losing the noise figure , it is very hard , coz if u want to stabilize the transistor u need to add reactive elements which already have paraistic resistance , and sure will add noise

if u can design ur amp , with matching so also make the amp conditionally stable , so the k is <1 but u sure that S11 and S22 < 1 , under ur sourece and load conditions


minimize noise figure amplifier

Thanks for your reply,

About the differential amplifier, I will first deisgn on ADS after that depending on seeing the results I could do board manufacturing.

Minimize the circuit means I am designing with Ideal components, at the input side if I use more components, when it will replace with real components resistance will be added and it would be have bad noise figure.

Today I have designed only for 3 elements shape of T network, Like Capacitance is in centre vertical two inductors are horizontol. I got the NF like over all 0.3 dB for broad band frequncies(0.5 to 1.5 GHz)

then I did Impedence match while adding Inductance at Source feedback, which has matched S11 and Stability better. But not at all frequencies, still LNA is not stable.

But How can I manage to do stable,(while I have tried to add resistane at input, output) still not giving me stable.

Actually i am designing LNA for astronomical purposes which should have atleast overall noise less than 1 dB. According to my supervisor less than 0.5 dB NF.

Looking for reply

Afzal Hussain Mohammad

amplifier gamma opt noise

man , which ur designing it is better to download the real models of the passive elements from murata and panasonic and avx , there are a design kits for these real componenets on the websites , use them while ur matching , it will help u alot

or another way , try to use as much as microstrips this will minimize the loss of the input but u need to use low loss substarte

about the astronomical purpose u need really a very low noise amps , even sometimes , they cool these amps for minimum noise performance

u need also to simulate the layout effects very well


atf-54143 hp

Thanks for your reply. I will soon implement with real components.

According to my supervsisor I can not use micro strip lines. Normally at less than 2GHz frequency lumped elemts are useful.

That is true in Astronomical area optimum NF LNAS are usable.

Do you have Idea about stability, to make the LNA stable at 1Hz to 20GHz, how could the componentsts will be added at out put(like resistor in parallel at output attched with capcitor series and Series Resistor in parallel with INductor) Can you explain me about this circuit what it behave in side those are in brackets.

Do you have any specific material that could help me about stability and biasing(also very samll amount of components)

Afzal Hussain Mohammad

using the lumped components to stabilize an lna

about the stabiliy and biasing u need to read the frist 2 chapters of
Practical RF Circuit Design for Modern Wireless Systems VOL II

is it very usefull about stability and how u make stabilyt networks


lna designing with atf 54143 eda board

Thanks Khouly,
I will read this 2 chapters.

Afzal Hussain Mohammad

Hi Khouli

I designed lna (upload in word document) , but the power is low. What do I do to increase it?

i have checked the doc , the gain is good , what do u mean the power is low ?
or u mean the gain is low ?


the operating power gain (Gp) is low. i want Gp>20 db

Not open for further replies.

Part and Inventory Search

Welcome to