Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics 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.

Problems with transient Simulation in Spectre with resonant tanks and rectifiers.

Status
Not open for further replies.

vamsitalla

Newbie level 3
Joined
Jul 31, 2012
Messages
4
Helped
0
Reputation
0
Reaction score
0
Trophy points
1,281
Location
Seattle
Activity points
1,321
I am trying to design a rectifier block for an inductive link. However, I am facing issues with transient simulation in Spectre, Cadence. As an example following is a simple resonant tank driven by an ac source with 50 ohm input impedance. The ac source is modeled by a sine voltage source with 50 ohm resistance in series. Series resonant tank consists of an ideal capacitor in series with an inductor (1 ohm parasitic resistance is included in the model). The circuit diagram is shown below:

https://images.elektroda.net/94_1343785574.png

The signal for different nodes are shown as follows:

VT("/GND!")
https://images.elektroda.net/11_1343785682.png

VT('net8')- Output of voltage source
https://images.elektroda.net/6_1343785783.png

VT("/net02")- At the top of inductor
https://images.elektroda.net/32_1343785855.png

Few details about the circuit:

1. The voltage source , amplitude = 1V, freq = 13.56 MHz
2. Following warnings during transient simulation.

Notice from spectre during topology check.
No connections to node `0'.
No DC path from node `net02' to ground, Gmin installed to provide path.

I am in a fix and can't figure out this basic circuit topology. any help in this matter will be appreciated.

Thanks,
Vamsi
 

D.A.(Tony)Stewart

Advanced Member level 5
Joined
Sep 26, 2007
Messages
7,059
Helped
1,721
Reputation
3,441
Reaction score
1,716
Trophy points
1,413
Location
Richmond Hill, ON, Canada
Activity points
51,811
Decide how much amplification you want. Right now you have a Q of 6 which is the impedance ratio of the reactance to resistance at resonance. I estimated the reactance ~300Ω /50 =6 , 2pi f L=300 = 1/(2pi f C). If you want a Q of 1 make the ratio of Zc/Rs=1

...that means make the L 6x smaller and the C 6x bigger...if you want Q=1

Not sure what this is for....maybe you want high Q....

Pay close attention to layout.. and keep parts close together and on a ground plane.

Where is your rectifier?...keep in mind that will drop the Q. if unbuffered. use a common collector NPN with load cap followed by PNP CC with load cap for zero offset or an OPA. with diode feedback and charge cap for a peak detector...
 
Last edited:

vamsitalla

Newbie level 3
Joined
Jul 31, 2012
Messages
4
Helped
0
Reputation
0
Reaction score
0
Trophy points
1,281
Location
Seattle
Activity points
1,321
Decide how much amplification you want. Right now you have a Q of 6 which is the impedance ratio of the reactance to resistance at resonance. I estimated the reactance ~300Ω /50 =6 , 2pi f L=300 = 1/(2pi f C). If you want a Q of 1 make the ratio of Zc/Rs=1

...that means make the L 6x smaller and the C 6x bigger...if you want Q=1

Not sure what this is for....maybe you want high Q....

Where is your rectifier?...keep in mind that will drop the Q. if unbuffered. use a common collector NPN with load cap followed by PNP CC with load cap for zero offset or an OPA. with diode feedback and charge cap for a peak detector...

I need high Q because I am trying to do transfer power through the link. I agree that rectifier would reduce the Q of the tank. However, the point of the post is not to discuss the circuit topology but to solve the problem with transient simulation in Spectre, Cadence. I have demonstrated a very simple circuit and shown the plots that I get from transient analysis. The plots are inconsistent and make no sense. If I simulate the exact circuit using LTSpice I get the expected and corrected output.

I believe there is an issue with my transient simulation circuit. For example, the ground signal should be 0V but it has ac signal. Similrily, the output of voltage source is not correct. Can anyone tell me why the transient simulation in spectre is screwed up and how I can fix it!

Thanks,
Vamsi
 

BradtheRad

Super Moderator
Staff member
Joined
Apr 1, 2011
Messages
14,232
Helped
2,816
Reputation
5,640
Reaction score
2,756
Trophy points
1,393
Location
Minneapolis, Minnesota, USA
Activity points
106,166
Is your ground attached in the middle of a wire? Or is it connected to a meeting of two wire ends?

The simulator might recognize one as usable, but not the other.
 

vamsitalla

Newbie level 3
Joined
Jul 31, 2012
Messages
4
Helped
0
Reputation
0
Reaction score
0
Trophy points
1,281
Location
Seattle
Activity points
1,321
Is your ground attached in the middle of a wire? Or is it connected to a meeting of two wire ends?

The simulator might recognize one as usable, but not the other.

You are a life saver... thanks a lot...that was the problem.. the ground should be connected in one of the corner and it fixed everything, this is a pretty bad flaw with Spectre in my opinion, it gives no warning!

Thanks a million for your help!

Vamsi
 

Status
Not open for further replies.

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Top