UHF Amplifier Problem

hi everyone
i'm working on a project which needs serial data transmission
i used hm-t FSK Module @ 915 MHZ frequency!but it's too weak!only 5 mw output power
i chose "bgy204" ufh rf amplifier module from NXP!it amplifies signals between 800 to 915 mhz!
BGY204 Datasheet pdf - UHF amplifier module - Philips
it's ok but the striplines are hard for me to construct
on page 8,explained how to preapre them but there is a problem
in my country,only exists fr-4 fiberglass pcb which has 4.4 dielectric constant
but the pcb's dielectric must be 2.2!i know any change in epsilon-r makes the circuit unstable but i can't design a new pcb and striplines with considering epsilon-r=4.4
could anyone plz help me?what am i doing?desing a new pcb and striplines or coordinating this pcb with new dielectric constant,4.4?
thanks for your regard

To design the 50 ohms transmission lines at the input and output of the module you can use the freeware program below:

https://web.awrcorp.com/content/Downloads/TXLine.zip

FR4 fiberglass, Er=4.4, should be fine for your application.

I totally agree with vosh.
Actually, it is not necessary to make same stripline as datasheet. You can get the characteristic impedance of the this stripline with material described in datasheet, then you can design your own input and output transmission line with your own materials, any type TL, microstripe, strip line etc, I think.

thanks alot guys
now what is the difference between stripeline or microstripe?which is better?
i bought fr-4 pcb yesterday and measured the di-electric constant by capacitance meter on my multimeter by solving "c=k*8.85*10^-12*A/L" equation
i found fr-4 di-electric constant 4.9!is it a correct method?
by the way...............
is numeral shape of striplines or microstripe important in designing or not?how about thickness of copper?
regard
"Armin"

Because your freq is low, use FR4 is no problem. I like use microstrip line.
Your method of measuring Er is not so accurate, and also FR4 Er is also varaiable between 4-5 in China. It is not so important. You can use 4.4 for design and then set Er to 4 or 5 to see the VSWR, then you know the difference.
For FR4, copper normally is 35um or 1oz, that's ok for 900M.

Microstrip has lower loss-tangent problem.
Microstrip has faster propagation time.
Stripline has better immunity to crosstalk.
Stripline has better EMI characteristics.

There are multiple ways to measure the dielectric constant:
**broken link removed**

Thickness of the copper affect the impedance, but not too much.

thanks so much
i found out hm-t module output impedance is not 50 ohm and the ouput power is about 4mw i think
but in bgy204 module catalogue emphasizes that the input impedance and power must 50 ohm and less than 2 mw
how can i attenuate the ouput power of "hm-t" and match the impedance between the ouput of hm-t and rf amplifier?

nobody wants to help me?

You can use a simple 3dB attenuator having the right impedances at the ends.

**broken link removed**

vfone said:

You can use a simple 3dB attenuator having the right impedances at the ends.

**broken link removed**

Click to expand...

thanks
what is the resistor type in this circuit??i asked because of inductive effect of coiled wire resistors at high frequencies
shall i use carbonized resistors for example?

Use SMT with 0603 package, 5% error, 1/8W power rating, is OK for 900M.

---------- Post added at 23:00 ---------- Previous post was at 22:59 ----------

SMT resistor

nimra said:

thanks so much
i found out hm-t module output impedance is not 50 ohm and the ouput power is about 4mw i think
but in bgy204 module catalogue emphasizes that the input impedance and power must 50 ohm and less than 2 mw
how can i attenuate the ouput power of "hm-t" and match the impedance between the ouput of hm-t and rf amplifier?

Click to expand...

nimra,

Are you clear about the out impedance when 4mW output power from hm-t module?
If you know, 30Ohm is assumpted, then you can refer to the PI-Attenuation/T-Attenuation mentioned by vfone

Here is for your reference: