Noise on Log Channels

Forum for discussing PicoLog
Post Reply
DJETH
Newbie
Posts: 0
Joined: Mon Jun 29, 2020 3:17 pm

Noise on Log Channels

Post by DJETH »

All - Checking if anyone has ran into this and has a solution...
I have been using Picolog and PicoScope for about 6 months and am using Picolog lately to record temperatures and "events" along with a variable frequency output. I am using a 5444D MSO (ID on Picolog is GX138/0098) and a 5442D (ID on Picolog is HT449/036). I have been trying to track down the source of some noise that comes and goes (see attachment "Jun 29, 2020 5-06-13 AM 1" and the CONFIG file)
Jun 29, 2020 5-06-13 AM 1.picolog
(5.19 MiB) Downloaded 50 times
Jun 29, 2020 5-06-13 AM 1 CONFIG.picolog
(6 KiB) Downloaded 46 times
but I have had no success. You can see in the temp readings (red and blue) that the noise comes and goes. It does show up on the other samples but these are easy to see. The bottom 3 channels have a 0 OHM BNC cap (homemade) installed and are sampling at different rates. The noise does not have the same pattern at the different sample rates. I don't see a way to turn on any filters in Picolog but I do have the 60hz (mains) set to 60hz. This capture was done on a HP workstation desktop (somewhat older but very stable). I have Tried: turning off breakers (A/C ect...), isolation transformer, different computers, different buildings (30 miles difference), removing the scope leads and capping (0 Ohm) the inputs (we do have a cell tower 1/4 mile away), one scope, either scope, both scopes, ect... I have looked for noise on Picoscope but I don't see any - I will send these separately but shortly...

Sometimes the noise may be gone for hours and sometimes there is 20-40 minutes of "quiet" between the noise that may last 20 minutes to a couple of hours.

I have tried a laptop with All channels capped (0 Ohm) with and without battery and still get the noise (See attachment "Jun 28, 2020 3-58-33 PM" and the config file with battery only - all channels capped)
Jun 28, 2020 3-58-33 PM.picolog
(736.5 KiB) Downloaded 54 times
Jun 28, 2020 3-58-33 PM config.picolog
(5.5 KiB) Downloaded 43 times
Attachment "Jun 28, 2020 4-09-35 PM DATA" and the config file
Jun 28, 2020 4-09-35 PM DATA.picolog
(1.2 MiB) Downloaded 44 times
Jun 28, 2020 4-09-35 PM config.picolog
(5.5 KiB) Downloaded 49 times
- Done on the laptop - This configuration Channel A (blue) of both scopes have the factory scope probes set on 1X and are connected to a 12 volt battery the rest are capped with 0 OHM caps - read the annotations because I ran with ac power and then disconnected the AC to laptop - then disconnected AC power to the scopes (that is why there is missing data on the bottom 4 channels) then plugged the AC power back on the scopes... I am running different sample rates...
Generally 100 MS is MOST always completely noisy... 200 MS the pattern can be seen quite well... at 500 ms the pattern is there but very light and 1 sec is hard to see

To add more confusion... See attachment "Jun 28, 2020 11-05-36 AM data" and config file - was done on the workstation desktop - all channels are capped but taking samples at different rates however both A channels are at the same sample rates B the same ect... If you look at B channels they do not have noise at the same time. There is an overlap and C channels have a different pattern

I am attaching ONLY a couple of Picoscope files because the settings are endless. Just look at the file name for the scope I used and the time base. I did capture these at the same time because I was running 2 instances of Picoscope at the same time. If anyone wants a certain set up let me know and I will capture it

Thanks for your help DJETH

DJETH
Newbie
Posts: 0
Joined: Mon Jun 29, 2020 3:17 pm

Re: Noise on Log Channels

Post by DJETH »

