VLF pulses

[WMXZ]

I have now a 2nd rshake setup

observation

2nd has no idle tone

but 2nd has now VLF pulses

In the 2min puls thread Branden mentioned bad connected geophone, but did not explain why, i.e. what the mechanism is behind this phenomenon.

Nevertheless I post a picture of both stations

one 20 min scale and one a zoom in.

one can cleasly see the pulse is a single spike.

BTW, these figures are made via matlab/GISMO.

[chris...@aol.com]

Hi Zimmer, 

    I suggest that you look for a pulse when a fridge or a deep freezer switches on ? 

I presume that the two devices are physically separated and are not on the same power point ? 

    Regards,

    Chris
From: WMXZ <zim...@nurc.nato.int>
To: RaspberryShake <raspber...@googlegroups.com>
Sent: Thu, Feb 16, 2017 3:56 pm
Subject: [RShake Forum] VLF pulses

--
Some useful links:
 
Manual: http://manual.raspberryshake.org/
Do It YourSelf Page: http://raspberryshake.org/do-it-yourself
Shop: https://shop.raspberryshake.org/
Website: http://raspberryshake.org/
 
Instagram: https://www.instagram.com/raspberryshake/
Hashtag: #rasperryshake
---
You received this message because you are subscribed to the Google Groups "RaspberryShake" group.
To unsubscribe from this group and stop receiving emails from it, send an email to raspberryshak...@googlegroups.com.
To post to this group, send email to raspber...@googlegroups.com.
Visit this group at https://groups.google.com/group/raspberryshake.
To view this discussion on the web visit https://groups.google.com/d/msgid/raspberryshake/166acfa3-7e9f-4de7-87e3-8ff4aa819b1f%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[WMXZ]

Chris,

they are in different rooms, (note: Stationview is Not correct (yet))

The 2nd is now at the same location where the first was before it was relocated to the actual room.

Maybe the power supply is of less quality and does not filter electrical spikes?

Walter

[chris...@aol.com]

Hi Walter, 

    You get quite a large pulse on the house mains wiring and a physical thump on floor as a low 

resistance is switched across the house electricity supply and the electric motor starts up. Put your 

hand on an electric fridge and switch it on ? This is a common 'candidate'. 

    Regards. 

    Chris

To view this discussion on the web visit https://groups.google.com/d/msgid/raspberryshake/6c8e642b-23fb-4f09-a6ef-eea265d4a81c%40googlegroups.com.

[WMXZ]

Chris,

I changed power supply, and noted that the spike update rate was initially (after power-up) higher and slowed down a bit. This was similar to first power up.

So I would guess it is not comming from the house but from inside the Rshake-RPI system. (fridge also does not startup every 2 minutes).

Interestingly, the sub-hz noise of the 2nd device (RA78B) is higher than the one of the first rshake (RDF14). This increased sensitivity to noise may influence the generation of VLF spikes. 

Walter

[chris...@aol.com]

Hi Walter, 

    OK, it sounds like OSOP have used a FET Opamp. From memory, they have some impulse 

noise problems.   

    Regards, 

    Chris

To view this discussion on the web visit https://groups.google.com/d/msgid/raspberryshake/b7ac2cdc-6023-442a-9b3b-64b877f98236%40googlegroups.com.

[Branden Christensen]

Hi Walter!

Good evening. 

I would say that it is not that your sensor does not have an idle tone, it is that the idle tone is at a frequency below 0.5 Hz making it difficult to see in a spectrogram plot.

Do you still have spikes after changing the walwart I would be happy to look at the data. Please tell me what the station name is. A comment: we did notice that normal microUSB cell phone changers caused spikes in the data because they are usually 2 amps and the RPi 3 Model B requires more. With 2.5 amp chargers which we ship RShake with, this problem disappeared.

Thanks for your feedback. It is important.

Yours, Branden

    Chris

To unsubscribe from this group and stop receiving emails from it, send an email to raspberryshake+unsubscribe@googlegroups.com.
To post to this group, send email to raspberryshake@googlegroups.com.

