Surely some of you have heard of this weird issue, with the "smoked" tail lights of North America spec Defender X shorting out and turning off without warning (until the vehicle is restarted) when a strong light - such as the headlight of a car following, for example - is shined directly into the light cluster. This has the potential of being a very dangerous manufacturing defect that could cause someone to be rear ended, or worse. There is apparently an upcoming recall, but it seems like Land Rover didn't quite do their homework on which batches of tail lamps may be affected... Watch this video from Simon at powerfuluk, where he not only recreates the malfunction on his bench with scary ease, but by blind luck also finds that Land Rover's recently issued recall is actually incorrect or incomplete as there may be more batches of defective light clusters out there that Land Rover has not identified... which is an even scarier prospect.


Bottom line is, if you have a Defender X in North America (or upgraded the tail lights of any other trims to the smoked / grey ones fitted as standard on the X), do your own tests - as you will see, all you need is a powerful flashlight and some patience - and don't necessarily rely on Land Rover's information for guidance.

 

ill be able to check it today with my escalade with its bright LED bright full beam.

 

I have tried this today on my MY 2021 110x and could not replicate it... maybe just luck ?
However, this is an easily fixable issue. A diode should take care of this without any problems. Now, as an owner, one should not do that, but if there are delays in parts ( which may have to be exchanged ) this may be a temporary solution.

 

Well that's just great!
Expecting delivery of my X in August

 

it should have been fixed by then from the factory.

 

The "fix"" won't come from the factory.
It will go through the dealers, and that is where they will be fixed. Is not just JLR is a common practice with automakers, EVEN when they know there is an issue.
Why ? Cheap.
Do you think they would put those vehicles back on the line, have robots reporgramed, and create a new line just for "fixes"? Never.
To give you an example: Mercedes has completely removed their quality control on OEM parts, 100%. Testing on a bench or some sort of testing or the parts they produce.
Why, because is cheaper for them to have you fix it, or even have it replaced, than for them to test it before it goes out the door. A study showed that they can save as much as 19-21 % in costs by doing so. For such a large company that is a LOT of money. Cost of replacing the bad parts coming back ? Less than 5%.
Back to JLR - for example, mine does not have this issue, although is an X 2021 model. Probably there are a lot of other owners in my situation.
If anyone is technically inclined, can fix this issue for under 2 $, inserting 2 diodes in line, and voila ... no reverse power to the computer.
Is it normal to do that ? Absolutely not. Then again, is a JLR and we would not have it any other way !

 

Not sure what you mean by "...in line...to the computer". As Simon said in the video, the shutdown of the lights is done within the unit itself. So you would have to take apart the whole assembly, which sounds like a PITA to me. Once you have done that, you may still not be able to get to the wires from the top LED cluster or circuitry responsible for shutting down power. I agree that if someone with a Defender X is electrically minded and they're waiting on this fix, it probably is worth investigating. Hopefully the light unit can be broken down far enough to get to everything. It is interesting that the US spec light unit is more sensitive than the European spec unit.

However, there is an easier fix that anyone can do though...

LEDs are light sensitive to any external light source that has a shorter wavelength or equal wavelength to the light it emits. The standard light cluster has a red-tinted plastic cover which blocks colors except red to the LEDs (i.e. yellow, green blue etc) and so when you shine a white light onto it, the amount of energy getting through to the red LEDs is greatly reduced compared to the red LEDs covered by the clear plastic cover.

So, the really easy fix that anyone can try is to cover the light clusters with ruby red lithographers tape. It would look crap, but it might just work!!

 

Here is my thinking ( and it may be flawed without having tested this myself ) ....just as a proof of concept:
- the shut down command of the lamp does not come from the lamp itself ( so I hope )
- if that is the case, a diode placed in line, would prevent a reverse voltage towards one of the controlling modules

I wish Simon was a bit more detailed, but I would love to take apart one of the lights ( just not mine - lol )

 

I don't think it would work. An LED is very inefficient, compared to say a photovoltaic cell, in converting light from an external light source into a potential difference across the LED. If you were to measure the voltage across the LEDs (powered off) when you shine the light, I think it would be very small compared to the normal source voltage when powered on. So I think the light just limits the current passing through the LED rather than actually reversing it. Your in-line diode would not actually see a reverse in current direction. This is why I mention red lithographic tape as a way to reduce the amount of light energy being transmitted through the glass without affecting the red light. I'm not suggesting it is a serious solution of course. But a solution that produces a similar effect could be found--perhaps inserting a filter between the clear light lens and the LEDs. I'm intrigued to see what JLR come up with as a retrofix. Probably depends on how many Defender Xs there are are out there with the problem. the lights are well over $1000 each I think.

It is a fascinating problem. Fortunately, my HSE Defender has the standard light cluster so I'm not going to take it apart.
Someone else can do that.

 

Yes, good point...
Something must have happened at some point into production line or design (?).
Like I said, mine is an X (MY21) but I do not have that issue. It will be a challenge for JLR to get it fixed quick.