Gunn diode hacking / General Science and Electronics / Forums

Forums

4hv.org :: Forums :: General Science and Electronics

« Previous topic | Next topic »

Gunn diode hacking

Move Thread

LAN_403

Conundrum

Fri Mar 30 2012, 06:36AM

Registered Member #96 Joined: Thu Feb 09 2006, 05:37PM
Location: CI, Earth
Posts: 4061

Hi all.
Just dug out of one of my drawers that Gunn module which I changed the diode on.
The original was evidently damaged, as it had zero resistance and no oscillation whatsoever.

Problem is, these are a right PITA to retune; the replacement diode is a 10 GHz and similar mechanical
size and operates around 150mA.

Any ideas?
I can get very weak output according to the mixer diode at around 10V.
Thanks!

Also, another um, "creative" idea. An infrared burner diode of the "open can" variety is essentially bulk GaAs with a PN
junction and a resonator cavity.
So, if I had say a LED'd IR diode which was then carefully butchered with Epoxy etc and mounted in the original cavity, then further tormented by precise applications of controlled reverse breakdown voltage while kept cold, at some point just before the bond wires give out there should be a point where the junction melts and resolidifies into something approximating a Gunn diode junction ie N+ N- N+ .... ?

edit: I might have actually made this work in the past as had a "Hot Diode" which knocked out WiFi when powered.

see

Sound possible?
My inspiration is from Jeri's and others experiments with electroforming of PN junctions in germanium etc.
In fact some of the very early point contact transistors were made in this way (!)

-A

Proud Mary

Fri Mar 30 2012, 09:10PM

Registered Member #543 Joined: Tue Feb 20 2007, 04:26PM
Location: UK
Posts: 4992

Gunn diode oscillator frequency is usually adjusted by varying the DC bias via a high Z stripline - the microwave equivalent of an RF choke.

Mattski

Sat Mar 31 2012, 03:39AM

Registered Member #1792 Joined: Fri Oct 31 2008, 08:12PM
Location: University of California
Posts: 527

You won't have much control over the doping profile if you're just heating it up to spread around the existing dopants. PN would turn into something (very approximately) like P P- N- N, then P- i N- as the donor dopants will cancel out the acceptor dopants. Most GaAs structures are grown layer by layer for maximum control of the doping profile, and to enable heterostructures.

But if you got your hands on a pre-doped wafer at a proper N- doping level and just worked on depositing ohmic contacts on it (that's what the N+ is for) then you would have a better shot at it.

Conundrum

Sat Mar 31 2012, 07:47AM

Registered Member #96 Joined: Thu Feb 09 2006, 05:37PM
Location: CI, Earth
Posts: 4061

Thanks guys, I'd heard about using YIGs which can be found in most old satellite LNBs.
seems that to get a larger change in frequency you just use more of them, and then apply a magnetic field to tune them,
shame I can't just get them ready made.
-A

Proud Mary

Sat Mar 31 2012, 09:21AM

Registered Member #543 Joined: Tue Feb 20 2007, 04:26PM
Location: UK
Posts: 4992

Don't you think it would be more usual to effect larger changes in Gunn frequency using a varactor diode?

If you want to get involved with Gunn and HEMT experiments, certainly the simplest way in is using active antenna oscillators.

Google active antenna oscillator or patch oscillator or active patch oscillator and you'll quickly find designs
giving circuit layouts and patch dimensions that you could put together in a day.

Detecting and measuring your microwaves is another matter!

Conundrum

Sat Mar 31 2012, 12:33PM

Registered Member #96 Joined: Thu Feb 09 2006, 05:37PM
Location: CI, Earth
Posts: 4061

Yeah, I recall reading an article about using "flat" plate capacitors as a radar detector.

another option is to use a satellite LNB, to pick up the reflection from a convenient wall.

-A

Mattski

Sat Mar 31 2012, 08:37PM

Registered Member #1792 Joined: Fri Oct 31 2008, 08:12PM
Location: University of California
Posts: 527

Mary's right, a varactor will be much easier for tuning. a YIG resonator will offer a much higher Q but it's more complicated. You can get YIGs ready-made, but they are not cheap. And something you salvage from a satellite TV receiver might not work well at the frequency you want, satellite TV I think is Ku band not X band.

Also keep in mind that if you are transmitting any high-power signals you generate you may well be violating certain FCC rules. At the very least you should direct any transmitter away from anything you might interfere with, and only run it for short periods of time.

Conundrum

Wed Mar 18 2015, 06:38AM

Registered Member #96 Joined: Thu Feb 09 2006, 05:37PM
Location: CI, Earth
Posts: 4061

Thanks for that!
I obtained four old MW (22GHz) burglar alarm modules, one works and three don't.
Reckon tin whiskering would be the problem as there is almost no voltage (300mV) across the diode and tin whiskers have been a problem ever since the first transistors were made.

EDIT: Tin whiskering could indeed be the problem.
Trouble is to nuke the whisker would likely burn the active region causing the diode to be permanently ruined.
Does anyone have another idea?

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob