help me to design bias cricuit!

biasing circuit for tim7785

I don't konw how to design a better bias circut for X band PA.
Anyone help me?
I have used λ/4 line with Radial Stub
but the simulation results isn't well enough.........

Try to make the collector/drain quarter-wave transmission line a little shorter, to be more inductive, and retune the output match. This will improve the load-pull behavior of the PA. Also check the behavior of the radial stub at 2fo and 3fo.

Thanks！But what's "load-pull behavior " meaning? I don't understand it much!
one more question:how to design the Radial Stub ?
the angle and radial？

Load pull consists of varying the load impedance seen by a PA while measuring the performance of the PA. This method is important for large signal, nonlinear devices where the operating point may change with power level or tuning of output match.

About radial stubs and small software to design it:
https://www.rfcascade.com/stub.html
**broken link removed**

thanx a lot!

footprint said:

I don't konw how to design a better bias circut for X band PA.
Anyone help me?
I have used λ/4 line with Radial Stub
but the simulation results isn't well enough.........

Click to expand...

make the 1/4 lamda as high impedance as u can

[quote="eladg"

make the 1/4 lamda as high impedance as u can[/quote]

but with large current ,how to consider the width of the high impedance line?

footprint said:

[quote="eladg"

make the 1/4 lamda as high impedance as u can

Click to expand...

but with large current ,how to consider the width of the high impedance line?[/quote]

u can solder another line or wire on the line so it will be higer or use substrate that have good heat transfer.

i build it for tim7785-30sl (id = 8A - idSS=22 AMP ), and it was on 5880 10 mil but i use 1 onz cupeer, i made the stub > 145 deg (use momentum ).

it was very good

good luck, and if u have problem let me know

hi
can u elobarate more about lamda/4 line
i mean like wht is this lamda/4 line???? wht is the main purpose &
how to calculate
thanx in advance

chakri
you can see this

vfone said:

Load pull consists of varying the load impedance seen by a PA while measuring the performance of the PA. This method is important for large signal, nonlinear devices where the operating point may change with power level or tuning of output match.

About radial stubs and small software to design it:
h**p://www.rfcascade.com/stub.html
h**p://www.microwaves101.com/encyclopedia/quarterwave.cfm

Click to expand...

Hei, eladg
can you tell me how to design the stub?
I find the R0 with lamda/4 is very good but it's too large!!!
and I can't understand "but i use 1 onz cupeer, i made the stub > 145 deg " means!

I always try to place the λ/4 bias line at a voltage minimum point in the matching network.

where is "voltage minimum point in the matching network"?
and how to find it??

Well, a bias line is a circuit component that should have as high an impedance as possible, so as not to load down the main line. With transmission type bias lines, there is only one frequency where it truly looks like very high impedance, such as the frequency where a grounded stub is λ/4. That simple bias stub at the frequencies where it is not λ/4 will look either like a shunt capacitance or inductance to the main line. This shunt reactance will reflect rf signals in the main line, causing what looks like limited operating bandwidth of the bias circuit.

So lets say we are at a frequency off from the center frequency, say where the gronded stub is 0.2λ in length. Its impedance of the bias stub is going to be around j 3.0 * Zs, where Zs is the stubs characteristic impedance. Lets say you used a stub characteristic impedance of 70 ohms, so Zs is j 210 ohms.

If you used this bias stub on a standard 50 ohm transmission line, loading it with a reactance of j210 ohms will cause some reflection.

Lets say, instead, that you used a transformer to transform down to a 20 ohm chanracteristic impedance, attached the bias line, and then transformed up to 50 ohms again. Now you have j210 ohms loading a 20 ohm line, which is a much smaller reflection than j210 ohms loading a 50 ohm line. The end result--your bias circuit has a 2.5 times as much bandwidth now.

More practically speaking, lets say you have a bipolar transistor with an output impedance of 10 ohms. If you attached you bias line right there at the transistor output, that bias line reactance will have very little loading effect.

Lets say, instead, that you use some matching network to match the transistor output to a 50 ohm system, and then added the same bias line structure, you would see 1/5 the bandwidth since it is loading down a much higher impedance point.

So think of the matching structure as a standing wave of voltages, find a practical place to add the bias line where you have the a voltage minimum, and stay away from where the voltage maximums would be, and you will have wider bandwidth.

O, I will try it like this! Thanx!

I 'm working on this problem ,if anyone wants to exange your ideas, welcome !!!!!
my msn:
calvindd@hotmail.com

But I don't konw how to design the R0 of Radial Stub ? who help me??