Topic: Interference and Noise

[Cutty Sark Sailor]

chk solder connections on B smds...

[W3DRM]

Looks like my 2B amp has gone south. I get a constant waveform on the channel regardless of the settings. I have dropped all inputs (Amp 1 and Amp 2) to 1.1 but the the waveform on channel 2B stays the same. Resetting everything has no effect either.

Sadly, I'm too busy with family here for the next week or so and won't have any time to work on it. I suspect something has happened with the E-field amp. It does sit in an outside enclosure and we have had temps over 100°F for the past couple of days. I need to place a temp sensor inside the enclosure to see just how hot (or cold) it does get in there.

You can see the crazy waveform on Cutty's signal info page (select the Minden station):

• http://frankfortweather.us/BoStaSig/

FIXED! I was wrong about the E-amp being outside. It is inside next to my RED controller board and connected to the E-preamp via a coax. We have family here and they are using the office as their bedroom. Seems as though someone plugged-in their iPhone charger into my power distribution box and then laid the cable across the top of the E-field amp. It picked-up the radiated signals from the wal-wart. I simply moved the power cable away from the E-field amp and the problem was solved!

[JonathanW]

Looks like my 2B amp has gone south. I get a constant waveform on the channel regardless of the settings. I have dropped all inputs (Amp 1 and Amp 2) to 1.1 but the the waveform on channel 2B stays the same. Resetting everything has no effect either.

Sadly, I'm too busy with family here for the next week or so and won't have any time to work on it. I suspect something has happened with the E-field amp. It does sit in an outside enclosure and we have had temps over 100°F for the past couple of days. I need to place a temp sensor inside the enclosure to see just how hot (or cold) it does get in there.

You can see the crazy waveform on Cutty's signal info page (select the Minden station):

• http://frankfortweather.us/BoStaSig/

FIXED! I was wrong about the E-amp being outside. It is inside next to my RED controller board and connected to the E-preamp via a coax. We have family here and they are using the office as their bedroom. Seems as though someone plugged-in their iPhone charger into my power distribution box and then laid the cable across the top of the E-field amp. It picked-up the radiated signals from the wal-wart. I simply moved the power cable away from the E-field amp and the problem was solved!

Should probably mention that as something "not to do" in the project documentation 

"No family with wall wart power supplies near the system"

[dfroula]

Just a note, I have found that faulty bridge diodes on non-switching mode, unregulated wall-warts can kick up a lot of line noise. An AM radio works great to trace those down, if you should have the misfortune of having one plugged in in your home.

Glad it was an easy fix!

Don
WD9DMP

Looks like my 2B amp has gone south. I get a constant waveform on the channel regardless of the settings. I have dropped all inputs (Amp 1 and Amp 2) to 1.1 but the the waveform on channel 2B stays the same. Resetting everything has no effect either.

Sadly, I'm too busy with family here for the next week or so and won't have any time to work on it. I suspect something has happened with the E-field amp. It does sit in an outside enclosure and we have had temps over 100°F for the past couple of days. I need to place a temp sensor inside the enclosure to see just how hot (or cold) it does get in there.

You can see the crazy waveform on Cutty's signal info page (select the Minden station):

• http://frankfortweather.us/BoStaSig/

FIXED! I was wrong about the E-amp being outside. It is inside next to my RED controller board and connected to the E-preamp via a coax. We have family here and they are using the office as their bedroom. Seems as though someone plugged-in their iPhone charger into my power distribution box and then laid the cable across the top of the E-field amp. It picked-up the radiated signals from the wal-wart. I simply moved the power cable away from the E-field amp and the problem was solved!

[weatherc]

Hi guys,

A partly repost of my post on BO-forum...

I use the station off-grid with solarpower. Theese solarstuffs generates "some" noises whats not figured out yet if they can be "cleaned out". I guess not.

