Varactor-tuned oscillator

"fourblackcats"

the temperature related frequency drift with varicaps is severe.

See fig 4

It is circa 2 exp -4 per degree C.

So, for a 20 degree C rise in temp,

C varies by 0.4 % or 400kHz in 100MHz.

........ Phil

Hi Ken,

I'll try a 2n2222 as you suggest to see what it does to the tuning range. Might as well try a few others too. I don't have a fast enough scope to make those measurements you mentioned, so I wonder if I could adjust the phase relationship with a variable capacitor to see the effect on range?

I'm using a common-emitter arrangement, so isn't the base voltage normally

Allen

even

load

No folks, I haven't yet found my "dumb" error, but I agree that it must be stray capacitance in the tank circuit. I'm going to try that and also different transistors and possibly a hyper-abrupt varactor as was suggested in other posts. Thanks for all the good thoughts.

Allen

eliminating

caps

was

and

Delta-C of 0.4% results in delta-F of 0.2%, 200 KHz. The square root thingie.

John

Hooray for my cousin 2n2222!

However, with a nominal f sub t of 300MHz, it only has an effective beta of about 3 at the top end of the range. Possibly you want to run in faster company yet.

Regards PN2222A NPN (Is = 14.34f Xti = 3 Eg = 1.11 Vaf = 74.03 Bf = 255.9 Ne = 1.307 Ise =

Ikr = 0 Rc = 1 Cjc = 7.306p Mjc = .3416 Vjc = .75 Fc = .5 Cje = 22.01p Mje = .377 Vje = .75 Tr = 46.91n Tf = 411.1p Itf = .6

Vtf = 1.7 Xtf = 3 Rb = 10)

"John Fields"

......... Phil

Try BB109 or 640.

In article , fourblackcats wrote: [....]

Most of the CE arrangements have extra capacitors in them. How are you avoiding having them effect the output frequency?

The best way to think about a transistor in an RF circuit is to assume that, that little plastic package has an ideal transistor hidden and that transistor is surrounded in a whole mess of Rs, Ls and Cs. All these extra parts are so arranged as to delay and reduce the output of the transistor.

There is a lag from the "base" lead to the actual base of the transistor. There is a lag from a change in collector current to what you see on the "collector" lead.

Ideally, you want the transistor to put energy into the tuned circuit without it effecting the frequency. If it wasn't for these delays, I assume your circuit would do just that. With these delays, the phase of the output of the transistor lags its input so it tends to pull the frequency down.

At low frequencies, yes.

But a transistor has a frequency Ft at which beta drops to unity. And it has a low frequency beta B. So beta is flat from low frequencies up to a frequency of about Ft/B, then it rolls off with 1/f all the way to Ft and beyond. All along this rolloff slope, collector current lags base current by 270 degrees, not 180, because of the time delay associated with the frequency rolloff.

I think frequency Ft/B is the "alpha cutoff frequency", the point where the common-base gain drops 3 dB, but I'm too lazy to look it up.

Actually, wirebonds and other nasties make things a bit messier, but the above is pretty much true. So most RF transistors operate well into their rolloff region, adding an extra 90 degree of phase shift. Added base and collector capacitance only increase phase lag from there.

John

I'd say you found a part of your problem. An ft of 200 MHz. means that you only have 6 dB of available gain at 100 MHz. That's gonna require a pretty heavy feedback to get it to kick off. Why not try something like a 2N918 or

Jim

And the more I think about it, the more I'm thinking that a device with ft of 200 MHz. is going to be a pretty fair size device; my bet is that the "stray" he's looking for is actually going to be Cob or some variant thereof.

Post the schematic to abse.

Jim

"John Fields" a écrit dans le message de news: snipped-for-privacy@4ax.com...

le

You also have diffusion capacitance: more voltage across the junction->more carrier there. That's just a capacitor. And since current rises more than voltage, its value is current dependant.

But is it really that much fun feeding the trolls and slipping them through our filters?

Thanks, Rich

"John Larkin"

........ Phil

"Rich Grise"

........ Phil

On Thu, 23 Mar 2006 00:50:52 GMT, Rich Grise Gave us:

That HAD to be how your lame ass got through.