Visit this group at https://groups.google.com/group/raspberryshake.
To view this discussion on the web visit https://groups.google.com/d/msgid/raspberryshake/b7ac2cdc-6023-442a-9b3b-64b877f98236%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

--

Some useful links:
 
Manual: http://manual.raspberryshake.org/
Do It YourSelf Page: http://raspberryshake.org/do-it-yourself
Shop: https://shop.raspberryshake.org/
Website: http://raspberryshake.org/
 
Instagram: https://www.instagram.com/raspberryshake/
Hashtag: #rasperryshake
---
You received this message because you are subscribed to the Google Groups "RaspberryShake" group.

To unsubscribe from this group and stop receiving emails from it, send an email to raspberryshake+unsubscribe@googlegroups.com.
To post to this group, send email to raspberryshake@googlegroups.com.

Visit this group at https://groups.google.com/group/raspberryshake.

To view this discussion on the web visit https://groups.google.com/d/msgid/raspberryshake/15a48d43a10-19df-10e17%40webprd-a44.mail.aol.com.

[WMXZ]

Branden,

after changing the walwart, the pulses are still there (over night they settled to 1 pulse every 4 minutes)

the station is RA78B,

which is collocated (adjacent room) to RDF14 (Note: Stationview gives wrong location).

It is based on Magnitude 6.5 +, my own RPI3-B1 running V3 Software and own housing (no 3-point contact to ground, but normal M2.5 screw used to fix the RPI)

At the moment power supply is 2A, 'original' was 2.5A.

Walter

    Chris

To view this discussion on the web visit https://groups.google.com/d/msgid/raspberryshake/b7ac2cdc-6023-442a-9b3b-64b877f98236%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

--
Some useful links:
 
Manual: http://manual.raspberryshake.org/
Do It YourSelf Page: http://raspberryshake.org/do-it-yourself
Shop: https://shop.raspberryshake.org/
Website: http://raspberryshake.org/
 
Instagram: https://www.instagram.com/raspberryshake/
Hashtag: #rasperryshake
---
You received this message because you are subscribed to the Google Groups "RaspberryShake" group.

To unsubscribe from this group and stop receiving emails from it, send an email to raspberryshak...@googlegroups.com.
To post to this group, send email to raspber...@googlegroups.com.
Visit this group at https://groups.google.com/group/raspberryshake.

[WMXZ]

Here is a wave form of a single pulse

The polarity may be reversed, but to me it looks like exponential loading of a capacitor follows by a nearly critically damped oscillations

Initial slope 1800 counts/s

Walter

On Friday, February 17, 2017 at 7:54:38 AM UTC+1, WMXZ wrote:

[chris...@aol.com]

Hi Branden, 

     Do you use any FET Opamps in the RShake, please ? These were known generators of small 

pulses - 'popcorn noise' - I never heard an explanation of exactly why this occurs - but it does ! 

    See http://www.ti.com/lit/an/slva043b/slva043b.pdf  and many others.

    Regards,  

    Chris

From: WMXZ <zim...@nurc.nato.int>
To: RaspberryShake <raspber...@googlegroups.com>
Cc: branden.christensen <branden.c...@osop.com.pa>
Sent: Fri, Feb 17, 2017 7:41 am
Subject: Re: [RShake Forum] VLF pulses

--
Some useful links:
 
Manual: http://manual.raspberryshake.org/
Do It YourSelf Page: http://raspberryshake.org/do-it-yourself
Shop: https://shop.raspberryshake.org/
Website: http://raspberryshake.org/
 
Instagram: https://www.instagram.com/raspberryshake/
Hashtag: #rasperryshake
---
You received this message because you are subscribed to the Google Groups "RaspberryShake" group.
To unsubscribe from this group and stop receiving emails from it, send an email to raspberryshak...@googlegroups.com.
To post to this group, send email to raspber...@googlegroups.com.
Visit this group at https://groups.google.com/group/raspberryshake.

To view this discussion on the web visit https://groups.google.com/d/msgid/raspberryshake/246031bb-5d22-4e59-8e45-e612d0e88e58%40googlegroups.com.

[Branden Christensen]

Walter:

Good day!

This looks like one of two possibilities I can think of:

• Wifi- Is wifi enabled on either of the units?
• Power- 2A is not enough. The power connector we ship with is: http://www.suntechpc.com/store/index.php?id_product=18&controller=product

Branden

[Brett N]

Chris,

We wouldn't do that. 

Opamp's noise spec is 5.6nV/rtHz, flat from very low frequencies to 100kHz.

Brett

[WMXZ]

Branden,

I switched back to 2.5A, pulses still there.

But to correct you, it worked overnight with a Samsung phone charger (my mistake) that is rated to be 4.75V and 0.55A (sic, not 2A as said before)

So max power requirement may be 2.5A, but RPI seems to work also with 0.5A.

NO WIFI on, as far as I know, took your img file as is and use wired Ethernet.

Router however talks WLAN to rest of house.

BUT WIFI or any outside RPI events would not explain why on power-up pulse rate  is increased and decreases slowly.

BTW, I flipped the geophone (was wrong wiring) but pulse still goes towards negative numbers (towards 0V, I guess)

Walter

[Branden Christensen]

Walter:

Do you have time to run a few more experiments? To start, please try this experiment: use the power supply from RDF14 (no spikes) on RA78B (spikes). Next experiment: swap their physical locations and see if the spikes stay with the location or follow the RShake.

Also, could you send some photos of the two units? I want to see if you have anything unconventional in your setup. 

Concerning the geophone: How do you know to say that the wiring is wrong? If you are using a RACOTECH geophone from OSOP and you connect the grey cable from the “+” on the geophone to the “+” on the Raspberry Shake board, the output signal will have the correct polarity.

Yours, 

Branden Christensen

Director, OSOP

USA: +1-206-414-1288; Panamá: +507-6747-1427

Get yours now! Visit shop.raspberryshake.org

--

Some useful links:
 
Manual: http://manual.raspberryshake.org/
Do It YourSelf Page: http://raspberryshake.org/do-it-yourself
Shop: https://shop.raspberryshake.org/
Website: http://raspberryshake.org/
 
Instagram: https://www.instagram.com/raspberryshake/
Hashtag: #rasperryshake
---
You received this message because you are subscribed to the Google Groups "RaspberryShake" group.

To unsubscribe from this group and stop receiving emails from it, send an email to raspberryshake+unsubscribe@googlegroups.com.
To post to this group, send email to raspberryshake@googlegroups.com.

Visit this group at https://groups.google.com/group/raspberryshake.

To view this discussion on the web visit https://groups.google.com/d/msgid/raspberryshake/2d792af0-8731-4377-9da0-6e615ea7c6ef%40googlegroups.com.

[WMXZ]

I can do the switches.

I connected the grey cable NOT to '+', but corrected this this morning.

Walter 

To unsubscribe from this group and stop receiving emails from it, send an email to raspberryshak...@googlegroups.com.
To post to this group, send email to raspber...@googlegroups.com.

Visit this group at https://groups.google.com/group/raspberryshake.

[Brett N]

Walter,

Your comment about no 3-point contact got my attention, if I understood what you were saying.  On broadbands, two hard surfaces in contact will generate noise pulses, probably from differential thermal expansion.  With three points solidly fixed in place, such expansion (very) slightly bends the legs and the base without generating noise.  Just something to consider.

Brett 

[WMXZ]

Brett,

I thought about this, also the way contact is made,  but more in terms SNR, However, pulse would then come in via the geophone.

But reversing the geophone connection did not change the polarity of the pulse, so IMO is should come from a point after the coil.

Walter

[WMXZ]

Branden,

switching the power supply did not change the situation.

Again, on power-up pulse rate is 1/minute slowing down while system heats up.

(Just realised that the ADC board is just over RPI processor)

Walter

On Friday, February 17, 2017 at 1:22:51 PM UTC+1, Branden Christensen wrote:

[Branden Christensen]

Hi Walter:

Thank you for that. We will get to the bottom of this. 

The following experiments I would like to ask you to do are:

1. Swap the locations of the two Shakes

2. Swap the RShakes (but leave geophone, RPi and everything else the same)

Thank you!

Branden

[WMXZ]

Branden,

I'm on it

have problems with swarm and RS community server accessing the Rshakes

on the console, there are a lot of local SWARM: connect() giving up

Q: is there a minimum time for the Rshake to establish full functionality?

Anyhow (e-mail needs too long to write),

got now data via RS-community but not direct via local SWARM

pulses are still on RA78B which now switched location with RDF14 and using OSOP power supply.

I still 4 OSOP geophones, should I try these.

I hesitate to open the Magnitude 9 one, except there is the possibility to get rid of the idle tone (AKA input mismatch)

Walter

[WMXZ]

Weird situation:

Can access my Rshake via RS-community server but not directly from SWARM.  error: connect(): give up

address is correct

obviously :5000 gives me access to Rshake.

[Branden Christensen]

Hi Walter:

I'm on it

have problems with swarm and RS community server accessing the Rshakes

 

OK, I am going to keep focusing on the pulses for now. Let's come back to this later if possible.

on the console, there are a lot of local SWARM: connect() giving up

Q: is there a minimum time for the Rshake to establish full functionality?

Anyhow (e-mail needs too long to write),

got now data via RS-community but not direct via local SWARM

pulses are still on RA78B which now switched location with RDF14 and using OSOP power supply.

OK, so we know that location is not the issue. Neither is wifi. Neither is power supply. The only remaining things to test are: geophone and RShake card itself. 

Please run these tests:

1. Change geophone from non-spiky to spiky RShake.

2. After returning everything back to original state, please change RShake board from non-spiky to spiky RShake. 

Branden 

[WMXZ]

Branden,

Did swap Rshake board and geophone (the one of the Mag9 was soldered, and I have no iron here).

I did not fasten the geophones, but put them to the side on the ground, therefore results may be more noisier.

they are all in the same room on the same type of floor separated by 2 m.

Observations:

both are little bit noisier (free standing geophones)

orig (RDF14) preamp less sensitive than 2nd (RA78B)

Idle tone in RDF14 gone in RA78B

VLF pulse are still bat barely visible in RA78B

-pulses must likely due to interferences of RPI and visible due to increased sub-Hz sensitivity of pre-amp

-idle pulse due to interferences of RPI and type of strapping down

Left before swap right after swap

[Branden Christensen]

Walter:

Thanks again for your help!

Your message has left me confused.

Can you do the experiments by maintaining everything but a single variable constant? Changing multiple parameters at once will not help us identify the problem.

After returning everything to normal, 

Please first:

Change only the geopohone and then

Change only the RShake board.

Also, the idle tone is a funtion of the ADC and will not go away.

Thanks Walter.

Branden

--
Some useful links:
 
Manual: http://manual.raspberryshake.org/
Do It YourSelf Page: http://raspberryshake.org/do-it-yourself
Shop: https://shop.raspberryshake.org/
Website: http://raspberryshake.org/
 
Instagram: https://www.instagram.com/raspberryshake/
Hashtag: #rasperryshake
---
You received this message because you are subscribed to the Google Groups "RaspberryShake" group.

To unsubscribe from this group and stop receiving emails from it, send an email to raspberryshake+unsubscribe@googlegroups.com.
To post to this group, send email to raspberryshake@googlegroups.com.

Visit this group at https://groups.google.com/group/raspberryshake.

To view this discussion on the web visit https://groups.google.com/d/msgid/raspberryshake/4c60c9eb-dfb7-47e9-a24b-67db24d401ca%40googlegroups.com.

[WMXZ]

Brandon,

I cannot remove the geophone from the original Mag9 board, as it has been soldered on the rshake analogue  board (the one with the 2.2 Hz tone), and I have no iron here.

Well, concerning the 2.25 Hz oscillation, it is gone now after swapping the boards.

So suggest conclusion: 2.25 Hz tone that was there for a month  may not necessarily be a simple ADC generated idle tone but may depend on additional factors.

Walter 

To unsubscribe from this group and stop receiving emails from it, send an email to raspberryshak...@googlegroups.com.
To post to this group, send email to raspber...@googlegroups.com.

Visit this group at https://groups.google.com/group/raspberryshake.

[Branden Christensen]

Walter:

Understood. 

What is the status of the VLP peak? Please provide as many details as possible as I am a bit confused as the current status. Feel free to continue troubleshooting this with me off the list.  

Yours, 

Branden

[WMXZ]

Branden,

As I cannot, for the moment, unsolder one geophone from the analogue board, I reverted all to the original settings (with one modification)

OK, pulses on RA78B are still there, but we know this is from RPi3.