- Doing settings the "normal way" to get it "quiet" ended up with like 4x2 gains H-field what resulted in ca 20 strikes/hour (what was less than 1%).
- Automode do not even get it out from inteference
- Auto noise just jumps up and down like a jo-jo in continous inteference
- The trick: Find the gains where Alptitude-filter gets those solar-noises filtered. In my case are it 10x5 gains on H-field and 8x4 for E-field. This gives a continous ca "raw" 200 signals/s. But as sent data went it in a 15-22% signals-to-strikes-ratio and as best today ca 850 strikes/hour and ca 70% of the 50-500 km range 
Even one step up or down in gains screws the filtering and plentty of signals are sent.

It do not detect that many stirkes really far away this way, farthest has been 2400 km away tought but main part of the detected strikes has been < 1500 km away but that was expected.
Another was how it acts now at the evening when acticity drops. And yes, the amount of sent signals has dropped too, at the moment is it ca 500 signals/s what ends in 5% signal-stirke-ratio.

// Henkka

[JonathanW]

So, new, low-level interference (H-field) at my location, started two days ago in the evening.  It's a roughly 52 kHz signal.





This is the first strike where the signal appeared (5:02:50 pm EDT), and it's magnetic field only.  Since then, it seems to come on during the day, and off at night, and it's stronger in the A-channel (Red) ferrite than in the B-channel (Green).

Based on the orientation of the ferrites in my system and eyeballing the relative strength and phase of the signals in the H-field channels, it appears the signal is coming fairly consistently from a bearing along which a street light can be located (in the opposite direction is one of my neighbors, a little farther away).  It's a rough fix, but relatively consistent.





So, my question is, could this be from a malfunctioning street light?  The night/day consistency makes me think that's the issue, but I'm curious if anyone has run across any similar signals.

[JonathanW]

Ok,  turns out the answer was a lot simpler...  We have a couple of Toshiba LCD TVs at about the same bearing from the H-field ferrites (opposite direction from the street lights) .   The manual says the input HDMI connection uses about 25 kHz for 1080p input.   I'm guessing the  signal is mixing with an already present  25 kHz signal coming from somewhere else (maybe some cfl bulbs).   The 50 kHz signal only happens when both the TV and the converter are on.

By the way--a big thank you to Mike for setting up his auto-refresh signal tracking page.   It's a big help in tracking down problems!

[meteo-melin]

Hello,

Can someone give me tips about the noise of station 1104 region 1.
It seems that there is some noise in the 1kHz and 65kHz frequencies.

Thanks a lot,
Regards.

[JonathanW]

Good information from Tobi in a post about interference mode today.

Without, I think, resolving all questions regarding the interference mode thresholds (and I know the developers are still working on things), Tobi indicates that setting gains such that strike impulses are greater than 1.5Vpp may lead to clipping and signal distortion.  That can become an issue, of course, when a storm rolls over a station and gains aren't reduced, but can also be an issue in general.  Better to lower the gains and the thresholds (within reason).

Since the signal display on the controller page is relative (is this correct?), using the signal plots on Mike's auto-updating signals page will tell you whether your absolute signal levels are getting too high.

[Knickohr]

Isn't it implemented ?

Look at the drop down of "%good" at about -3h. It happens, when the storms are close to my station (<100km).

Station still not in interference mode 

Thomas

[CF20852]

I'm still working on 60 Hz (and harmonics) interference here at Station 1093, North Bethesda, Maryland.  I decided it would be useful to rig up a way to be able to move my H-field antenna and Amplifier 12 around the house and listen to the Channel A and B outputs using earphones to see if I could find a quiet spot in the house.  (Sadly, all I've found so far is places that are more noisy than where I have the station set up.)  I bought an In-Line 1/8" Phone Jack, P/N 274-0274, from Radio Shack, and made up a two-foot-long single-ended Ethernet cable according to the T568A wiring convention (http://en.wikipedia.org/wiki/TIA/EIA-568).  On the Amplifier 12 control connector, pin 4 (T568A blue wire) is the AMP-B output, pin 5 (T568A blue/white wire) is ground, and pin 7 (T568A brown/white wire) is the AMP-A output.  I connected pin 7 (brown/white wire) to TIP on the phone jack, so AMP-A drives the left earphone, pin 4 (solid blue wire) to RING on the phone jack, so AMP-B drives the right earphone, and pin 5 (blue/white wire) to SLEEVE on the phone jack.  See http://www.connecthowto.com/20091227/connect-how-to-stereo-plug/ .  I'm still operating tethered to a 9V wall-wart powering a 5V linear regulator that in turn powers Amplifer 12 through the mini USB connector, but the next step is to power the 5V regulator from a battery.  The pot on Amplifier 12 works fine as a volume control, and with my earphones, I can get plenty of volume, probably enough to make a teenager cringe.

