Queries on Simualted and measured Biasing Values of LNA design in ADS

[pusparaga]

Dear Sir,
I have designed LNA using ATF54143 transistor, have set biasing values Vds= 3V , Vgs= 568mV, and Ids= 53.9mA for the nonlinear device model of ATF54143 transistor and simulated using ADS2011.10.
I have generated the layout and PCB fabricated for LNA, when I measured the fabricated PCB biasing Vgs, Ids, and Vds. Vds = 3v is only coming correctly. Remaining Vgs is showing negative voltage of -500mV, and drain current of Ids = 1mA.
When I designed and simulated LNA using nonlinear ATF54143 transistor, it was showing positive gate bias of 568mV and Drain current of 53.9mA. But When I have given 3volts to designed and fabricated LNA, it was showing - 500mV and 1mA drain current.Due to this -500mV at the gate , LNA is showing same output what ever the input it has, it is not showing amplifying signal. Some LNA is showing negative gain, when -500mV appeared at the gate.
What is the reason behind this negative voltage and 1mA current at gate and drain. Even though it was showing positive voltage of 568mV and 53.9mA at gate and drain, when I simulated using ADS2011.10.

 

[volker@muehlhaus]

Remaining Vgs is showing negative voltage of -500mV, and drain current of Ids = 1mA.

Do you understand that you have to set the gate voltage for your circuit?

 

[pusparaga]

Do you understand that you have to set the gate voltage for your circuit?

Volker Sir,
I didn't understand, what you asked.By setting proper bias voltages Vgs,Vds, and Ids, amplifier will amplify the signal otherwise it won't amplify. I have set the difference of the gate and source voltage(Vgs) as closer to positive 568mv and difference of drain and source voltage as closer to 3V(Vds). In the screen shot, you may notice the biasing values.

 

[volker@muehlhaus]

By setting proper bias voltages Vgs,Vds, and Ids, amplifier will amplify the signal otherwise it won't amplify.

Correct. And in your hardware, YOU need to set the gate voltage to approximately 550mV (positive), so that the correct drain current will flow. If the gate voltage is wrong (negative), then check your circuit and change your bias voltage to have the correct value.

 

[vfone]

LNA is showing same output what ever the input it has, it is not showing amplifying signal. Some LNA is showing negative gain, when -500mV appeared at the gate.

I think your LNA is oscillating, at a frequency that you are not looking for...

 

[pusparaga]

YOU need to set the gate voltage to approximately 550mV (positive), so that the correct drain current will flow. If the gate voltage is wrong (negative), then check your circuit and change your bias voltage to have the correct value.

Sir,
When I simulated USING ADS , Vgs was +568mV only. After the PCB design and component placing, it is showing the negative 500mV(-500mV). Which one I have to trust, whether simulation is wrong or PCB fabrication is wrong, I am in confusion.
Already PCB designed and Components are placed, I don't have the option to change anything. Only option, I can the components.
Why it is showing - negative at the Vgs=- 500mV, while measuring, even though Vgs , it was simulated with positive result, what could be reasons.
How can I get the positive at the Vgs, while measuring. Is there any changes can I make.Can you suggest me, how to get the positive voltage at the Vgs while measuring as closer to the simulated Vgs= +568mV.

 

[volker@muehlhaus]

If you look at the ATF54143 datasheet, figure 1, you can see that you need a positive gate voltage. With your negativ gate voltage, no drain current can flow.

The negative gate voltage can be from wrong circuit design, or you measure it the wrong way, or maybe the amplifier oscillates, as vfone has suggested.

I would first test the gate bias voltage without the transistor (simple DC test with a multimeter).

 

[RealAEL]

If the problem is with the assembled circuit board it would have been far better to post pictures of the layout or the final assembled board itself. Posting a picture of only part of the schematic 'that works' does not help anyone work out what the problem is.

I certainly agree with Volker that taking the transistor off the circuit board to check the gate bias while it is not loaded by the transistor is a sound starting point to diagnose the problem. Then make sure that you re-attach the device the correct way around, remembering that the broad source tab is on the same side of the package as the drain and not opposite it according to the Avago datasheet.

 

[pusparaga]

If the problem is with the assembled circuit board it would have been far better to post pictures of the layout or the final assembled board itself.

RealAEL Sir,
I am herewith attaching my screen shot of LNA design generated layout, which has given to PCB design and got the PCB design. As I discussed in the earlier postings, in the simulation Vgs was showing positive 568mV (+568mV), while measuring it was showing -500mV.
Now I can change the components only, that option only I have. I request you to help me to get positive voltage closer to the simulated values at the Vgs for the fabricated PCB design measurement. Can you suggest me, what could be done to achieve the measured biasing voltages at the Vgs closer to the simulated Vgs values.

 

[volker@muehlhaus]

Looking at your layout, the DC gate voltage must be positive, because it's a simple voltage divider from the drain voltage. Maybe vfone is right and your amplifier oscillates (-> measure with spectrum analyzer).

In your PCB, there is one obvious mistake: you do not have any capacitor to ground at the voltage supply lines (meander). Add something like 100pF to ground at the DC supply (if the meander lines are 1/4 wavelength lines: at the DC side of each meander).

 

[RealAEL]

The first step would have to be, as previously mention by both Volker and myself, to take the transistor off the board and check the unloaded Vgs. If it is not correct adjust the resistors as necessary. Only when that is correct re-attach the transistor.

If you suspect oscillation may be the issue then simply attaching a spectrum analyser to the output to see if there is any 'unexpected' output tones would be a simple test.

 

[pusparaga]

to take the transistor off the board and check the unloaded Vgs. If it is not correct adjust the resistors as necessary. Only when that is correct re-attach the transistor.

