I have a working (sofware and hardware wise) project that interfaces with a scanner radio. Everything was fine until I took the project mobile. Previously the radio had its antenna mounted outdoors and well away from the display so I never noticed any inteference problems. In the vehicle the display is on the dash and the antenna is a few feet away. I noticed broadband noise was generated in the VHF (108-150 MHz appx.) range. I was planning on embedding a small GPS module in the project behind the display as well. I couldn’t get it to work. Thinking it was a bad module or wiring I was at a loss then I realized if I moved the GPS about 18 inches from the display it blinked and received a GPS signal lock.
I then looked at what is emitted from these things using an SDR receiver from 25-200 MHz. It’s terrible. There is a massive broadband noise source emitted in the VHF range. I’m not seeing as much at 1545 MHz L band GPS but I’m thinking the huge broadband noise is swamping the preamp in the GPS. I tried numerous different types of GPS modules and it does it to them all. I tried wrapping the entire display in copper foil tape, still no improvement.
I’m not sure how these are part 15 compliant. Has anyone ran across this issue? Are there any known useful mods (filter caps, inductors) that mitigate it? I have tried adding. 1, .01 uF caps at the connector back pins to ground and tried a 47 uH inductor in line with power but it didn’t help.
At this point I have to abandon the project if I can’t figure it out. Putting my finger on the board seems to couple more energy out if it’s near the large flat memory chip and near the real time clock. I disabled the RTC chip by lifting the VCC pin but that didn’t help. I don’t know if it’s the memory reading and writing that’s doing it or maybe the FPGA or what but it’s frustrating as hell not being able to do anything about it. Obviously, many items like vehicle GPSs and other similar items use touch panels (cell phones) so it can’t be something that can’t be addressed.
The noise in the VHF range is typical for most MCU boards which are clocked around 48MHz - 72MHz. I’d say that this is not forcibly a typical Nextion problem. Any STM32 MCU with a SPI link to a LCD controller could do that. Best practice here is shielding.
I have Teensys and plenty of other devices running similar clock speeds that don’t do this. It’s a full on broad band jammer. I tried shielding with copper tape with conductive adhesive and it just lauged at me as it continued to radiate. I also noticed that the corner anchor points aren’t grounded. There are quite a few areas that aren’t grounded. They implemented what appears to be a strange ground plane layout. It’s a shame because the software is nice and the protocol is very user friendly. Unfortunately, at this point I will have to abandon it all and move to 4D Systems or something else.
o my god, i come to know now that my project is working that i will probably have to change the display and then reprogram on other compiler etc etc because not compatible with nearby GPS! crazy, they created a jammer, a signal generator, a scrambler.
Has anyone solved it and how? thanks
There are reasons why HMIs from 4D Systems have a massive metal shielding around which makes them much more expensive.
Each LCD panel is, due to its multiplexed column and line addressing, a directional phase modulated planar antenna. And this adds to the wiring (flat ribbon cables and PCB) radiation. Did you ever look at the EM report part of the CE certification available here? This could have told you in advance if a Nextion HMI was a suitable component for your SDR or GPS project or if additional precautions have to be taken. Fortunately, you discovered this during prototyping and not only after mass production started.
I’m looking at a 4D Systems 4.3" Diablo next to a Nextion 4.3" and I’m not seeing any massive metal shielding on the 4D, actually I see no shielding at all. The 4D board is small and just floats on the back of the LCD only held on by the plastic frame. Peel off the plastic frame and it flops about. In theory, the Nextion should yield far less RFI due to it all being mounted to one large PC board. However, I noticed that the ground plane layout of the Nextion is a bit bizarre. Places one would think should be ground aren’t and are just sort of left floating.
I’m beyond frustrated. I gave up the project for about a year due to the frustration and decided to give 4D Systems a try. After getting my head around how they do things it’s flat out horrible. You can’t even update the color of a text field or a font dynamically. You would have to write a secondary field with different attributes over that same location.
You can’t alias object names in the code for reference either. You are left with Strings1, Strings2, etc… blah blah. A guy would need to draw a map just to read his own code.
I wish Nextion ran on the 4D Systems’ beautiful hardware.
I should visit the forums more regularly
I wouldnt trust any emissions complaint certificate as far as i could throw them from nextion…
Saying that though, The board itself is rather basic in design. And given that its just am MCU with a built in crystal and external voltage regulator, Im kinda surprised its THIS bad.
Makes me wonder how much noise my pcbs make
You say the problem became worse when you took it mobile. Is what ever your using as a power regulator causing trouble? Inductors can emit a fair bit of RF noise.
Buck/boost circuits can cause trouble…
I have a UV lamp i use for curing my resin prints, and it knocks out the DAB radio when ever I turn it on. Typical Chinese poop.
Good points but I don’t think the power supply is the issue. I’m using the same mobile supply with a pi and DSI display now with zero RFI issues. The GPS is working fine too.
1 Like
This forum is in no way affiliated with NEXTION®, ITEAD STUDIO®, TJC®, or anyone else really. All product names, logos, and brands are property of their respective owners. All company, product, and service names used in this website are for identification purposes only. Use of these names, logos, and brands does not imply endorsement from the respective rights holder(s).
