EMI and Ferrite Cores
-
@Brains said:
@scottalanmiller We could try for testing, but that bag has to be 100% closed right? I assume if it wasn't, it would still block a majority of it
Correct, EMI is like light, not like sound, so having a small hole will do very little to let EMI in. Any amount of coverage will do a lot.
-
A simple Faraday Bag is a sheet of aluminum foil. You may wish to insulate any metal surfaces and connectors prior to prevent possible shorting (never know).
Just wrap the device up with some foil (tin foil hat brigade) and see if that helps.
-
Thanks. I'll give it a go and see the results. Ill update in a couple of days
-
Well. No dice unfortunately
Tests performed:
- Turned off all other lights. Incandescent (no florescent) attached to 4 dimmer switches
- Turned off all other surge protectors
- Cell Phones turned off
- Unplugged all other electrical devices in room
- Replaced extension cable with shielded extension cable
- Wrapped unit in aluminum foil
- Wrapped Cables in aluminum foil
- Wrapped USB Extension cable join with built in cable and majority of built in cable with aluminum foil
- Wrapped Sensors and User's Hands in aluminum foil
- Moved Device to table where printer sits <~~Cut the 180Hz signal from 3 to 1.5
- User reports similar results when walking around the building with laptop. Some places its not as bad, but it is always there
- Checked nearby offices for random items plugged in - None
Changes made:
- Moved all computer items to new surge protector, plugged into power cleaner
- Left shielded USB extension cables up there
Possible avenues
- Ask Maintenance to disconnect the power line running outside of the wall that is powering User's Cubicle Outlets and only use wall outlet
- Speak to Electrician to see if there are filters that can be placed in electrical room in case the whole building has "dirty power"
- Would explain why User can reproduce this with the laptop (Running on Battery) in various places in the building, but it does not reproduce at his home.
- Bring in Professional (No idea what he would be a professional in)
Important Notes
- Problem is only existent when using sensors far apart from each other (more opportunity for EMI to get in the middle)
- Doesn't happen if sensors are 2 inches from each other on the same arm.
- Happens when 1 sensor on each hand and hands are shoulder-width apart
- Parking Garage Below Him - Maybe something to look into to? Wonder if there is any equipment below him?
- This is happening on both devices
- J&J SEMG sensors
- Sampling Rate = 1024 per second
- Thought Technology SEMG
- Sampling Rate = 2048 per second
- J&J SEMG sensors
-
Also when I worked for the cable company we had "Signal Leakage" meters. Is there a similar meter that can assist me in pinpointing the source or direction the interference is coming from?
-
@Brains said:
Also when I worked for the cable company we had "Signal Leakage" meters. Is there a similar meter that can assist me in pinpointing the source or direction the interference is coming from?
I see that you have dimmers - these themselves can be noisy. if they are off, they should not generate noise.
-
@g.jacobse said:
@Brains said:
Also when I worked for the cable company we had "Signal Leakage" meters. Is there a similar meter that can assist me in pinpointing the source or direction the interference is coming from?
I see that you have dimmers - these themselves can be noisy. if they are off, they should not generate noise.
So much that you can actually HEAR them!
-
any other ideas or suggestions guys? This is pretty far out of scope, and I have limited experience this deep in.
-
@Brains said:
any other ideas or suggestions guys? This is pretty far out of scope, and I have limited experience this deep in.
Unfortunately this is an electrical engineering thing and not an IT thing. While we know of the issues, I don't think that anyone here is trained or qualified to help. Have you looked into bringing in an EE for this?
-
Im talking to some of my EE friends from college, but they havent been of any significant help yet. I would like to pass this off to a professional qualified to deal with it, but I dont even know where to start. Any ideas?
-
Did some EE, still do for a hobby.
Having trouble figuring out what the actual problem is in your post - 180hz signal shouldn't be there? What should? How do you know it's 180hz?
Could you link to the products you're using? From below you didn't really give what model etc. I found these, but not sure.
http://www.jjengineering.com/C6.htm -
@MattSpeller said:
didn't really give what m
Thanks Matt. I appreciate the offer of help.
I have attached a screenshot from the monitoring software. Yes the 180Hz signal should not be present, or at least prevalent in the pictures. He is reading very low frequency electrical impulses and this signal is throwing off the tests.
He is sending me the model numbers of the equipment in the morning. In the meantime, could you help me pick out some ferrite cores that may be able to help filter this 180Hz (3rd Harmonic?) interference? I was about to order some, but then I noticed that there were different types for different frequencies, and I cant make heads or tails of which one I need.
-
@Brains Alright, well from what I remember I don't think chokes work that way - at least how you want it to in this instance. I will have to read up on them again to make sure I'm correct there!
Onwards; 180hz is super low and an odd one. 3rd harmonic of the 60hz powerline, no doubt that could be the source. Unusual to only see the harmonic though, do you also see 120 & 60? (60 should be a huge spike if it's AXT (alien cross talk) from a transformer or something nasty). You've already tried isolating it from outside noise and it sounds like you did a pretty good job of it.
Next up I'd look at the device it's self and read it's manual. I'm willing to bet it has internal filters for this kind of thing - if not I'm almost certain that it'll mention how to deal with AXT as it's such a sensitive device you won't be alone in having issues with interference.
-
Thanks for the followup Matt.
I went through some of the logs of tests performed. When there was a transformer close, it definitely cycled every 60/120/180, but with everything turned off, this is what we see
He has spoken with the manufacturers of both devices and they have been of no help unfortunately. When I get the model#'s tomorrow, I will try to give them a call and see if we can have a detailed conversation about our options. I do not expect this to be productive as the user using this equipment is pretty intelligent and understands some of this better than I do and was not able to glean any information. I am very good at cutting through red tape and double talk from tech support, so maybe I will be able to get something helpful.
FYI he is usually using a desktop to perform these tests, and the problem is apparent on both the PC and laptop (Laptop is unplugged of course)
-
Very interesting
If you have access to an oscilloscope or a signal gen you could probe it out a bit further... very intrigued and I hope you let us know how it plays out!
-
-
Page 5, try self calibration
http://www.thoughttechnology.com/pdf/manuals/SA7510 rev. 7 ProComp Infiniti User Manual.pdfTry replacing the fiber - that shit breaks so easily and I'm guessing the Doc isn't familiar with using it
-
@MattSpeller so you think there may be a small break in the fiber that could be letting the EMI in, its just not bad enough to cause the unit to stop working?
Ill send this over to the user and ask him to try it
-
@Brains just trying to think of the most common things to break - fiber is always top of the list.
-
@MattSpeller awesome. He is receptive and will take care of that this weekend or Monday. I am open to any ideas, I was also wondering if it might be a case of "dirty power" in the building? Maybe we have some feedback from the AC Units being sent back along the line? Since this happens in many different places in the building, maybe we could install something to "clean up" the power. Before today I always considered those power cleaners a scam, but this seems like the intended usage if I ever saw one