The attached photo shows the connections to the phone jack.  I have a stereo cable plugged into the jack in the photo.

As Don points out in his message below, it may be more convenient to attach the phone jack to the test connectors, depending on what you have in your junk drawers.  I happened to have a half-dozen Ethernet plugs, some Ethernet cable, and a crimping tool handy.

Of course, the bandwidth of this system is limited--given my age and exposure to rock concerts, airplane engines, small arms, and lawnmowers, maybe 10 kHz if I turn the volume up high enough.

Update 8/4/2014:  I implemented the battery power option and went on walkabout through my house yesterday afternoon, and found a reasonably quiet spot in the northwest corner of the house on the lowest floor.  It seems to be immune from the bathroom dimmer switch on the top floor, and the LED lights in the kitchen don't seem to interfere either.  There's still something, though...  But right now, at 5:20AM, my noise level is down to less than 60 mV pk-pk (as measured by the controller) and my gain settings are 16 x 4 x 40 = 2560.

[dfroula]

The test pins near the RJ45 connector on the amplifier board can also be used to wire in a 1/8" stereo jack, perhaps more conveniently. I found it best to connect a small headphone amplifier with a high-impedance input to prevent loading of the amp output and to provide impedance matching to the headphones.

I also found that connecting long leads to the output pins on the amp (for monitoring on an oscilloscope, for example) can produce oscillations in the op-amps when a lightning pulse is detected.

Don
WD9DMP

[bidouilleur]

hello All, joined the system about 3 weeks ago in France (region1), have been plagued by 2 peaks and I can't find the origin at all.  Think I resoldered the 2 boards about 5 times, fooled around in the house with am radio, cutting electricity etc etc etc ... these 2 peaks are there and no way to get rid off them.
Also when I turn the antenna's the signal goes from one channel to the other, makes me think it is not HW related
I do have an electric fence about 300 yards away but we already tried turning it off, same result ...
Peaks are I'd say around 1 and 64 kHz ....

you can see it at http://wouters.fr/blitz/plaf.php?station=1104 in action, driving me nuts.


Living in little village with a few houses around, not even very close ...

would appreciate your thoughts and eventually an idea where to start digging for a solution ...
thanks

[Cutty Sark Sailor]

hello All, joined the system about 3 weeks ago in France (region1), have been plagued by 2 peaks and I can't find the origin at all.  Think I resoldered the 2 boards about 5 times, fooled around in the house with am radio, cutting electricity etc etc etc ... these 2 peaks are there and no way to get rid off them.
Also when I turn the antenna's the signal goes from one channel to the other, makes me think it is not HW related
I do have an electric fence about 300 yards away but we already tried turning it off, same result ...
Peaks are I'd say around 1 and 64 kHz ....

you can see it at http://wouters.fr/blitz/plaf.php?station=1104 in action, driving me nuts.


Living in little village with a few houses around, not even very close ...

would appreciate your thoughts and eventually an idea where to start digging for a solution ...
thanks

I might suspect someone having a 64kHz security device nearby... you might try Google 64kHz to look for a possible device.  64kHz is used for a lot of things like that, security scanners, etc... sound systems, etc...

[CF20852]

This photo shows the power line communications interference generated by an Enphase Energy Envoy communications gateway communicating with a set of Enphase microinverters.  The power line communications signals are to the left of center.

[bidouilleur]

will have a look for alarms around, we don't have any for sure

and the line interference, I see these generally around 6 am and 10 pm when we get the signals to go to night or day prices ... but that is only for a minute or two

[CF20852]

This:


Seems to be caused by this, between my back-fence neighbor's garage doors:


Here's a spectrum plot of the above waveform:


Or it could be that fluorescent light in combination with the other two my neighbor has on the front of his house... 

The interference appears when my neighbor's outside lights come on at about 6PM and disappears when they go off at about 11PM.

