Series Resonant VHF Oscillator Blues!

If I was that circuit, I sure wouldn't want to oscillate.

Get Randy Rhea's paperback "Oscillator Design and Computer Simulation."

John

Reply to
John Larkin
Loading thread data ...

I can't understand how it could ever oscillate. And, since you are scanning from 87 to 108 MHz. I'm going to make the educated guess that you copied this circuit out of some pirate FM radio transmitter book and are trying to make it work. People make intentional mistakes in these sorts of documents strictly to KEEP people from making something that will work on the hope that you will buy one of their "wired and pretested" units.

THere are dozens of good FM wireless microphone designs. THis isn't one of them.\\

BTW, believing Electronic Workbench at VHF is really very silly. THe capacitors don't have inductance and the inductors don't have capacitance. Resistors have neither. In the real world, this is exactly the opposite of the truth.

Jim

Reply to
RST Engineering (jw)

Sheeesh! I regularly simulate at above 2GHz. [snip]

...Jim Thompson

--
|  James E.Thompson, P.E.                           |    mens     |
|  Analog Innovations, Inc.                         |     et      |
 Click to see the full signature
Reply to
Jim Thompson

I agree with the rest of the guys; I wouldn't suspect this circuit of oscillating.

What's your simulation time step? You can get just an L-C to oscillate if you set that coarse enough. Hell, you can get an R-C to oscillate.

John

Reply to
John Larkin

I simulate picosecond stuff now and then. Of course, I poke in estimated parasitics and sort of assume that I don't really have it

100% right. But the sims do help get it sort of close to working, and train one's instincts about what the causalities are.

Small surface-mount parts pretty much work as lumped components in the, say, 200 ps domain. There's sort of a wall at around 100 ps, kind of like the sound barrier.

John

Reply to
John Larkin

Hello Kevin,

ARRL Handbook. The Art of Electronics (not transmitters but BJT oscillators)

Well, a 2200pF cap is pretty much like a short at VHF frequencies. Also, I can't make out a proper feedback path.

Regards, Joerg

formatting link

Reply to
Joerg

Hello Jim,

I can't either.

There are also some good "FCC Part 15" sites with schematics and all, for the low power xmit community in the US. But it seems Kevin lives in Ireland where that stuff is most likely not legit.

Simulators are also not very good in predicting the effects of Murphy's law. Although I prefer Murphy's stout...

Regards, Joerg

formatting link

Reply to
Joerg

Hi all,

formatting link

I have a link to a screen shot of an RF circuit I simulated in EWB 2001 and also built two prototype models. I can't get the dam thing to resonate. I've scanned from 1 ~500Mhz and no transmission at all. Could some one point me in the direction of a book or web page that details a BJT VHF transmitter that works! Can anyone see any flaw in the design I have? Should there be a ratio of L to C or XL to Xc in the resonator?? Please help.

Cheers, Kevin.

Reply to
Kevin

Hello Kevin,

I'd contact the author and ask him if he or someone else got it to run reliably.

None. Anything above 100MHz I calculate, design and then prototype. Simulation does happen up there but only for phase shifting gear, chip level circuits and similar stuff. Not for discrete oscillators.

For the rest I use Spice. If you want to try that LTSpice is free and very good. Downloadable from Linear Technology's web site.

Regards, Joerg

formatting link

Reply to
Joerg

Hello Jim,

Well, yes. That's why I excluded chip level circuits ;-)

The same goes for picosecond stuff which is simulated a lot. Mostly because it becomes next to impossible to measure. You move a wet toothpick close to the circuit and everything already changed. Let alone a scope probe.

Regards, Joerg

formatting link

Reply to
Joerg

details

are those 2 caps realy meant to be 2.2nf not 2.2pf ? then there is a chance you will enough feedback in the series LC feedback network with hopefuly 180 phase shift, otherwise they just kill everything.

Colin =^.^=

Reply to
colin

Hi ,

formatting link

The above link is the same oscillator with the feed-back loop closed and the bode plotter removed. The oscillator is forced to start with the AC voltage source into the 1 Meg resistor.

When I build this exact model I cant get it to work. Any ideas? I have noted the 2.2nF caps Joerg, thanks for pointing that one out.

Cheers, Kevin.

Reply to
Kevin

The design was taken from one of my many books, Complete Wireless Design by Cotter W. Sayre.

Reply to
Kevin

Hello John,

In one case the causalities were more like casualties. A client had a worrisome field failure rate on an ultrasound front end. Even with the fastest scopes in their lab we couldn't see much. Then the sims showed a tiny little carrier lifetime effect in a fast diode that was most likely causing the preamps to die. A small circuit change and the problem went away. It really puzzled them because they could not see any difference on the scope.

Yes, that's where I prefer hybrids. It avoids the "body effect". Unfortunately this technology dropped in popularity. We used to do stuff with them that is otherwise only possible on a chip, at a lot more $$$.

Regards, Joerg

formatting link

Reply to
Joerg

"Kevin" schrieb im Newsbeitrag news:_NQfe.52914$ snipped-for-privacy@news.indigo.ie...

and

details

Hello Kevin,

I can't see why it should oscillate.

Why such big 2.2nF caps? It seems they are 2 decades too high.

Where is the feedback?

Please provide a link to the article where this "oscillator" circuit has been used.

Best regards, Helmut

Reply to
Helmut Sennewald

and

details

Lose the forcing oscillator thing and drop the 2n2's to 220p. It then runs quite fairly at 110MHz, 3Vpp. Need to allow though, about 2uS for it to startup. Maybe your sim is bypassing this.

Reply to
john jardine

audiophoole stuff then......(sorry, I'm bored)

martin

After the first death, there is no other. (Dylan Thomas)

Reply to
martin griffith

Dear Kevin,

Definitely, even in you closed-loop schematic, your C1 and C2 caps are far too large, as stated by others. Looking at the formula in Complete Wireless Design (I love that book!) on page 228 for these caps shows them to be in the 66pF range at 100MHz, not 2.2nF. I didn't even look at your other passive's values vs. Page 228. I have designed a few oscillators at HF, VHF, and UHF (three using the formulas in Sayre's book) and they all worked fine. *However*, as the frequencies increase, so do not only the problems with real-life vs. ideal component's reactances, but so does PC layout criticality. In other words, you will always have to tweak a circuit design when the frequencies get over a few megahertz -- the higher the frequency, the more you will have to tweak. Tweaking will become less radical if you model the oscillator (or any circuit) more accurately by using S-parameter models or equivalent circuits in place of the ideal components (just specifying 'Q' is not enough). The models made by Modelethics are the ultimate devices to use for your passives. And Spice is not the correct tool for your VHF frequencies though: Eagleware's Genesys, AWR's MWO, Agilent's ADS, Ansoft's Serenade would be the real tools you would need. But these programs can be very, very (very!) expensive, so you will want to simulate with the Puff linear simulator that came on the CD for Complete Wireless Design (at least it came with my copy), and then use open-loop simulations with s-parameter models for all the passives and the transistor. Its really the only accurate way to do this. After you complete the oscillator design via this method, you can then see the oscillator's closed loop output amplitude and spectrum in Spice, but at these high frequencies this is not really a good option at all.

Cheers,

Bill

Reply to
billcalley

Hi Bill, Thanks for your reply. I actually got the oscillator work!! I changed the value of the cap from nF to pF and it work fine. I have since built the model on PCB with an mps5179 as the active device and it worked a treat at106Mhz. I cant get s param, files for the mps5179 so I'm going to redesign with the bfg97 device. I have the s param files for this device and I also have just stocked up on them from farnell.

Reply to
Kevin

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.