Matching Circuit Help
I need to match 912MHz SAW device to 50 Ohms within its 10% fractional bandwidth, which has only 1 port. I have done the matching network design already. Theoretically, it should work, as seen in attached pdf. Then I optimized using ADS layout then ADS momentum. It looked good. Then I built the board and the results weren't as expected.
I have detailed my design process in the attached pdf and I have included the results. I need some advice on what to do as I'm stuck. Forgive me for using somewhat informal and less professional language on the attached pdf. I have neatly arranged it the best that I could. If there's something that I should clarify, please let me know. I'm stuck and the only place left I could think of to ask for help is in here.
I just learned how to optimize using momentum today but i've used ads for more than a year already.
Thanks for the help in advance!
Share the design file if possible.
The details provided by you are insufficient.
hi. i have uploaded my ads project in the form of a zip.
http://dl.getdropbox.com/u/602818/912MHzSAW_prj.zip
file descriptions; please look under networks folder; ads2008 was used by the way and i don't know if there are any compatibility issues
ver2optimize.dsn: using just ADS layout, I input the file into a data box where the effect of the SMA connector was taken to account (theoretical). then i optimized the structure for 50 ohm matching looking into port 1
ver2optimize50ohm.dsn: copying the optimized network from the previous file, i took out the data box and replaced it with port2 (Z0 = 50). now i know what the structure reflection coefficient looks like looking into port 2 while terminating port 1 with a 50ohm load since port 1 is still terminated Z0 = 50. now that i have this information, i recorded few data points which would become my 'goals' when i optimize my structure.
ver2optimize50ohm_1b.dsn: after setting up my structure through momentum and importing my momentum component back to ads layout for optimization, i optimized to match the s22 of the previous file. please see dds file as to how close my goal response and momentum optimized response match up. also i used momentum mw instead of rf as the latter gave me some sort of warning. i also input 0 for edge mesh so it'll determine what the edge dimension should be and i think i did 40 sections per wavelength in doing simulation
finally, if we look at the last 2 plots of my pdf file i attached earlier, you will see real and imaginary s22 after terminating port 1 with a 50 ohm load that i have. the results are off. if i attached the load i'm matching to to port 2 and looked at port 1 where i expect 50 ohm matching, i'm so off.
do you recommend me doing hfss instead? if i do every strip of line manually by fabricating each and connecting them connector, would that work too? i thought about doing it but it might mess up the whole thing as things changes when i have tee structures for example.
thank you for taking the time to look into my problem... any help / advice is greatly appreciated.
I do not have ads, so can not run your files. But some basic quesitons:
"912MHz SAW device to 50 Ohms within its 10% fractional bandwidth, which has only 1 port. "
what do you mean by 1-port? Is this a 2 terminal resonator, a 2 port filter? In your first post, you do not show any ground connections in the artwork. Are there no ground connections to the saw device?
The SAW device I have is a sensor composed of a transducer with reflectors on both sides of the transducer. The only port I have is connected to just that one transducer and some information is extracted depending on the reflector response, which could change with temperature for example, but not a drastic change in impedance at all. The case ground of the device was connected to ground of sma solder cup. I already did something similar (matching the device to an antenna) for 250MHz SAW device, where the design is detailed here http://purl.fcla.edu/fcla/etd/CFE0001205 . But since I used varactors and variable inductors, that was easier for me. Now that I'm in 912MHz, lumped components could be too small for good inductors and capacitors so I opted to do microstrip.
Thanks![/url]
Do you have a data sheet of the SAW device?
An interesting paper, but unfortunately no time to read it now. Maybe someday.
I do not understand what you mean by a match. The saw device, whatever it is, does not have a 50 ohm resistor "hidden" inside of it somewhere. the definition of match is to
1) transform the 50 ohm external impedance to the devices equivalent resistance
2) conjugately cancel the reactive impedance
Since you do not have a defined device resistance, you can not match anything.
It sounds more like you just want to bounce a signal off of the SAW device, and measure the return phase shift. If that is what you are trying to do, why do you need an impedance match.
This is confusing.