Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronic Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Register Log in

Single Balanced Mixer design: need help in ADS

Status
Not open for further replies.

haizal

Newbie level 5
Joined
Apr 25, 2005
Messages
8
Helped
0
Reputation
0
Reaction score
0
Trophy points
1,281
Activity points
1,406
dear friends..

I need help with my design...This is my information of my design:

1) I have designed 90° hybrid coupler at 2.4GHz,that consist of 4 port.The isolation between port 1 and port4 is about -3dB.
-> S21=231=3dB
-> S11=S41=-40dB
->phase different between port 2 and port 3 is 90°
And perhaps it would work when i integrate it with RF/LO and diode mixer (HSMS8101)

2) I have integrated all of those components.At
port 1 :i have represent RF signal power source at 10dBM at 2.4 GHz and at
port 4:I have represent LO with power source at 10dBm at 2.4 GHz.Since i need to implement it for zero-IF receiver the frequency that I set is the same.

Port 2 and port 3: i connected it to single diode that parallel with each other and terminated at 50ohm at each end.



3) My problem is..when i done with circuit simulation with harmonic balanced,

i) When i meaasure output level that come out form RF/LO at port 1 and port 4 is not 10dBm anymore.What happen to the source?i got about 5.8dBm is far from the generated input source of 10 dBM..Is it measurement is ok?

ii) Another problem is...when i measure connection of diode at the other end..the phase different output from diode 1 and diode 2 is not 90° anymore..Can anybody help me and suggest me how to solve this thing and is this simulation is valid?

iii) My aim for this project is to get I &Q simulation result at both end...so i'think it is important to have 90° phase different after pass by the diode mixer . Can anybody help me with suggestion for my project.

regards,haizal
-thank you
 

Mazz

Advanced Member level 2
Joined
Nov 3, 2001
Messages
676
Helped
131
Reputation
262
Reaction score
50
Trophy points
1,308
Activity points
5,463
Single Balanced Mixer design: need help in @DS

The problem is due maybe to non correct impedance levels.
Check with S parameters the impedances with diodes connected.
Maybe you have some power reflected, so you measure a lower value.
The coupler perform well if matched: also phase difference can be explained in this way.
Be careful that from RF port your system should be linear, so keep RF signal at lower level than P1dB of the mixer (maybe 10 dBm is too high).
I hope it can help.
Mazz
 

haizal

Newbie level 5
Joined
Apr 25, 2005
Messages
8
Helped
0
Reputation
0
Reaction score
0
Trophy points
1,281
Activity points
1,406
i have try to put RF power and LO power at 0 dBm..and i still doesnt get 0dBm at the input..when i do the simulation it goes about -7dBm..what happened?

i think my design of hybrid coupler is fine...when i connect port 3&4 to mixer diodes its look like something weird..it is necessary to get the output different 90°..i just being confused with my simulation result..can anybody help me..


for ur information directly i connected the 2 diodes(hsms8101-from agilent) which one to each output port of hybrid coupler and terminate is with 50 ohm terminator. Is it ok with my setup?and it is necessary for me to do impedance matching?since i'm need to implement it in zero-IF.(RF freq=LO freq=2.4GHz)
 

yolande_yj

Full Member level 2
Joined
Jan 20, 2005
Messages
145
Helped
0
Reputation
0
Reaction score
0
Trophy points
1,296
Activity points
1,635
Single Balanced Mixer design: need help in @DS

I think Mazz is right. It should be due to the reflection. Can you show how you "meaasure output level that come out form RF/LO at port 1 and port 4". Do you masure the voltage and current at these ports and calculate the power using the equation like: 0.5xreal(V x conj(I)) ? This power is the incidence power, not the source power. I think showing a graph of your design is much better.
 

flyhigh

Advanced Member level 1
Joined
Nov 12, 2003
Messages
409
Helped
68
Reputation
136
Reaction score
29
Trophy points
1,308
Location
somewhere over the rainbow
Activity points
4,772
Hi,

I think your mismatch comes from diodes that are cousing imbalance of the coupler. It is designed for 50 Ohm environment, so any severe difference to diode impedance from this value can couse coupler to behave poorly. If you want to match diodes, you should do it at 2.4GHz as this is the frequency of your RF and LO that is important to maching. You can alternatively choose a different diode that has impedance close to 50 Ohms.

Terminating diodes to 50 Ohms on IF side might not be optimum too, maybe some transformer is needed there to present optimum IF impedance to the diode. Also, it is almost allways beneficial to have IF filter (although your IF is 0) as this can block higher mixing products and improve overall mixer performance.

flyhigh
 

haizal

Newbie level 5
Joined
Apr 25, 2005
Messages
8
Helped
0
Reputation
0
Reaction score
0
Trophy points
1,281
Activity points
1,406
dear mr flyhigh and yolande...

thanx for both of ur advice..

i just check out my mixer diodes data sheet...
it stated there Zif= 150 Ω..is it true that i need to do impedance matching because that this Zif≠50Ω?and is this Zif represent the diodes impedance itself?just enquiry and need help for u all..if it is need for me to do matching ..what are suitable matching technique..is it stub matching or quarterwave?


mr yolande..

i think, for the 1st step i just looking at the power level and frequency where IF is drop..before further simulation on baseband signal...i will paste the diagram of my simulation after this..for now I just selected the power using what available in dqata display (e.g:dBm),can you give me a comment and suggestion for what i'm doing..
thanx

Added after 43 minutes:

this is my simulation file...

one more thing i just asked my lecturer about Zif impedance..he told me that Zif is only output of the diode at if frequency..it might effect at the if frequency..and since IF occur at lower frequency..impedance matching at IF is not so important since i only need to extract the base band signal..he said that the normal RF diodes it impedance is 50 ohm..and this diode is tested before and normally used at 50 ohm..any suggestion with my design?
 

scuba

Member level 1
Joined
Nov 15, 2005
Messages
40
Helped
1
Reputation
2
Reaction score
0
Trophy points
1,286
Activity points
1,612
i see your describe about your project.
and is't wrong with your S41..
could you please tell your difference of measure result and simulation?
my project is different a lot.
 

yykk

Member level 1
Joined
Aug 29, 2003
Messages
39
Helped
0
Reputation
0
Reaction score
0
Trophy points
1,286
Location
Hong Kong
Activity points
276
Hi all,

I think a AC ground (at 2.4 GHz) should add at the end of each diode and then through a lowpass filter to obtain the IF.
 

Status
Not open for further replies.
Toggle Sidebar

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Top