I have seen plenty of published circuits for audio frequency white/pink noise generators.
What is involved in producing noise up into the low GHz region?
I don't care if it is flat, or other expensive things, so long as it is of sufficent strength for equipment testing.
Mark Henderson
If you aren't too picky, a pseudo-random sequence generator built with a EC LinPS shift register and an ECLinPS exclusive-OR should get up to a GHz or so. The clock oscillator might be interesting. John Larkin plays with that sort of stuff.
Horowitz and Hill's "The Art of Electronics" from section9.33 to the end of the chapter 9 gives chapter and verse. Using a sinc x (sine x/x) weighted resistor sequence to make FIR filter is covered. When I did that - a year or so before the book was published - I found out the hard way that you hav e to apply a Hamming window to your sinc x values to avoid Gibbs oscillatio ns - ripple - on the low-pass filter profile.
-- Bill Sloman, Sydney
For low level stuff, just a resistor. A 50 ohm resistor makes about 1 nV/rootHz at room temperature.
More noise, people generally use a "noise diode" which is just a selected zener diode. A 10 volt zener biased to a few mA will make something like 300 nV/rtHz. Bandwidth depends on junction capacitance and dynamic impedance.
Photodiodes are nice noise sources, because the Johnson and shot noise levels are predictable.
People used to use gas discharge tubes in waveguides to make microwave noise; don't know if they still do.
-- John Larkin Highland Technology Inc www.highlandtechnology.com jlarkin at highlandtechnology dot com Precision electronic instrumentation
Oh, zenering the b-e junction of a microwave transistor will probably make wideband noise. Regular zener diodes have a lot of capacitance.
Hmmm, there's a 3 GHz spectrum analyzer on my bench just now. Maybe I'll try it.
-- John Larkin Highland Technology Inc www.highlandtechnology.com jlarkin at highlandtechnology dot com Precision electronic instrumentation
You beat me to it. That's what I was going to suggest. Grumble.
I built various fluorescent tube based RF noise source over the years. The microwave variety were stuffed into waveguides. The lower frequencies were pickup coils (or foil capacitors) wrapped around the tube with amplifiers of varying denomination for isolation. No need for waveguide. I just wrapped it in aluminum foil shielding.
More on the topic: My favored tube was the small 6 inch, T-5, 4 watt, tube, running on about 100V of DC. There are also smaller CCFL (cold cathode fluorescent) tubes used for LCD backlighting, which might work, but I haven't tried them yet.
Still more: or look at what others are doing for making noise: Most of these zener diode or reversed BE junction noise generators will belch something that looks like noise to at least 1GHz, but don't do well much beyond that.
If you have money:
-- Jeff Liebermann jeffl@cruzio.com 150 Felker St #D http://www.LearnByDestroying.com Santa Cruz CA 95060 http://802.11junk.com Skype: JeffLiebermann AE6KS 831-336-2558
A 50 ohm resistor feeding into a few cascaded MAR-x MMICs ?
Of course you need to shield the resistor and first stages very well to avoid external signal intrusion.
6.8v zener driving a MMIC product. Scroll down to: "N-gen Wideband Noise Generator. (100 kHz - 500 MHz)" Schematic: If you replace the MAR-1 with something that works up into the "low GHz" region, it should be suitable. About $60 for the kit.
-- Jeff Liebermann jeffl@cruzio.com 150 Felker St #D http://www.LearnByDestroying.com Santa Cruz CA 95060 http://802.11junk.com Skype: JeffLiebermann AE6KS 831-336-2558
If you do, please let us know how it turns out. Sounds like a tidy solution.
Mark Henderson
For low noise stuff, I expect that you can do a lot worse than a Tek SD-42 plugin with ~ 1 mA of photocurrent. DC-6.4 GHz, pretty accurately known frequency response, no impedance mismatch funnies when you switch the source, NF of 0 dB dark, 3 dB with light.
I'll have to give that a whirl myself. Lots easier than LN2.
Cheers
Phil Hobbs
-- Dr Philip C D Hobbs Principal Consultant ElectroOptical Innovations LLC Optics, Electro-optics, Photonics, Analog Electronics 160 North State Road #203 Briarcliff Manor NY 10510 hobbs at electrooptical dot net http://electrooptical.net
-- I've got some good stuff. You should really envy me and pine for my 'gift'. John Fields
I work in an electronics engineering company. We have test equipment. Nothing unusual about that.
-- John Larkin Highland Technology Inc www.highlandtechnology.com jlarkin at highlandtechnology dot com Precision electronic instrumentation
-- http://www.youtube.com/watch?v=Oi84maqHgxg
Johnny! Stop that.
There's a 22 GHz spectrum analyzer on my bench (or it would be if I could lift it out of the rack). ;) Cost $960 plus shipping. Even mutts can afford that.
Cheers
Phil "Not a dog person" Hobbs
-- Dr Philip C D Hobbs Principal Consultant ElectroOptical Innovations LLC Optics, Electro-optics, Photonics, Analog Electronics 160 North State Road #203 Briarcliff Manor NY 10510 hobbs at electrooptical dot net http://electrooptical.net
The Rigol DSA815 looks pretty good. About $1300, and you can lift it with one hand. The tracking generator is $200 more, pretty good price for a 1.5 GHz signal generator.
Rigol make pretty nice gear.
-- John Larkin Highland Technology, Inc jlarkin att highlandtechnology dott com http://www.highlandtechnology.com
All the SDR-type analyzers have really crappy phase noise compared to a YIG-tuned boat anchor, though. Like 30 dB crappier below 10 kHz offset.
If I were doing RF installation work, I'd probably get one, but since I'm rarely more than ten feet from my rack, there's not much incentive to compromise.
Cheers
Phil Hobbs
-- Dr Philip C D Hobbs Principal Consultant ElectroOptical Innovations LLC Optics, Electro-optics, Photonics, Analog Electronics 160 North State Road #203 Briarcliff Manor NY 10510 hobbs at electrooptical dot net http://electrooptical.net
Here's a BFT25 b-e junction as a zener. Zener voltage is 5.2 at 1 mA, and the noise level looks best at about 1 mA.
Looks like about 25 nV/rootHz. Mediocre.
-- John Larkin Highland Technology, Inc jlarkin att highlandtechnology dott com http://www.highlandtechnology.com
Thanks John. That will given me a benchmark to aim for.
Mark Henderson
You might look up noise diode datasheets from Noisecom and their arch-rival Noisewave.
-- John Larkin Highland Technology, Inc jlarkin att highlandtechnology dott com http://www.highlandtechnology.com
A spark gap can be good from DC to light...
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.