Kia Niro Forum banner
  • Welcome to the Kia Niro forum! We discuss all models of the Niro, including the Hybrid, PHEV and EV versions. We are glad you stopped by. Feel free to browse the various topics, along with out FAQs. To enable posting, you need to register for a user account. There is no cost for this. Just click in the upper right corner where it says Login/Join. We look forward to your continued parcipitation.

Radio frequency interference susceptibility of Niro smart key system

1 reading
1.8K views 5 replies 4 participants last post by  sc489  
#1 ·
So here's the scenario: had to drop by the Home Depot tonight to pick up some supplies and when I came out, walked up to the car and pressed the button on the door handle and nothing happened... I thought I must not have hit the button hard enough and tried again.. Nothing.. Hmm.. Got out the remote, pressed the unlock button.. Nothing.. Pressed it again, nothing, pressed the other buttons on the remote, also nothing...

So, first thought was: Is it the 12V battery? Am I also a victim of this? The car looked dark and dead, no light in the door handle that normally illuminates when you walk close by with the smart key, so dead battery was a reasonable thought, but then this is also a 3 month old car and I drive it every day and it never sits for long unless I have a battery maintainer connected, so it can't be dead already.. It must be something else in those dodgy Hyundai/Kia systems that I'm still leery of, coming from the utter reliability of the Toyota world 😝..

So what can I do? Well if it was the battery, I could just turn around and go back into the Home Depot and buy an overpriced battery boost box from their meagre automotive section (I honestly don't know why they bother, who buys automotive supplies from Home Depot? Guys with dead Niros in their parking lot, that's who..), but first I wanted to get my purchases stowed, so I got out the physical key, figured out how to lever off the cap on the door handle and unlocked the car, and when I pulled the handle to open the door, the alarm went off.. OK, physical key doesn't disable the alarm, but also, the alarm is actually going off, so not the 12v battery... But now how do I turn off the alarm? My smart key buttons were not working, so I tried pushing the engine start button with the corner of the key as specified in the manual for a dead key battery and that worked: alarm turned off and the car went into accessory mode. I then also noticed that when I took the key away from the start button the screen would display the "Key not in vehicle" message, and it would go away when I brought the key back near the start button.. OK, so dead battery in the smart key? Again, this is only a 3 month old car, so the coin cell should not be dead so soon, but I've gotten bad coin cells even from name brands from reputable stores, so it still could have happened.. But then I noticed that the key has a flashing LED on the side when you pressed any of the buttons, and it was working: flash when I hit a button, but no appropriate response from the car, so not the key battery either..

At this point, I got the car unlocked, my purchases in the back seat and it wasn't the 12v battery, and it could start and drive just fine, so I went back home, where if it was going to happen again, I at least am at home instead of a parking lot in the dark, and I have access to all my tools and my diag scanner if I needed them.

Of course, when I got home, everything worked just fine: the smart key lit the little light in the door handle as I approached, and the button locked and unlocked the doors, and the buttons on the smart key also worked fine.. So what happened?

Well if you read the Subject: line of this tale, it was Home Depot's fault.. Huh? You say, that's not what the Subject line says, and how is it Home Depot's fault? Well, OK, maybe it's not really Home Depot's fault, but Hyundai/Kia certainly dropped the ball as far as RFI susceptibilty testing when they designed or chose the supplier for the smart key system.. When I got to the store, I thought I had scored, as I got the closest parking spot right across from the entrance, but I forgot until I got home and everything worked normally again that all the Home Depot locations I've ever been to use a fancy radar-based door opener for their sliding doors, and I remember years ago,a cheap radar detector I had going off whenever I drove in front of Home Depot and Walmart stores... So my car, being right across from the door was being blasted by the door sensor and this interference was basically jamming the smart key. Now, my premise in laying the blame at Hyundai/Kia's door lies in the presumption that the door opener used in literally thousands of Home Depot locations in Canada and the US, would meet FCC/ISED Canada regulations for output power and out of band emissions, so the fact that the car systems got swamped by this signal would imply something in the circuitry (probably related to the RF filtering in the input stage, insufficient bypass capacitors, or lack of physical RF shielding on the circuit boards.. Basically, this should have been something that was tested for either by Hyundai/Kia themselves or by whatever supplier sold them the smart key subsystem... In fact I looked up the system Home Depot uses, and its signal is way up at 24GHz and has an output power of 20dBm (100 milliwatts, or about half the power of a typical cell phone).. Bandwidth is 250 MHz, so there's a lot of opportunity to generate spurious emissions at other frequencies, but again, this is what the FCC regulations are for: before a vendor can sell a RF product, they must have it tested by an independent lab to validate that it won't generate enough power in signals out-of-band that would cause interference to other surrounding systems. Anything lower than the limits specified in the regulations is presumed to be low enough to either not cause interference in the first place or be simple enough to mitigate with measures such as filtering and shielding by the producers of other technology.. The Hyundai/Kia smart key uses 125kHz for the transponder and 433 MHz for the transmitter, which are both way down, well clear of the radar door opener's center frequency of 24.175 Ghz and not even a nice even multiple of 433 MHz where you could blame bad harmonics.. The 125 kHz was also actually working fine as I could still start the car by pressing the start button with the key... Instead, I kind of suspect a direct susceptibility to the 24GHz signal itself and not to any of the out-of-band components, so bad engineering and/or subpar testing.. This is why you don't just test success path...