However, 2.25 Hz tone is now gone on RDF14.

Here comes the only modification I have done:

The RDF14 is a Mag9 Reward and was used as assembled and shipped by OSOP in December.

No work was done by me.

When dismounting the geophone I realized that the blue cable was below the bracket, pressing onto the geophone.

Now, remounting the geophone Rshake combo, all cables are free between geophone and analogue board.

As the 2.25Hz tone is now gone, I conclude that the vicinity of the wire to the coil may have some influence which in the end resulted in the 2.25 Hz tone

 

Walter

[WMXZ]

our post crossed.

Does the board work also with an original RPi- B board, I still have one that I can use next week?

[Branden Christensen]

Walter:

I am sorry but I can no longer follow this trend. It would be much easier if we could Skype or whatsapp. Do you have Skype or whatsapp? Please respond to me off the list.

Yours, 

Branden Christensen

Director, OSOP

USA: +1-206-414-1288; Panamá: +507-6747-1427

Get yours now! Visit shop.raspberryshake.org

--

Some useful links:
 
Manual: http://manual.raspberryshake.org/
Do It YourSelf Page: http://raspberryshake.org/do-it-yourself
Shop: https://shop.raspberryshake.org/
Website: http://raspberryshake.org/
 
Instagram: https://www.instagram.com/raspberryshake/
Hashtag: #rasperryshake
---
You received this message because you are subscribed to the Google Groups "RaspberryShake" group.

To unsubscribe from this group and stop receiving emails from it, send an email to raspberryshake+unsubscribe@googlegroups.com.
To post to this group, send email to raspberryshake@googlegroups.com.

Visit this group at https://groups.google.com/group/raspberryshake.

To view this discussion on the web visit https://groups.google.com/d/msgid/raspberryshake/99846a11-0169-47e8-ac0e-ec6da5d9a7c1%40googlegroups.com.

[WMXZ]

Correction to my post.

idle tone re-appeared once covered with box. It seems that in free air, air movement generate sufficient noise in the 3 Hz band so that no idle tone occurs.

- uncovered rshake noise increases but idle tone disappeared

- covered with large digikey box, noise decreases idle tone reappears.

[Branden Christensen]

Nice work. I have seen this also. Upon closer inspection you would find that the idle noise did not disappear, it was only masked by noise.

Skype?

Branden Christensen

Director, OSOP

USA: +1-206-414-1288; Panamá: +507-6747-1427

Get yours now! Visit shop.raspberryshake.org

--

Some useful links:
 
Manual: http://manual.raspberryshake.org/
Do It YourSelf Page: http://raspberryshake.org/do-it-yourself
Shop: https://shop.raspberryshake.org/
Website: http://raspberryshake.org/
 
Instagram: https://www.instagram.com/raspberryshake/
Hashtag: #rasperryshake
---
You received this message because you are subscribed to the Google Groups "RaspberryShake" group.

To unsubscribe from this group and stop receiving emails from it, send an email to raspberryshake+unsubscribe@googlegroups.com.
To post to this group, send email to raspberryshake@googlegroups.com.

Visit this group at https://groups.google.com/group/raspberryshake.

To view this discussion on the web visit https://groups.google.com/d/msgid/raspberryshake/0cb47041-2f21-47ef-a7ce-9a8e0a04be18%40googlegroups.com.

[WMXZ]

Final update

pulses disappeared after blacklisting wifi and bt

according to http://www.runeaudio.com/forum/pi3-and-dac-pro-t3293-50.html#p13076 (different problem)

You can disabling the drivers by editing the file /etc/modprobe.d/disable_rpi3_wifi_bt.conf:

##wifi
blacklist brcmfmac
blacklist brcmutil
##bt
blacklist btbcm
blacklist hci_uart




i.e. one of them (wifi or bt) is interfering with ADC or RShake board.

Suggest OSOP to add the disable_rpi3_wifi_bt.conf file to img. To use wifi simply uncomment the two blacklist commands

before someone asks: I used OSOP img file V3 and connected to RPI only over cable and never activated wifi of bt.