Here are the scope attachments. These should show the noise floor for the scope... They all look good to me ????
Attachments
5444 5us - 20200629-0003.psdata
(41.82 KiB) Downloaded 45 times
5444 1ms - 20200629-0002.psdata
(2.6 MiB) Downloaded 51 times
5442 5us - 20200629-0004.psdata
(36.64 KiB) Downloaded 51 times
5442 1ms - 20200629-0001.psdata
(2.3 MiB) Downloaded 49 times

DJETH
Newbie
Posts: 0
Joined: Mon Jun 29, 2020 3:17 pm

Re: Noise on Log Channels

Post by DJETH »

Here is some more testing I did...
I started to take slower samples in the log. The file "Jul 1, 2020 6-04-21 AM NOTED" has sample rates of 1 to 3 seconds and shows the noise (at different times...Weird...). This log shows the noise only 1 time and for a short duration. My bet there would be more noise at the faster sample rates as I recorded in the past. I did leave Axis 3 that at 200 millisecond sample rate but does not show the noise because the peak to peak is 0 - 12 volts DC and the voltage base is too large to see the noise.

File "Jul 2, 2020 5-20-14 AM" has no noise at all but the sample rates are 4, 5, and 6 seconds. First log with no noise I have had but a way slower sample rate. I'm thinking that logging with the 5000 series scopes with sample rates under 1 second show noise (that should not be there). Anybody else have this issue? or can cap the inputs on their 5000 scope and run a test for me?
Attachments
Jul 2, 2020 5-20-14 AM.picolog
(1.33 MiB) Downloaded 42 times
Jul 1, 2020 6-04-21 AM NOTED.picolog
(862.5 KiB) Downloaded 50 times

DJETH
Newbie
Posts: 0
Joined: Mon Jun 29, 2020 3:17 pm

Re: Noise on Log Channels

Post by DJETH »

Well its not local interference... Here is a log that I took today about 30 miles from the other location... All inputs are capped (and set at 50mv) and am recording at 4 different time bases 100ms, 200ms, 500ms and 1 sec. You can definitely see the pattern. This is what I am seeing captured and transposed on my temperature and frequency logs.

Hardware/Software problem... Pretty sure...

NOTE: I captured using the 5444D MSO, Blue Pico USB cable, and Pico power supply. The location, computer, ect was all different.
Attachments
Jul 7, 2020 12-20-08 PM.picolog
(785 KiB) Downloaded 44 times

Martyn
Site Admin
Site Admin
Posts: 3846
Joined: Fri Jun 10, 2011 8:15 am
Location: St. Neots

Re: Noise on Log Channels

Post by Martyn »

As a first look the PicoLog traces are greatly zoomed on the vertical axis (120uVolt) when compared to the PicoScope traces (10mVolt) so the noise automatically looks bad.

I would also suggest sampling at 1msec per channel in PicoLog and select "Latest Sample" and not "Average" to see what you get as this will be a closer match to the PicoScope setup.
Martyn
Technical Support Manager

DJETH
Newbie
Posts: 0
Joined: Mon Jun 29, 2020 3:17 pm

Re: Noise on Log Channels

Post by DJETH »

Martyn,

I will set this up and post the results... Channels A and C at 1 msec sample rate and B and D at 200 msec all set to latest sample when i do the recording... I have not seen the noise pattern on any sample rate less than 50 msec (that I remember) so should be interesting. What do you think about the pattern? Quiet sometimes and noisy other time's.

DJETH
Newbie
Posts: 0
Joined: Mon Jun 29, 2020 3:17 pm

Re: Noise on Log Channels

Post by DJETH »

Martyn,

when I looked at the first log using the "Latest Sample" setting I was reminded why I have not been using it...Too loud. Look at "Jul 17, 2020 11-20-56 AM 200MS Latest Sample". The temperature readings are pretty unreadable and need to be averaged. I turned the Latest sample to average on Channel B only. See log file "Jul 18, 2020 6-27-55 AM Temp 200ms" to get a comparison. No doubt average is the way I want to go. however you can see the noise come and go as I stated and have shown in the above posts.

