The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl Simulators >> RF Simulators >> CG and matching about mixer https://designers-guide.org/forum/YaBB.pl?num=1277737192 Message started by steven_l on Jun 28th, 2010, 7:59am |
Title: CG and matching about mixer Post by steven_l on Jun 28th, 2010, 7:59am I am confused about conversion. 1 http://www.designers-guide.org/Forum/YaBB.pl?num=1122952471/2#2 If I setup E in port, then in reality, the signal at the gate of the mosfet is 2E. I wonder if the E means PAC mag of RF POIRT in cadence? Why the gate of the mosfet is 2E? Assume the port is ac coupling to the gate and there is no match net work, which means signal at the right side and left side are different. One side is E, the other side is 2E. Why? how can a port with E can generate a 2E at the gate of mosfet? Then Voltage gain of transient means Vout/Vin= Vout/E. Why not Vout/2E? Because the signal at gate is 2E~~ 2 There are many voltage definition: For PAC, PXF and Pniose, it is Vout/(E/2) For transient and PSP, it is Vout/Vin, Vin is the signal voltage at a input node of DUT. I wonder if the input node means the + terminal of port or vsin? It seems that Vout/(E/2) is used for power based system, while Vout/Vin is for voltage based system or for low frequency. As mixer is a RF block, it seems that the voltage conversion gain Vout/(E/2)is correct. Say, PAC plot in Direct Plot is correct. But why “http://www.designers-guide.org/Forum/YaBB.pl?num=1237656418/0” says , CG plot from Direct Plot is wrong? 3 I parallel a 50 Ohm resistor with PR port and LO port, respectively. This is for broadband resistive matching. But as http://www.designers-guide.org/Forum/YaBB.pl?num=1194006190 says, 50 Ohm does not one part of circuit. So the stimulation results are not correct. But in reality,without 50 Ohm resisitor how can I realize broadband match? My mixer is resistive mixer, input signal from tens of MHz to 2GHz. Can I use off-chip match network? But the imput impedance is time variant. How can I match the input impedance to 50Ohm for such a wideband? |
Title: Re: CG and matching about mixer Post by pancho_hideboo on Jun 28th, 2010, 8:06am You use "E" as half of open circuit voltage of port. Your usage of symbol "E" is not common. Generally "E" is used like following. See http://www.designers-guide.org/Forum/YaBB.pl?num=1122952471/2#2 "E" is used as an open circuit voltage of port generally. steven_l wrote on Jun 28th, 2010, 7:59am:
steven_l wrote on Jun 28th, 2010, 7:59am:
steven_l wrote on Jun 28th, 2010, 7:59am:
Conventional voltage gain is Vout/Vin not Vout/E, even if you mean "E" as half of open circuit voltage of port. steven_l wrote on Jun 28th, 2010, 7:59am:
steven_l wrote on Jun 28th, 2010, 7:59am:
steven_l wrote on Jun 28th, 2010, 7:59am:
Both gains are correct. But Vout/Vin is more useful and practical gain definition than Vout/(E/2) in IC chip. Here "E" is an open circuit voltage of port. steven_l wrote on Jun 28th, 2010, 7:59am:
manfred wants to get Vout/Vin. But he gets Vout/(E/2) using "Direct Plot". I mean this is not correct. Here "E" means an open circuit voltage of port. steven_l wrote on Jun 28th, 2010, 7:59am:
steven_l wrote on Jun 28th, 2010, 7:59am:
For your case, 50ohm resistors are part of your circuit. So the simulation results are correct. See http://www.designers-guide.org/Forum/YaBB.pl?num=1207009725/1#1 The followings are general notes for you. - Always describe correct tool's name and vendor's name which you use as tool or simulator. - Don't do multiple posts which are same content. - Don't request source code or behavioral model without any efforts. - There are many simulators which have analyses called as PSS, PAC and Pnoise. - Describe in detail with using correct terminologies. - Warnigns are different from Errors. - ADS is not name of simulator. - There is no tool which name is Cadence. - Don't use Direct Plot of Cadence ADE blindly without knowing definition. - All gains in Direct Plot of Cadence ADE are "right", "true" and "practical" voltage gain. - Don't mix up Simulation with Post Processing. They are completely different phase. - MATLAB are different from Simulink. - Learn measurements using actual instruments. Not "EDA Tool Play |
Title: Re: CG and matching about mixer Post by steven_l on Jun 28th, 2010, 10:11am pancho_hideboo wrote on Jun 28th, 2010, 8:06am:
I can not understand. In my opinion, if a RF port in cadence has E in PAC mag, it mean the port generate E Vp-p voltage to the gate of mosfet through AC capacitor. In deed , the input impedance is very large. But why it is 2E instead of 3E or 4E. I can not understand it through intuition. Can I get the result from mathematical derivation? pancho_hideboo wrote on Jun 28th, 2010, 8:06am:
Which net represents Vin? I am sorry. I do not know why I can not upload picture. So I can not describe my question clearly. So I can only describe my circuit in following words. Apploogized Commonly, Vsin series a AC capacitor. And the AC capacitor series the gate of mosfet. Does Vin net mean the net between AC capacitor and gate of the mosfet? While port is made of a ideal source and a internal resistor. If PAC mag of the port is E, the internal resistor is 50 Ohm. I series the port and gate of mosfet through a AC capacitor. Then signal at the gate of mosfet is 2E( I can not understand until now). In these two cases, source signal means the signal at the gate of the mosfet. And the gate net is the input net. Also the gate signal is the input signal. pancho_hideboo wrote on Jun 28th, 2010, 8:06am:
You say Vout/Vin is more useful and practical than Vout/(E/2) in IC chip. I wonder does the CG in many IEEE paper represent Vout/Vin or Vout/(E/2)? In my receiver, LNA, mixer and IF filter are all off chip block. The LNA and filter need to buy. The mixer is need to design in IC. So in the case Vout/(E/2) is more useful? pancho_hideboo wrote on Jun 28th, 2010, 8:06am:
Sorry PR is a mistake. I means 50 Ohm resistor paralleling with RF PORT and LO PORT. pancho_hideboo wrote on Jun 28th, 2010, 8:06am:
You say 50 Ohm resistors are part of my circuit. Yes, mixer is a wideband mixer. But I do not want this resistor as possible as I can. It seem than match net work is the only way. But for a wideband, it may be a little difficult work. Any other good method? |
Title: Re: CG and matching about mixer Post by pancho_hideboo on Jun 28th, 2010, 10:27am See http://www.designers-guide.org/Forum/YaBB.pl?num=1231076499/11#11 steven_l wrote on Jun 28th, 2010, 10:11am:
But it is not appropriate to argue Vpp since PAC is a small signal analysis. steven_l wrote on Jun 28th, 2010, 10:11am:
If you do, you can understand. steven_l wrote on Jun 28th, 2010, 10:11am:
pancho_hideboo wrote on Jun 28th, 2010, 8:06am:
steven_l wrote on Jun 28th, 2010, 10:11am:
Here "E" means an open circuit voltage of port. steven_l wrote on Jun 28th, 2010, 10:11am:
See http://www.designers-guide.org/Forum/YaBB.pl?num=1266168115/1#1 I'm not kind and patient enough for explaining these too basic issues. If you can not still understand, other people who is very kind and patient will help you. |
The Designer's Guide Community Forum » Powered by YaBB 2.2.2! YaBB © 2000-2008. All Rights Reserved. |