How to Model an SRD Diode

[Marios91]

Hello guys,
I am designing an UWB pulse generator for my Diploma thesis,but i do not have a step recovery diode model to use in my simulations.After 5 months of searching i have concluded that unfortunatelly is almost impossible to find such a model.In a paper i saw that they measured a actual srd and then used the results into the simulation.If i possess a srd what measures should i do in order to model the srd?

 

[volker@muehlhaus]

https://www.macom.com/products/product-detail/MA144769-287T
SPICE model on the "Technical Resources" page.

 

[Marios91]

Have you tried this,is it working?

 

[biff44]

I looked at that so-called spice model, and it is not accurate at all. there is not transit time/charge storage effects shown, which are pretty much the entire point of an SRD.

you need some sort of charge storage thing so that, briefly as the driving sine wave excitation reverse biases the diode, diode current STILL FLOWS for a short while. THEN the diode current abruptly drops to zero as the last bit of charge stored in the diode intrinsic region is finally absorbed (either drifted to be collected by the reverse bias Efield or recombine). it is this abrupt drop in reverse current to zero that makes the SRD form a sharp impulse spike in the driving inductor that is part of a multiplier circuit.

 

[Marios91]

I agree with biff44, i gave a shot with the given model and it didn't work.I am going to buy some SRDs, if i make mesurments on it ,is there any way to use these measurments in my simulation? I am using the Agilent ADS 2009 simulator.

 

[volker@muehlhaus]

Ok, for the old ADS version (2009 and earlier) there is a document and example in the ADS support center, title "ADS 2005A: A model of Step Recovery Diode (SRD)", Example ID: 251048

**broken link removed**

"Background:

In 1995, Zhang and Räisänen proposed a simple extension to the standard diode model which is capable of simulating the “snapping” behavior specific to SRDs. Although the model may suffer from convergence difficulties in harmonic balance simulations of some circuits, it may still prove useful.

The Zhang and Räisänen model was used in Eagleware App Note 22 Nonlinear Modeling of Step Recovery Diodes Using Verilog-A within the context of developing a user-written Verilog-A module. The actual implementation was left to the user.

ADS Implementation:
Here we provide the Zhang and Räisänen model, implemented as a subcircuit and packaged as an ADS design kit. It is immediately ready to use without any programming. Simply download the SRD Design Kit and install it using DesignKit > Install Design Kits… menu pick in the main ADS window."

 

[afz23]

The referred paper is verilog based SRD model,which need to be implemented in ADS.
If some one can help implement this in ADS,I will be very grateful.

 

[volker@muehlhaus]

The referred paper is verilog based SRD model,which need to be implemented in ADS.

The original paper by Eagleware describes a Verilog implementation, but later it was implemented in ADS without Verilog. The ADS example can be downloaded from the ADS support pages, as I described above.

 

[vfone]

The SRD CVB1031 datasheet from Skyworks gives more data about the Spice model.

https://pdf.dzsc.com/20130115/CVB1031_2456268.pdf

 

[afz23]

SRDs are the most nonlinear devices available,and
toughest to model,I have worked with SRDs for
designing the
higher order frequency multipliers.I used Teklec SRDs
for my designs,only problem is that,you need to provide
post fabrication tuning options to make it work.
I haven't come across an SRD model,which really
works
,i.e simulation results matches with practical circuits.

Whether these ADS SRD models are accurate,worth
trying,
Please give your opinion,share your experience working
with these models.
Any way the keysight link on ADS model does not open.

 

[Marios91]

Yes unfortunatelly most of the Spice models i have come across aren't accurate enough to work properly and the link for the ADS model given by Keysight is down.I had found it in a chinese tracker but you needed invitation to join it.The model suggested by vfone looks promising ,i'll give it a shot on tuesday and tell you the results!

 

[Marios91]

Moreover, i tried the verilog file a couple of months ago and it hit me with some syntax errors(the compiler didn't recognised a specific function).Was there anyone with the same problem?

 

[volker@muehlhaus]

and the link for the ADS model given by Keysight is down.

It is possible that my link has customer ID that doesn't work for you. But if you log in properly, the download is available. Here si what to search for:

 

[afz23]

I have checked the model,as suggested by Volker at Agilent(key sight) website.
The model is working fine,is able to generate impulse/comb,can be tried for designs.