In any case, this is starting to get technical and going to turn into an IEEE journal letter if I continue, so I'll stop here, but I just wanted to share the story so if anyone else ever has issues unlocking their car at a Home Depot or maybe near the control tower of an airport, maybe it's just RF interfence causing the smart key to not work properly, and if it happens, don't panic, try the "dead key battery" procedure and see if you can get your car started and see if the problem goes away when you drive to a different location.. Or just give up hunting for the close up parking space and just park in the boonies (not going to happen 😜)..
 
#2 ·
All electronic systems are susceptible to RF interference unless they are engineered as a perfect Faraday Cage which is not possible. Is the effect repeatable if you park in the same spot again? The problem could be a software engineering bug (human error) of which there are many certainly in the infotainment software.
 
#4 · (Edited)
Agreed, all systems are susceptible to RF, but that's the reason for FCC and various industry regulations as well as for conducting testing by manufacturers.. By limiting out of band emissions to a low level and restricting in-band power output, it makes systems generating RF less likely to cause interference, and by testing, it allows manufacturers of other systems to determine if their designs are vulnerable to interference and to put in mitigative measures (could just be some bypass capacitors in certain areas, a bandpass filter in their own RF input stage, or in the extreme case, an RF can (ie, a Faraday cage) soldered around the most sensitive components)... I'm an electrical engineer and one of my former roles was RF design for manufacture for a cellular equipment vendor, which is one of the toughest design jobs as not only do you have to ensure that the system works under normal conditions, but that it will also work under adverse conditions, ie, if there is interference present, and on top of that, you have to ensure that manufacturing tolerances eg circuit board trace thickness, variations in board layer bonding, component value tolerances, variations in pick and place locations and component shift during soldering, how the board mounts in the chassis, etc, etc don't impact the performance of the circuit.. The only way to tell is basically a heck of a lot of testing.. We'd put the circuit in a chamber with a signal generator blasting various types of waveforms at various frequencies, at below FCC permitted power levels, at beyond permitted power levels, all to ensure that our system didn't fail and if it did, we had to find a way to get it working again, all while constrained by BOM costing, costs for mechanical engineering changes, and the corporate drive for ever more cost reduction in general...

IMO, this obviously didn't happen for the smart key system, but it should have: vehicles operate in conditions where they can be expected to encounter all sorts of RF: other vehicles on the road with radar collision avoidance and driver assist systems, speed radars, toll collection transponders, 2-way radio traffic, etc.. These systems really should have been more thoroughly tested... Edit to say that the Forum highlighted another thread by someone who reported that their smart cruise just turned off without human intervention every time he passed a particular spot on the highway, and the consensus was some sort of RF interference in the area causing it, so just another example.. Could be just lack of RF knowledge as a car maker is probably full of mechanical, electrical and software engineers, but usually not a lot of RF, or maybe it's a corporate cultural thing that they don't test these conditions because it adds cost, but it's an obvious thing to an RF engineer..

Also, nothing wrong with either the 12V battery or the coin cell and I read codes when I got home, and nothing other than the fact that I have no TPMS in my winter tires..vvI'm going to go back to the Home Depot this evening as I need a few more things and see if I can get the same parking spot if I go later in the evening on a weekend and l am sure that my smart key will stop working again..
 
#5 ·
I recall similar issue with Car Alarms on both Honda and Hyundai's when I was on Mt Soledad in San Diego (park at the summit) the CBS and ABC TV affiliates plus microwave towers are all over the area. It wouldn't happen all the time but often enough to be noticed.

Back in the 80s whenever the Navy had gunnery practice on San Clemente Island beach residents reported Car alarms wouldn't unlock or went off and garage doors opening and closing or disabled.

It was assumed that was radar related.
 
#6 ·
I recall similar issue with Car Alarms on both Honda and Hyundai's when I was on Mt Soledad in San Diego (park at the summit) the CBS and ABC TV affiliates plus microwave towers are all over the area. It wouldn't happen all the time but often enough to be noticed.

Back in the 80s whenever the Navy had gunnery practice on San Clemente Island beach residents reported Car alarms wouldn't unlock or went off and garage doors opening and closing or disabled.

It was assumed that was radar related.
High power radar is a great source of RF interference. I used to work on a research site having radars and could hear the radar sweeps breaking through on FM radio well out of band presumably due to harmonics generated in the radio circuits.