Varactor selection and simulation with LTspice?

[MrSlack]

I'm designing a simple 40m VFO for a direct conversion QRP rig. Cut one of the VFO section is a simple W1FB inspired VFO with some part subs for what I have lying around, suitable padding for my tuning range and a more stable power supply. I've just slung this together this morning using a polyvaricon. It worked first time (miracle) and appears to kick out some relatively clean looking sines. I don't have a spectrum analyser (yet) so I'm not sure of real spectral purity but the simulation shows the 2nd harmonic at around -32dB which isn't terrible for a first attempt. Seems to be quite stable as well. I can see some drift at the moment but nothing scary unless I wave things around it or touch the polyvaricon (see later).

Unfortunately tuning is more than a little bit tetchy and not at all linear. I put this down to the crappy chinese ebay polyvaricon. I don't want to fish out for an air variable and a reduction drive for this plus they are a bit large and unweildy for what is going to end up a small device. So after sticking my face in TAOE 3e for a bit and some googling I'm going to attempt to use varactor diodes. I've never used one before but they appear to fit the bill nicely and I can move the control portion to a 10 turn pot and a stable voltage reference. I suspect this will kill the other issues with the polyvaricon as well as the varactor reference voltage has a relatively high output impedance and can be kept away from fingers.

Now I designed the entire thing initially in LTspice. Moving forwards however there is a varactor symbol in LTspice but there are unfortunately only two varactor models and they appear to be unobtainium. So questions:

1. What is a reliable generic jellybean varactor or series of varactors I can source easily that has a SPICE model?

2. Is there any way in LTSpice to plot an input voltage sweep against the first harmonic of the FFT of an output so I can check this for linearity? TAOE 3e points to non-linearity of tuning capacitance vs voltage but it's on a log/log scale so my brain can't process it well. It'd be nice to run a simulation run for 100mV steps vs 1st harmonic at least.

Any help appreciated

[MikeLogix]

Hello, I personally have not tried this but you should be able create your own model from a varacter data sheet and modifying the "standard.dio" file in LT Spice.
The file should be located in your ltspice directory under \lib\cmp

If you look in the file you will see single .model lines, and one for the MV2101
Here is the .model line for the MV2101
.model MV2201 D(Is=1.365p Rs=1 Cjo=14.93p M=.4261 Vj=.75 Isr=16.02p Nr=2 Bv=25 Ibv=10u Vpk=25 mfg=OnSemi type=varactor)

Now you can find a varacter you want to model, add a new .model in the file.
To decode all the parameters use the graphic below:

Also for a good app note explaining how to add the parameters and meanings click the link below.
https://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=1&ved=0ahUKEwiCyJTjy8LMAhUS32MKHal7AowQFgggMAA&url=http%3A%2F%2Fwww.skyworksinc.com%2Fuploads%2Fdocuments%2F200315B.pdf&usg=AFQjCNHjjgKCZDvBdQJBJH5TrfItNcBzRg&sig2=k4qxdtm1lR7JKJW7LMO6kg&cad=rja

The app note explains the parameters for the varacter that affect tuning range etc...
Note: The app note is for a different spice, but the abbreviations are still the same as for LTSpice. I think 
 Let me know if this works out for you:)
 

[nugglix]

Hi!

Had the same problem and found some help.
Also was pointed to the above mentioned reference and I could fetch the
attached little experiment in LTSpice.

I fiddled a bit and got close to the values from the data-sheet for my varactor.
(got a bag of them from china :)

Packed model and plot settings into one zip file.

[MrSlack]

Thanks both for your replies - much appreciated! I can progress with the modelling of this at least.

I actually breadboarded the circuit with a couple of BB149's from RS and measured it. I couldn't get the dynamic tuning range I wanted for the voltage I had available so I'm awaiting delivery of some other ones now (BB208) and have changed the tuning front end entirely.

[nugglix]

Hi!

Just stumbled upon this one, there are a lot of varactors in the extended lib.

http://ltwiki.org/?title=Components_Library_and_Circuits

Might be of interest for you.

My simulations are confusing, getting different frequencies when using startup and when not.
Strange...  need to build a few versions I guess.

Cheers

[MrSlack]

Excellent - thanks for that. I had no idea that site even existed.

[MrSlack]

A further update on this...

After thoroughly destroying a whole strip of varactors, I looked at some other diodes. Turns out your common garden red LED does the job quite well. I used the ideas here as inspiration: http://www.hanssummers.com/varicap/varicapled.html

So thanks to some inspiration from another forum member as well, I now have the following tuning arrangement:

* Stable tuning voltage reference provided by a 78L05 and divider to provide output up to 9v.
* Course tuning via 10k pot
* Fine tuning via 10k pot
* Summing amp based on the other 1/2 of the  LM358
* Hand selected (from about 50) generic red 5mm LED used as a varactor diode.

This gives me actually not too non linear tuning over 6.9-7.5MHz with no expensive air variables, nasty little SMD varactors or reduction drive. Drift of about 400Hz between 15oC and 25oC which will be improved when my CPC order arrives as I didn't use NP0 caps in a couple of places (because everyone was out of stock) and the assembly is poorly breadboarded deadbug at the moment.

Now for the mixer, front end attenuator and bandpass...

[olvinjanoisin]

I did some measurement when I searched diodes for using them as varicaps for vco. Found SS34 having pretty good range.
Unfortunately I get no any good result from leds.(!!!?) Red 5mm led gave just ~6pf@0v and yellow~3.4pf.

The BA124 capacitance diode was used as reference.