Another "RF Signal Generator"

[pwdrive]

Judas Priest's 'Breaking the Law' would have been more appropriate I suppose

- Joe

Or maybe The Clash - 'I fought the law and the law won'? (Well not yet it didn't!):

Breakin' rocks in the hot sun
I fought the law and the law won (twice)
I needed money 'cause I had none
I fought the law and the law won (twice)...

Well done Joe - looks very neat, and the scope traces look nice and clean. I don't think Ofcom will be knocking on your door any time soon!

http://www.youtube.com/watch?v=OgtQj8O92eI

Lawrence.

[Yorkie]

Well I never! Thanks for that Lawrence, and what a great version. In my ignorance, I had been unaware who wrote and first performed the song, and assumed that that 'The Clash' version was the first. Very interesting and rather sad write up about Bobby Fuller on Wiki: http://en.wikipedia.org/wiki/I_Fought_the_Law

It was one of the songs in Ade Edmonson's repertoire in his 'Bad Shepherds' 'punk meets folk' band, which has now come to an end. Good while it lasted - they covered songs such as London Calling, Tainted Love, Whitehall's Up Against the Wall, Up the Junction, Down in the Tube Station at Midnight and so on. I won't take the thread even further off topic - suffice to say that anyone who's interested will find the 'Bad Shepherds' on youtube.

Hopefully, Joe is still 'at large', still actively experimenting, and hasn't yet had his collar felt by Ofcom!

[pwdrive]

Am repllying to Davids last post in the chat room. (bobby Fuller)

Lawrence.

[Joe]

Yes, I'm still at large (as well as still being large!).

I had a further play this morning and there seemed to be an intermittent fault. I couldn't see any dry joints, shorted or broken tracks or anything so decided it was time to get some proper connectors fitted as all the croc clips wouldn't be helping. I also fitted the power supply components.

When I next tried it I found that the LM317 wasn't regulating. This had me well an truly puzzled and eventually I took it off the board to try to get it going on a breadboard. This quickly revealed that despite being a new National Semiconductor one the chip itself was u/s.

I can now find no signs of oscillation so one of the three-legged fuses may have copped a packet, though they shouldn't have as I was on a current-limited supply. After the afternoon's battle with the LM317 I'd had enough and will investigate tomorrow.

- Joe

[Joe]

OK, I admit defeat, or at least confusion.

I've been fiddling about with this for the last few days and have discovered that if I connect my 'scope to the rail indicated by the arrow below there's not only a waveform but the thing works. As soon as the probe is taken away it shuts up again. It doesn't matter exactly where it's connected so it isn't that touching a particular component and temporarily making a dry joint conduct.

   

I did wonder about the health of the second transistor, the only one shown on the above snippet, but changing it has made no odds and the original tests good.

Any ideas as to what I should investigate next? Like blanket cap changes it would have been far quicker and easier to just make another but not as satisfying.

Cheers

Joe

[pwdrive]

Hi Joe, I asume the B+ is the correct voltage, that aside I am trying to figure out what the osc. configuration is, it looks like a series tuned part Collpitts? I would say that it's bursting into life when connecting your scope probe because you are adding extra capacitance to the series leg, it sounds as though it's just below critical feedback, maybe check the inductor and the split capacitor network, asuming the transistor and DC voltages are something like. I would normally expect to see the end of the series leg connected to earth and not the emitter, any oscillator gurus out there?
Edit: Maybe it's a Clapp.

Lawrence.

[Joe]

Thanks, Lawrence.

As I said in the original post, my version has two parallel-connected 1000µH inductors - one mounted vertically and one horizontally to avoid mutual inductance. There was only 0.3Ω difference in the DC resistances of these so they seemed OK.

I found that connecting my DMM where the 'scope had been had the same effect and then discovered that connecting a 47pF capacitor there, ie between the highlighted rail and ground, seemed to fix it nicely.

I was puzzled that it had worked initially. I can only think that as it was on an 18" long board there was enough stray capacitance kicking about.

I've spent so long today looking at it that I've given myself back-ache so I'll have a better idea of what's what tomorrow and I can also get on and case it up.

Cheers

Joe

[pwdrive]

HI Joe I was waffling on a bit in my last post, I could not figure out what type of osc. it was, if the bottom of the series leg was connected to ground then I would say it's a modified Collpitts ie: Clapp.
The osc circuit is a new one on me, having said that I sometimes have a short memory..

Lawrence

[Skywave]

Hi Joe: welcome the world of invisible components at R.F.!
Yes, I think you are correct: your initial lash-up had various bits of stray L and C that are no longer there in your tidied-up version: a very common occurrence at R.F. Having said that, I see that your R.F. choke arrangement is different to what the original design was: that could also account for what you have found. I have made a post here in response to a Thread started by David (Yorkie) that discusses such a situation as regards R.F. chokes. At least you took the wise precaution of arranging your two chokes to avoid mutual inductance coupling: not everyone would have thought of that.

Al.

[pwdrive]

Maybe the Q of the tuned circuit is a bit low.

Lawrence.