My wife wanted to know why I had to go over there and take a picture, so I explained to her that I was documenting an interference problem and I might offer to swap out one of the neighbor's lights.  She rolled her eyes and muttered "oh, please..."

The things we do for science.   

The red channel line at 24 kHz is probably NAA in Cutler, ME, which you don't see in the waveform plot because although the green waveform is the same signal as represented in the spectrum plot, the waveform image and the spectrum image were captured at different times.

[JonathanW]

Yep.  I've noticed a consistent ~24 kHz interference signal from GE-brand Reveal CFLs.  I'd imagine many (most?) integrated CFLs have electronic ballasts that operate at a similar frequency.

I use CFLs far from my BO probe, and incandescents near.  I haven't tried LED bulbs yet.

[miraculon]

Yep.  I've noticed a consistent ~24 kHz interference signal from GE-brand Reveal CFLs.  I'd imagine many (most?) integrated CFLs have electronic ballasts that operate at a similar frequency.

I use CFLs far from my BO probe, and incandescents near.  I haven't tried LED bulbs yet.

I have a LED bulb in our bedroom. When it is on it creates a lot of hash noise on the NOAA weather radio nearby. I realized that this is 162.55 MHz and not our frequency range of interest.

These noise problems seem to be getting worse....

Greg H.

 

[JonathanW]

As Chip mentioned:

http://en.wikipedia.org/wiki/VLF_Transmitter_Cutler

I'd bet this 2 MW beauty is the source of the 24 kHz signal I've consistently seen since I started operating my station.

[b.e.wilson]

If you have some very loud local AM broadcast stations, both of which are within about 100 KHz of each other, they can mix in the first gain stage to give a heterodyne signal when the first gain stage is amplifying. This signal will have a carrier frequency matching the difference in the two AM station frequencies. In the example below, I have two stations, 1450 and 1400 KHz, each at S9+40dB, mixing in the first gain stage (when the gain is set at 8 or higher) to give a 50 KHz heterodyne. Modulating this heterodyne are the audio modulations of both AM stations, one at a low frequency (the broad up-down movement, I think) and one at a high frequency (the vertical expansion and contraction of the envelope).

In the frequency plot you can see the 50 KHz heterodyne, with audio sub-bands on either side, as well as the pure audio frequencies at low frequencies below 10 KHz. I don't have an explanation for the 30 KHz or 80 KHz signal, but they may be related (0 KHz + 30 KHz = 30 KHz; 50KHz + 30 KHz = 80 KHz).

Diagnosis: go to manual and set the gain of the first amp stage very low, in my case 1 or 2. This makes the first gain stage far more linear and the mixing of the signals is eliminated. The second gain stage is immune to heterodynes as it is amplifying the bandpass-filtered signal.

EDIT: Solution. In my case, the loop antennas had a self-resonant frequency close to the 1400 KHz interfering frequencies. The loop antennas, then terminated with a high resistance value (R1 and R16, 2.2kOhm), had a high Q value, so they became very efficient at pulling in that signal. Richo (project dev) suggested I lower the value of R1 and R16 to 100 Ohms. I used resistors I had on hand, 150 Ohm, and the intermod disappeared at all first-amp gain settings. By using a lower-valued termination resistor, the value of Q for my loop antennas was dramatically reduced, flattening the frequency response curve, and thus the efficiency of the loops at the resonant frequency. This should be considered the proper solution to an intermod problem. Lowering the first-stage gain is a diagnostic method and temporary solution only.

[dwmatt]

Finished my build and hooked everything up.  Getting really bad interference and can't tell if it is internal to the unit or external.  I've been running around with an AM radio to see if I can find anywhere in my house that isn't noisy (having a hard time with that.)  I think part of the problem is the crappy USB wall-wart I am using (which will be replaced asap,) and the other part is distance between the controller and the amplifier (need to get a 75ft shielded cable asap as well.)

I don't see any sine waves on the rogue gallery, but my 1A channel periodically sees a long sine wave, my 1B channel sees a 60hz wave, and I've grounded the controller (but not the amplifier.)  I know that some of the noise I am seeing right now is the electric company's smart meter (which I shouldn't see when I move the amplifier to its ultimate location about 50ft away from the smart meter.)

What is really weird is I'll see a good couple minutes of clear, with one or two strikes per minute, then a few minutes of 60/s per minute, then clear, then 60/s per minute.  Interference mode seems to activate every few minutes.  Won't start submitting my data until I get a better handle on the noise.

[miraculon]

Finished my build and hooked everything up.  Getting really bad interference and can't tell if it is internal to the unit or external.  I've been running around with an AM radio to see if I can find anywhere in my house that isn't noisy (having a hard time with that.)  I think part of the problem is the crappy USB wall-wart I am using (which will be replaced asap,) and the other part is distance between the controller and the amplifier (need to get a 75ft shielded cable asap as well.)

I don't see any sine waves on the rogue gallery, but my 1A channel periodically sees a long sine wave, my 1B channel sees a 60hz wave, and I've grounded the controller (but not the amplifier.)  I know that some of the noise I am seeing right now is the electric company's smart meter (which I shouldn't see when I move the amplifier to its ultimate location about 50ft away from the smart meter.)

What is really weird is I'll see a good couple minutes of clear, with one or two strikes per minute, then a few minutes of 60/s per minute, then clear, then 60/s per minute.  Interference mode seems to activate every few minutes.  Won't start submitting my data until I get a better handle on the noise.

I have two suggestions that might help. (or may not, but they are worth checking)

1: Log-in to the controller and select the "Signals" tab. Below the "oscilloscope" charts, there is a button labelled "5V Sampling" in the Signal Tools area. Enable it and the display will switch over to a power supply monitor. If your Peak-to-Peak noise is more than a millivolt or two, you need to look for a cleaner supply. (there have been some suggestions here in the forum.) Remember to disable the 5V monitor if you want to look at antenna signals again.

2: Disable the PWM for the display backlight. There are several ways to do this, either time-outs or just setting it to 100% duty cycle. This is under "Settings" tab, "Buzzer/LCD" sub-tab. Under the "LC Display" option, the first item is "Brightness". The PWM for the display is a known noisemaker and it is best to go to 100% or 0%. You can also set it to turn off after a time period. I was using this for a while, but I leave mine at 100%.

From your description, it sounds like you have the amplifier connected via UTP vs. the needed shielded STP Cat5. When you get your cable test continuity on the shield with a meter if you have one (or a Cat5 tester). I got a defective STP cable that caused me problems before I figured out that it was bad. The grounding of the controller probably won't do much if you are using UTP Cat5.

Also, is it possible you have in intermittent connection on the antenna to the connector block? Also check your solder connections on the SOIC8 PGA amplifier ICs. A microscopic void in the solder joints can cause problems.

Good luck hunting it down.

Greg H.






Greg H.

[dwmatt]

I replaced the cable with a shielded cable, but didn't see any improvement (though I now moved it out of the computer room and to a far quieter area.)  Removed the solder with a wick from SIOC8, resoldered SIOC8 and then SIOC1, and now I am getting 16x10x40 gain on both channels, with no high noise, but I am not sure it is working as I can see what appears to be small magnetic interference from two speakers 30' away (not enough to trigger, but I see the freq changing as the speakers get louder/quieter, but no 60hz and the 1st channel isn't doing a long sine any more.)

However, I am also strikes on LightningMaps, and my controller isn't seeing any of them, even though I should see them (others in California, far north of me are.)  Maybe I busted it instead of fixing it.  Those SIOCs are tricky.  Wish there was an easy way to unsolder them and resolder them without a lot of heat to potentially damage them or the components around them.

*Another Update: Fixed.  I've been recording strokes, and even though my gain isn't to my liking yet and there is still a lot left to do, the amplifier appears to be working properly.  Now if I could just find the source of the noise at dusk and dawn...someone has a sodium lamp around here, probably the school, and it is pushing 60s/s every evening as the lamp warms up.  Oh well.

[uhf]

I have no idea what I'm doing. It appears I have some sort of interference, but most of the time the web gui causes my browser to crash. There has to be a better way to troubleshoot. Anyone have any idea what this chart means??