I am restating the Issue I have. When I record at 200ms - 1 sec using the average setting I get a noise that comes and goes. I have narrowed this noise down to the scopes and/or software and am asking what the noise is, what is causing it, and can it be eliminated or filtered out? See PICTURE "Jul 18, 2020 4-06-42 PM data 1ms to 200ms compare LS Avg PICTURE". This file was 38MB so I was not able to upload the file only a picture. All 8 channels (2 scopes) are using the factory supplied probes set on 1X. All 8 probes are connected to the same 12 volt battery. Look at both channel D's (200ms avg) and you can see that the recording is very clean for about 20 minutes and noisy in the others. If I were to continue recording this pattern would repeat. One interesting note is that the two scopes are not "clean" in the same time frame.... I have seen this before also. I can understand that Latest sample is needed in some instances but I am looking for average on this project.

You requested that I record at 1ms time frame so I included it on the same graph. Both channel A's and dark green channel C are recorded at 1ms with Latest sample. These have a voltage change of roughly 250mv. At 200ms latest sample roughly 130mv and at 2sec and 5 sec roughly 100mv. The 2 D channels about 40mv (clean area) if you can get rid of the noise. File "Jul 18, 2020 5-40-10 PM 1ms to 5 sec" all channels connected to same 12 volt battery. Dark Green channel C you can see the noise pattern (200ms avg). Channel A has the 1 ms time recording you requested. Brown channel D I bumped the bnc connector going into the scope and caused that .005 volt drop you see. the rest are random.

I did do a log at 1ms with the inputs capped but I let it run too long and the files is 158MB (see picture "Jul 17, 2020 5-59-07 AM - Picture". I can redo this if you need me to.

I am also attaching "Jul 18, 2020 9-41-28 PM compare LS to Avg". All leads same as above on the same battery. The 5 sec and 200ms are so the Latest sample and average can be compared... Of course the 200ms avg has the pattern. The others are random time bases.
Attachments
Jul 18, 2020 9-41-28 PM compare LS to Avg.picolog
(495.5 KiB) Downloaded 34 times
Jul 18, 2020 6-27-55 AM Temp 200ms.picolog
(1.97 MiB) Downloaded 37 times
Jul 18, 2020 5-40-10 PM 1ms to 5 sec.picolog
(4.68 MiB) Downloaded 35 times
Jul 17, 2020 11-20-56 AM 200MS Latest Sample.picolog
(619 KiB) Downloaded 37 times

DJETH
Newbie
Posts: 0
Joined: Mon Jun 29, 2020 3:17 pm

Re: Noise on Log Channels

Post by DJETH »

Sorry... But the website would not allow me to post my two 500kb .png pictures so I put them into a pdf
Attachments
2 pictures.pdf
(504.72 KiB) Downloaded 41 times

DJETH
Newbie
Posts: 0
Joined: Mon Jun 29, 2020 3:17 pm

Re: Noise on Log Channels

Post by DJETH »

Bump... Martyn... checking if you have anything yet?

Martyn
Site Admin
Site Admin
Posts: 3846
Joined: Fri Jun 10, 2011 8:15 am
Location: St. Neots

Re: Noise on Log Channels

Post by Martyn »

I have asked for comment from the development team, I will post a reply once I have a response.
Martyn
Technical Support Manager

DJETH
Newbie
Posts: 0
Joined: Mon Jun 29, 2020 3:17 pm

Re: Noise on Log Channels

Post by DJETH »

Excellent... Thank You...DJETH

Martyn
Site Admin
Site Admin
Posts: 3846
Joined: Fri Jun 10, 2011 8:15 am
Location: St. Neots

Re: Noise on Log Channels

Post by Martyn »

