90MHz RF noise - how do I get rid of it on my cables?

There are double shielded cables costing just a tiny fraction more.

Rene

--
Ing.Buero R.Tschaggelar - http://www.ibrtses.com
& commercial newsgroups - http://www.talkto.net

Rene,

Are these "double shielded cables" still just two pins? What effect does the double shielding have?

There is also a large radio station in the area that broadcasts on

89.7FM - I think this may be what I'm seeing. Even if I do use a double shielded cable, whenever that cable is brought out to a device or PCB, the traces and all unshielded lines pick it up.

-William

I don't have any solutions but I have occasionally noticed the exact same thing. Same magnitude (small), same frequency (90MHz, give or take), and well you didn't mention scope so I don't know what you're seeing it on. Mine is a Tek 475.

I haven't tried, say, demodulating the thing and listening to it, but it's probably the nearby college station (90.3). I don't really like that theory, because it's not all that powerful a station.

Tim

-- Deep Fryer: A very philosophical monk. Website @

Likely the amp is oscillating (kill power, or touch its summing nod with a screwdriver and see if the frequency changes) or it's an FM station.

Maybe it's not "real" in the sense that it's only there when the scope is connected. The scope is a big antenna.

PMTs put out huge signals, even from single photons, so noise isn't usually a big issue.

Is the AD8015 located close to the tube, or is there a cable between? The TIA gain could have a high-frequency peak if it's driven from a significant capacitance, and might even oscillate. Usually a load resistor and a voltage amplifier is good enough to condition a pmt output; a tia may be overkill.

Just curious, but what's the physics?

John

Given an inverse square or inverse cube power to distance relationship (depending wether it approximates planar behavior or spherical) "nearby" counts for a lot more than "High power" does - and most college radio stations are down in the 90 mhz region, and most colleges have one - which will be "nearby" to any location on campus.

---

(replying to OP WilliamCox without bothering to paste in OP's text)

As for the "giant metal cage" that is what a screenroom is, and one test we performed to be sure the screenrooms were in good shape was to go into them with a portable radio and close the door - if the radio did not lose signal on FM, something needed work. If it didn't lose signal on AM, something major was wrong. So, a screenroom is one approach to get rid of the noise, yes.

Sounds like you need to get back to basics for shielding - no reason you should be running anything less well shielded than a standard BNC (typically RG-58) cable for your signals - your amplifier should be in a fully enclosed metal box with a BNC connector on it, and that box might well enclose as much of the PMT as possible, too. The lab I used to work in ran coaxial cables inside of copper pipe for as far as was possible for extra shielding.

Make use of the people around you, I'd bet some of the old hands can get you sorted out on practical applications of shielding. That's a part of the education you should be getting on your way to your Master's, but you won't get it if you don't ask, unless you have a really observant and helpful advisor with lots of time to see what you're up to and fighting against - but most will wait for you to figure out that you need to ask, even if they see that you need help.

Learning to ask for help (help, not the "do my masters thesis project for me" spams we see here from time to time, which yours is not an example of) is also part of the education you should be getting. While asking here is a step, someone local to you might be of more direct help than the folks in the newsgroup. If you do, in fact, have a station at your college, check to see if it goes off-air in the wee hours of the morning - and if it does, come in at that time and see if the signal also goes away. I don't recall any part of a running nuclear reactor that should be tossing out 90MHz signals - but the techs that work with it would be, if there are any.

-- Cats, coffee, chocolate...vices to live by

If it's an outside signal getting into your measurements, you can try wrapping a few turns of your scope probe cable around some ferrite. Big EE cores are very handy for taking care of these sorts of problems. Sometimes, wraping other wires around ferrite going to you circuit can be helpful, like power supply wires. Ferrite, it's religion! It can heal a multitude of problems with just a touch.

If you have a spectrum analyzer handy, you can look for the offending signal and measure the frequency. Look in your local broadcast listing to see if any local station matches the frequency or tune your FM receiver to the offending frequency. You can also stick your finger in the opamp circuit to see if the frequency is stable. If it changes, then your circuit is doing something evil.

Mark

That's really easy to test. Get an FM radio, tune in that station, look at the 90 MHz on the scope,, and see if it "dances" along with the audio.

Good Luck! Rich

When I was at the U. of MN, the radio station used the power wiring in lieu of an antenna.

So it could be coming in on the power lines.

Cheers! Rich

That seems like it would work exceptionally poorly relative to a real antenna.

Although I suppose it's a lot better than nothing...

Lots of universities do that. Low power, but easy distribution primarily just to the campus.

Like WTBS (Tech Broadcasting System) at MIT until Ted Turner bought the call letters for mucho bucks ;-)

...Jim Thompson

--
|  James E.Thompson, P.E.                           |    mens     |
|  Analog Innovations, Inc.                         |     et      |
|  Analog/Mixed-Signal ASIC\'s and Discrete Systems  |    manus    |
|  Phoenix, Arizona            Voice:(480)460-2350  |             |
|  E-mail Address at Website     Fax:(480)460-2142  |  Brass Rat  |
|       http://www.analog-innovations.com           |    1962     |
             
         America: Land of the Free, Because of the Brave

"Jim Thompson" wrote in message

Interesting... I was aware of, e.g., wireless systems in mines using "leaky coax" -- since it's still a well-controlled medium -- but not just the random power cabling within something as large as a campus!

I usually try putting ferrite torroids to form common mode chokes on the various signal leads and scope/uut power supply leads too.

it isnt usually a cure but sometimes helps just that little bit.

Colin =^.^=

What is the design bandwidth of your system?

Rich Grise hath wroth:

Ummm.... you're not going to see anything on an oscilloscope. The modulation is FM and the deviation is not (easily) visible. You might see something if there are any AM modulated subcarriers but those are very difficult to see on a scope. You might check the local FM stations operating schedule and see if the interference follows their schedule. Also, if you have a computing scope, you can possibly convince it to demodulate the FM.

Also, it's not coming down the power lines. Power line radio is commonly done for the AM broadcast band (about 1MHz), but not FM broadcast. FM is over the air. Most colleges have FM broadcast stations of one form or other:

All of these require FCC licenses. You can check the license information on the FCC CDBS lookup at:

The easiest way to get rid of such junk is to use differential amplifiers throughout. The common mode rejection will get rid of most sources of junk. Unfortunately, the AD8015 has a differential output, but not input.

Do you really need 240MHz bandwidth? If not, a simple low pass filter or ferrite beads at the input to the AD8015 will help. If you just want to get rid of the one frequency, a 90MHz series resonant LC circuit, between the input and the Vbp pin, will either get rid of just the 90MHz, or turn it into an oscillator. Anything that reduces the amplifier gain at 90MHz will work.

However, if you just need to get this thing done and not worry about elegance, look into adding lots of shielding or building a shielded box. Something fairly crude, like a carboard box, embalmed in aluminum foil, will suffice for an initial test. Any leads going in and out should be bypassed with feed through caps, ferrite beads, and/or small inductors. If visibility is an issue, solder together a cage made from brass wire mesh. There are also commerical versions:

--
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

The transmit antenna is only half the broadcasting solution; there is also a ground connection, the 'counterpoise'. Every time an amp of current goes up the antenna mast, an amp comes up from the ground connection.

Of course, it DOES get into the power wiring, everywhere there is an earth ground connection in the vicinity of the transmitter. If you really need to decouple from it, you use a shielded room with a single-point ground, and isolation transformers and filtering.