One main difference of my setup with respect to standard OSOP configuration is that I use my own housing and the geophone is on other side of rpi (there where the power supply is), so cables are routed differently.

finally: Idle tone is still there now on RA78B (now with correct location in station view, i.e. RDF14 and RA78B are not longer collocated, location here for RA78B seem much noisier)

To unsubscribe from this group and stop receiving emails from it, send an email to raspberryshak...@googlegroups.com.
To post to this group, send email to raspber...@googlegroups.com.

Visit this group at https://groups.google.com/group/raspberryshake.

[Branden Christensen]

Excellent work Walter! And thank you for persevering. 

Yesterday I was successful in reproducing this problem on two units. I will now test your fix and report back. 

Yours, Branden

[Branden Christensen]

Walter:

Good afternoon.

We have confirmed that these VLF pulses, which we were able to reproduce at OSOP:

can be solved by this modification at /etc/modprobe.d/disable_rpi3_wifi_bt.conf which will be included in the next update:

##wifi

blacklist brcmfmac

blacklist brcmutil

(disabling blue tooth is not necessary)

Thank you Walter. Nice breakthrough.

I have also updated the documentation to discourage the use of Wifi, even more so than I had in the past. The trouble with Wifi is that it produces a lot of RF that makes itself into the Raspberry Shake. The Raspberry Shake has no RF rejection (which was by design- a cost-based decision that contributed to the possibility of creating a seismograph as good as Shake for under $400/ unit). 

Yours, Branden

[dave d]

Is the wifi issue caused by proximity or circuitry ?    If circuity, would disabling the internal wifi and adding a cabled usb wifi dongle resolve the issue?

On Thursday, February 16, 2017 at 8:56:25 AM UTC-7, WMXZ wrote:

I have now a 2nd rshake setup

observation

2nd has no idle tone

but 2nd has now VLF pulses

In the 2min puls thread Branden mentioned bad connected geophone, but did not explain why, i.e. what the mechanism is behind this phenomenon.

Nevertheless I post a picture of both stations

one 20 min scale and one a zoom in.

one can cleasly see the pulse is a single spike.

BTW, these figures are made via matlab/GISMO.

[WMXZ]

I disabled WIFI in the RPI (as explained on 20th Feb) and using a cable to a 1 m distant TP-Link WIFI extender, which receives WIFI from the router.

So, yes, the physical separation of the WIFI from the RPI solved my problem.

I have two RShakes, but only one showed the WIFI issue, so, not all RPI 3 Model B seem to be equal. Also I do not know where on the RPI the WIFI antenna is located and if the path of the Geophone cable matters.

Using a WIFI USB dongle 1 m apart may be an alternative solution worthwhile to be investigated.

Walter

[Branden Christensen]

Hi guys:

From what we can tell, the wifi issue is due to thw 3 model B built in antenna being so close the Rshake board.

Branden

--
Some useful links:
 
Manual: http://manual.raspberryshake.org/
Do It YourSelf Page: http://raspberryshake.org/do-it-yourself
Shop: https://shop.raspberryshake.org/
Website: http://raspberryshake.org/
 
Instagram: https://www.instagram.com/raspberryshake/
Hashtag: #rasperryshake
---
You received this message because you are subscribed to the Google Groups "RaspberryShake" group.

To unsubscribe from this group and stop receiving emails from it, send an email to raspberryshake+unsubscribe@googlegroups.com.
To post to this group, send email to raspberryshake@googlegroups.com.

Visit this group at https://groups.google.com/group/raspberryshake.

To view this discussion on the web visit https://groups.google.com/d/msgid/raspberryshake/24a0c14c-d951-478e-aab4-1aef9f777f34%40googlegroups.com.

[Heinz-Bernd Eggenstein]

Some makers have found out that the RPi3 *almost* has a connector for an external antenna...just some careful soldering required ;-)

https://hackaday.io/project/10091-raspberry-pi-3-external-antenna

Do this at your own risk...when done wrong this will destroy your Pi. It's much easier to instead add a USB WiFi dongle via an USB extension cable, but those who feel confident enough with their soldering skills might want to try anyways.


Cheers
HB

To unsubscribe from this group and stop receiving emails from it, send an email to raspberryshak...@googlegroups.com.
To post to this group, send email to raspber...@googlegroups.com.