| Author |
 Topic  |
|
|
codref
8 Posts |
 Posted - 12/29/2018 : 06:30:24
|
Hi! I recently bought one of your instruments (the EMF-390), quite happy about it but the RF readings, when connected to the USB, are influenced by the attached computer interference.
For example, the frequency reported is always 672Mhz (or near it), and the values reported (both total and instant) are increased.
When I turn down the computer, of course, the reading get back to normal values.
I tried to use several USB cables (i.e. with or without ferrite bead) and also to switch computer (laptop with/without AC and raspberryPi with/without AC). Always the same results.
I'd like to setup a 24/7 monitoring system, I already wrote the software but, with those readings, it's quite useless.
Can you provide me some suggestion about how to insulate the equipment? Right now the USB connection is only good to download the data stored in memory - but it's not realt-ime monitoring!
Thanks!
Davide |
|
| Reply #1
ZLM
1261 Posts |
Posted - 12/29/2018 : 09:11:02
|
What is your correct reading and what is your wrong reading?
|
 |
|
| Reply #2
codref
8 Posts |
Posted - 12/30/2018 : 02:33:32
|
You can see here a short video showing the interferences.
https://youtu.be/uDp8JqARLf0
Without USB connected I read ~3mW, with the connector plugged (and the pc turned on), about 5-8mW. The unusual thing is that the meter reports mostly RF in the ~672 MHZ.
The laptop had AC adapter connected and grounded. |
|
|
| Reply #3
Staff Member
USA
7 Posts |
Posted - 01/03/2019 : 22:32:29
|
Have a EMF390 as well.
Exact same issue as you're having.
Mine oscillates between 432,664,and 882 MHz.
Also have yet to be able to successfully download internal
recorded data into any of the 4 Windows 7 Pro computers
that are laying around here.
What OS did you write your code for?
Nice if it's Python.
Intended to set this up on a Raspberry PI3 then
use Azure/Power BI to create reports.
It's a spectacular device for the money just a bit
rough around the edges.
Please advise if you get an answer about the noise transfer
from the USB interface.
73's
Staff Member
|
|
|
| Reply #4
ZLM
1261 Posts |
Posted - 01/04/2019 : 22:19:55
|
That does not mean wrong reading.
That I believe is a correct reading after USB connected.
Some RF signal been transferred from the USB cable to the meter.
If you can eliminate the signal from USB cable, then the reading will not change. |
|
|
| Reply #5
codref
8 Posts |
Posted - 01/06/2019 : 12:19:08
|
Thanks ZLM, if you do not have any other suggestion, I'll try with other workarounds, but I think the problem is bigger than my capacities - dealing with interferences at board level is not easy.
In any case I appreciate the serial connectivity, it for sure has potentials.
@StaffMember, I was writing a tool in Rust, unfortunately, to export the EMF390 data the developer relied on "C unions type" (for the floating points), which are not easily supported by higher level languages like python.
As soon as I finish the development (I got stuck with the interferences and so I started to doubt about the whole USB thing), I will push the code (and the executables) on github. Rust can cross-compile for linux as well as for windows, so no big deal! |
|
|
| Reply #6
codref
8 Posts |
Posted - 01/07/2019 : 11:47:28
|
I think I found the problem (and the solution).
Since the most probable interference source could only have been the battery charging circuit, I just removed the battery!
Now the readings are quite similar to the ones without the USB cable plugged.
Since the backlight, without the battery in place, flickers a bit, I suggest to turn it off completely via the config menu. |
|
|
| Reply #7
codref
8 Posts |
Posted - 01/07/2019 : 11:51:55
|
There's a label which warns about removing the battery, what's the reason?
I saw that the realtime clock stopped for a while (no big deal in my scenario).
The configuration seems saved in non volatile memory, as well as the saved data (which is still there) |
|
|
| Reply #8
ZLM
1261 Posts |
Posted - 01/08/2019 : 08:18:09
|
@codref, thank you for your solution. That means the interference from charging circuit. In this, the interference should go away if battery fully charged.
For the warning label, some user just do not know the battery has +/- and install it in wrong direction and then ask for the replacement.
|
|
|
| Reply #9
Neko2
Spain
9 Posts |
Posted - 01/24/2019 : 01:16:38
|
| in my GQ emf380v2 do not change the readings when I connect the usb and it is charging the battery. |
|
|
| Reply #10
Neko2
Spain
9 Posts |
Posted - 01/24/2019 : 01:20:30
|
| Only change EF value from 0.0 V/m to 1.2 V/m. RF and EMF not change. |
|
|
| Reply #11
EmfDev
2250 Posts |
Posted - 01/24/2019 : 10:27:02
|
| @Neko2, that's still normal. |
|
|
| Reply #12
Staff Member
USA
7 Posts |
Posted - 09/06/2019 : 15:01:38
|
quote:
Originally posted by codref
Thanks ZLM, if you do not have any other suggestion, I'll try with other workarounds, but I think the problem is bigger than my capacities - dealing with interferences at board level is not easy.
In any case I appreciate the serial connectivity, it for sure has potentials.
@StaffMember, I was writing a tool in Rust, unfortunately, to export the EMF390 data the developer relied on "C unions type" (for the floating points), which are not easily supported by higher level languages like python.
As soon as I finish the development (I got stuck with the interferences and so I started to doubt about the whole USB thing), I will push the code (and the executables) on github. Rust can cross-compile for linux as well as for windows, so no big deal!
Thanks for the reply.
Will look into RUST?
Real-time Monitoring is exactly what we need it for.
Please let the forum know as to your progress.
73's
Staff Member
Digitellum
Scottsdale, AZ
|
|
|
| |
Topic |
|
Wrong readings while connected to USB
