how to match imdeance of pcb traces ?

Hi , i am designing the a pcb trace between module and antenna sma connector. (i am not designing an antenna) .(sma connector)
the connector is has 50 ohm impedance . and after using online calculators my pcb trace is becomming around 90 ohm .

i am designing connector for gsm frequency (433 mhz in india) and lora (866 mhz in india)

how can i match the impedance of the trace ?
can i use some caps or inductor here to bring my impedance to 50 ohm ?

some info

impedance calculator

microstrip antenna

trace width = 1.016 mm

pcb thickness = 1.6 mm

cu thickness = 1 oz or 35 um

dielectric material = fr4
er = 4

trace lenght = 7.6 mm
(straight trace . no bend , no curve , staight as an arrow )

Good question.
Do you have matlab ?

design-a-50ohm-impedance-net-for-rf-signals

analog.com/media/en/training-seminars/tutorials/MT-094.pdf

Microstrip Calculator

microstrip-width-calculator

microstrip-impedance

Standard FR4 1.6mm will be about 3.2mm trace width for 50 ohm, from memory.

MarkT:
Standard FR4 1.6mm will be about 3.2mm trace width for 50 ohm, from memory.

Close to get exactly 50 ohm you need 2 mm trace width .

now my question is how does the trace lenght come into play ?

cuz the lenght will have some parasitic cap and resistance.

why are all the online calculators and thier formula dont inclue lenght into the equation ?

raschemmel:
Good question.
Do you have matlab ?

nope i dont have it .

thanks for these links.i have gone thru all of them but none of them help me with my original problem .

can i match my impedance using external caps and inductors ? and if yes then how ?

manveen_singh:
Close to get exactly 50 ohm you need 2 mm trace width .
now my question is how does the trace length come into play ?
cuz the length will have some parasitic cap and resistance.
why are all the online calculators and their formula don't include length into the equation ?

Because what you are proposing to do is called a Transmission line.

Need to read up on it. :roll_eyes:

All this fuss is about 7.6mm stripline on 433Mhz?
Better get the impedance of the hight of the solder blob on the chip/module and SMA connector right as well.
Leo..

Oh I strongly dislike shaving solder blobs anymore since we stopped using leaded solder.

Wawa:
All this fuss is about 7.6mm stripline on 433Mhz?

what do you mean by that ? am i taking impedance matching too seriously ? is 7.6 mm lenght too small to make a difference ?

what do you mean by that ? am i taking impedance matching too seriously ? is 7.6 mm lenght too small to make a difference ?

I think so. I think it is the same as 1 mm or 2 mm.

manveen_singh:
what do you mean by that ? am i taking impedance matching too seriously ? is 7.6 mm lenght too small to make a difference ?

At that frequency, yes.
Different story if it was 2.4Ghz or higher.
Nothing wrong with learning how to do it right though.
Leo..

You need to look at "transmission lines" again. Unless you match the input to the line and the output of the line to the line impedance. you are wasting your time. 50 Ohm transmission line must be driven by 50 Ohms and terminating with 50 Ohms. Anything else wastes power.

Paul

Paul_KD7HB:
Unless you match the input to the line and the output of the line to the line impedance. you are wasting your time. 50 Ohm transmission line must be driven by 50 Ohms and terminating with 50 Ohms. Anything else wastes power.

Paul

Yes that i 100 % agree upon .

thats my original question how to match ?
i will be looking at transmission line.

and rf design guides.

i have found a reference design that ues r and caps to match the impedance,
i was also hoping to find the inductor there somewhere in the circuit. but i didn't find it.

reference desgin

pg 18

so now my question is again how to choose the value of these caps.
basically how to tune this ckt ?

this is also my 100th post so now i can

manveen_singh:
i was also hoping to find the inductor there somewhere in the circuit. but i didn't find it.

It is actually the "resistor" with a value of zero Ohms. :grinning:

manveen_singh:
i have found a reference design that ues r and caps to match the impedance,
i was also hoping to find the inductor there somewhere in the circuit. but i didn't find it.

reference desgin

pg 18

so now my question is again how to choose the value of these caps.
basically how to tune this ckt ?

this is also my 100th post so now i can

Your GSM antenna is the inductor!

Paul

For the sake of discussion , I thought it was agreed, or at least stated that this post is academic because
the impedance difference for 7.6 mm at 433 Mhz was negligible.

Did I misunderstand that ?

Though I don't have better knowledge on it. But I would like to learn.

manveen_singh:
Close to get exactly 50 ohm you need 2 mm trace width .

No, 3.2mm is pretty accurate for 1.6mm thick FR4 microstrip, I've measured it and calculated it. Perhaps you are thinking of 1mm thick substrate? The OP says 1.6mm FR4

now my question is how does the trace lenght come into play ?

Characteristic impedance is independent of length - the loss depends on the length of course.

cuz the lenght will have some parasitic cap and resistance.

That's not the case if the surrounding circuitry is matched to 50 ohms, which it ought to be! All the parasitics are tuned out if the impedance remains constant along the signal path.

raschemmel:
For the sake of discussion , I thought it was agreed, or at least stated that this post is academic because
the impedance difference for 7.6 mm at 433 Mhz was negligible.

Did I misunderstand that ?

No, but what if the board layout changes in the future? Using the correct impedance trace will mean no repercussions to later changes in that trace's length. There's no cost to using the right trace width, no
trade-off to be made. Its just good RF practice.