The development team have looked and don't see any particular issues.
They have suggested that you do a test with PicoScope 6 where you set the time across the screen to a much longer value, you would then see that the trace results are similar with PicoLog 6. Try with 10 Secs/div for starters and set the number of samples for each screen at the top to 100kS so that you get the same 1msec sample interval that PicoLog uses.
Martyn
Technical Support Manager

DJETH
Newbie
Posts: 0
Joined: Mon Jun 29, 2020 3:17 pm

Re: Noise on Log Channels

Post by DJETH »

Martyn sorry for the delay and thanks for getting this to the development team…
The team seems to be misunderstanding my problem. 1ms sample interval is the time base they are wanting me to test (which I did) but the time base I keep talking about is between 200ms and 1sec (which I also tested). I did some further tests using both Picolog and Picoscope at the same time and I did not get the same pattern as they indicated I should.
To make sure they understand my issue I have marked up a log from a couple of days ago.
See “Aug 4, 2020 3-50-52 PM - Data and Notes” for the picture below


And here is the one I took last night (I don’t think I need to mark up this one ). "Aug 5, 2020 8-02-47 PM 16hrs"


So there is no confusion… I ran 2 scopes at the same time. One using Picolog and the other Picoscope. Inputs A of both scopes are connected to the temp sensor output voltage and both channel B’s to a different temp sensor output. Both channel C’s are connected to the same 12 volt battery and both channel D’s are individually capped on the BNC input. Picolog is recording at 200ms sample interval… This is a 16 hour log and the repeating pattern is evident.
According to the Picoscope traces (several different set ups) (attachments will be on this and the next post because of the size) I don’t get the same pattern of noise and quiet… HOWEVER look at Channel C (the 12 volt battery) in any of the Picoscope attachments and you can see that the scope trace Peak to peak is pretty close to the peak to peak of the Picolog when the noise is there. Picolog is supposed to be averaging…. It looks like the software/hardware is struggling to do this consistently.
Pleas have your development team indicate which sections are normal (or have not particular issues) on the above picolog the noise or the “no noise” part.
If you need me to do any other tests… different time bases, environmental locations, computers, ect… I’m happy to do them. I would like to get the noise removed if possible.
Attachments
20200625-0004 (21).psdata
(65.43 KiB) Downloaded 2 times
20200625-0004 (40) 10ms.psdata
(1.76 MiB) Downloaded 3 times
20200625-0004 (34).psdata
(1.76 MiB) Downloaded 2 times
20200625-0004 (31) 1ms.psdata
(1.76 MiB) Downloaded 3 times
Aug 5, 2020 8-02-47 PM 16hrs.picolog
(3.88 MiB) Downloaded 5 times
Aug 4, 2020 3-50-52 PM - Data and Notes.picolog
(198.5 KiB) Downloaded 3 times

DJETH
Newbie
Posts: 0
Joined: Mon Jun 29, 2020 3:17 pm

Re: Noise on Log Channels

Post by DJETH »

the other PicoScope attachments... You can look at the time stamps for the Picoscope and compare to the same time on Picolog...I don't see the same pattern regardless of the setup.
Attachments
20200625-0004 (30).psdata
(726.5 KiB) Downloaded 3 times
20200625-0004 (27).psdata
(124.18 KiB) Downloaded 3 times
20200625-0004 (26).psdata
(124.39 KiB) Downloaded 2 times
20200625-0004 (19).psdata
(69.4 KiB) Downloaded 2 times

Martyn
Site Admin
Site Admin
Posts: 3846
Joined: Fri Jun 10, 2011 8:15 am
Location: St. Neots

Re: Noise on Log Channels

Post by Martyn »

With PicoScope 6 can you set 500s/div , 5MS , 14bits, with four channels enabled and collect a buffer. Open the properties panel on the right (Views->ViewProperties) to ensure that the scope is sampling at 1kS/sec or 1000S/sec
Martyn
Technical Support Manager

Post Reply