RealAEL Sir,
As you told, I followed it. I took transistor off and checked the Biasing, now it is coming correctly Vgs = +567mV and Vds= 2.99 V, but Drain current is showing "1mA". Later I reattached the transistor and getting the Vgs as positive 567mV and Vds = 2.99Volts, and also Drain Current is showing 1mA. What has happened to this, why earlier it was showing negative -567mV at the Vgs. What could be the reason, even circuit is correct. After took transistor off, then it is showing very closer results to simulated values of positive Vgs and Vds.
But now I am facing the problem of 1mA current is showing at the Drain. Simulated drain current is 53.9mA, how to get the closer to measured drain current to simulated drain current of 53.9mA. Why it drawing 1mA current, even though it is simulated with 53.9mA , what could be reason. Please help me to get drain current of measured value closer simulated 53.9mA.Even I changed the new transistor and replaced with earlier correct working ATF54143 transistor also, it is showing the drain current of 1mA and Vgs= +567mV and Vds=2.99mV.

 

[volker@muehlhaus]

As you told, I followed it. I took transistor off and checked the Biasing, now it is coming correctly Vgs = +567mV and Vds= 2.99 V, but Drain current is showing "1mA". Later I reattached the transistor and getting the Vgs as positive 567mV and Vds = 2.99Volts, and also Drain Current is showing 1mA. What has happened to this, why earlier it was showing negative -567mV at the Vgs. What could be the reason, even circuit is correct.

It seems that you reversed the + and - wires of your multimeter :idea:

But now I am facing the problem of 1mA current is showing at the Drain. Simulated drain current is 53.9mA, how to get the closer to measured drain current to simulated drain current of 53.9mA. Why it drawing 1mA current, even though it is simulated with 53.9mA , what could be reason.

You control the drain current with the gate voltage, see datasheet Fig 1. So you would need to change the gate voltage to control the drain current.

But if you measure 1mA current without the transistor, and 1mA with the transistor, something is wrong. Maybe you measure the current in the wrong way? Do you have a supervisor at the university who can have a look?

 

[pusparaga]

It seems that you reversed the + and - wires of your multimeter

Sir,
I checked the properly the polarity of the multimeter, it was correct only. Polarity of the multimeter is not reversed. Now due to positive voltage of the gate, amplifier is amplifying, earlier amplifier was not amplifying.

 

[RealAEL]

I took transistor off and checked the Biasing, now it is coming correctly Vgs = +567mV and Vds= 2.99 V, but Drain current is showing "1mA".

Now I understand what is happening. You are NOT measuring the drain current. With the transistor removed there is zero Drain current. So you have to be measuring the total supply current and the 1 mA is the current through resistor divider you are using to set the Gate voltage. Therefore when you re-attached the transistor and still got 1 mA current that means that the Drain current was still zero. The device was not re-attached to the board correctly and you would have had a open circuit connection.

So now if the device is attached correctly and the total current is closer to the desired 55 mA (54 mA Drain current plus 1 mA bias network current) you should be able to fine tune the resistor network to set the device bias as required.

 

[pusparaga]

The device was not re-attached to the board correctly and you would have had a open circuit connection.
So now if the device is attached correctly and the total current is closer to the desired 55 mA (54 mA Drain current plus 1 mA bias network current) you should be able to fine tune the resistor network to set the device bias as required.

RealAEL Sir,
As you said, I followed it. I have re- attached transistor to board correctly( I have checked the continuity using the Digital multimeter of all 4 leads transistor to adjacent connected MLINS, they are absolutely connected, there is no doubt in device is re-attached perfectly to the board).
Now also It is showing 1mA current at the Drain, once again negative voltage -568mV is appeared at the Vgs( I didn't reverse polarity of the multimeter, polarity of the multimeter is not reversed, it is correct only), some times it is showing 0 Volts at the Vgs.
Even I took transistor off, and checked the biasing Vgs was showing positive 568mV(+568mV) and drain current was 1mA, and when re-attached the device to board, it is showing negative Vgs(=-568mV) and 1mA drain current, some times it is showing 0 V at the Vgs, when reattached the device to the board properly.
Even I tried with the new good transistor also, before attaching device, it is showing bias perfectly at Vgs positive +568mV and 1mA current at the Drain. After reattaching, it is negative 568mV(-568mV) at the Vgs and 1mA current at the drain.
Now I am totally confused , what I have to do. I am disappointed more by seeing these results. Can you help me get out this problem. How to get the positive voltage at the Vgs and 54mA at the drain after connecting the transistor.
Even I tried, by adjusting externally Vgs positive voltage of 550mV.At the drain,it was showing more than 200mA current, device is spoiled.

 

[BigBoss]

pusparaga...
The device will not show you a Vgs voltage ..

YOU WILL apply a proper Vgs Voltage to the device , YOU WILL..!!!! The transistor will not supply Vgs by itself, you must apply a Vgs voltage to get desired Drain Current...

 

[pusparaga]

pusparaga...
The device will not show you a Vgs voltage ..

YOU WILL apply a proper Vgs Voltage to the device , YOU WILL..!!!! The transistor will not supply Vgs by itself, you must apply a Vgs voltage to get desired Drain Current...

Sir,
I have applied 3volts supply voltage, the difference of gate and source voltage (Vgs) is appearing +567 mV as simulated. The simulated Vgs( the difference of gate and source voltage) which can be noticed in the screen shot of posting #1.

 

[BigBoss]

Sir,
I have applied 3volts supply voltage, the difference of gate and source voltage (Vgs) is appearing +567 mV as simulated. The simulated Vgs( the difference of gate and source voltage) which can be noticed in the screen shot of posting #1.

Can you post application schematic of your board ?? Not simulation schematic !!
I wanna look what you did...

 

[pusparaga]

Sir,
Herewith I attached screen shot of Board