LED headlamps

[jifflemon]

For ever good product out there that I've been able to find, there's 100:1 ratio crap ones. LED's, are absolutely in this category. I've tried literally 100's of them trying to find the elusive good ones. Dashboards bulb, to sidelights, brakelights, headlamps and spotlamps, I've tried them all. Most of them have been hilarious comical from the H1 LED's that were physically too large to actually fit spring to mind, or dashboard LED's that have a lifespan than can measured in seconds.

LED headlamps are an area that trigger many people. Whilst a nice crisp white light is desirable, poorly aligned headlamps or stupid SUV's that have the headlamps physically higher mean getting dazzled is a real risk. Now add into the mix that a even a poorly fitted halogen bulb will throw off a beam pattern and lead to dazzling and we could be running into a minefield here.

Beam pattern? Well, in the UK, the beam pattern looks like this:


The dip beam pattern, at a distance of about 3m, should drop to 5-8 cm below the headlamp centre and slightly left to avoid dazzling oncoming traffic. There's a kick up to the left that illuminates the kerbside of things. For Europeans, this is reserved. You can flatten the kick up (for example, when driving abroad) using tape or beamBenders. That pattern is formed by bouncing the light off the headlamp reflector and through the lens on the headlamp glass itself. Its a delicate science that relies on the light coming from the right place, which is why bulbs have very precise dimensions to ensure the light source is always in the same place (something that some Chinese manufacturers think is optional - More light must be better right?)

So, I finally bit the bullet and bought some non-Chinese bulbs to test. These ones in fact.


First things first, they're not cheap. A generic pair of halogens H4's can be had for about £7. I normally run Osram night breakers, which are around £15 a pair (and a VERY good bulb). These Philips LED's come in around £50.

So, next up, power. LED's have another benefit - they draw less current whilst producing more light. How much more? Well, a H4 bulb pulls around 4.2 amps, the LED around 1.8. Just to put that into some perspective, the LED headlamp is now drawing about the same current as a brake light bulb.

So, science stuff now. Whilst you can align headlamps using nothing more than a flat wall, some of us have a headlamp alignment tool (yeah.... I know!).

So, I installed a single bulb so that I could compare side to side. I'm comparing not only the beam pattern, but the light colour and Lux output. For the latter, focus less on the number, more on the comparison! You'll have to forgive the images, but taking photo's through glass isn't the greatest! Also, don't panic that the beam pattern appears to be LHD - it's being bounced off a mirror!

So, the Halogen bulb.




And then the LED




That, is quite a bonkers difference. It's a much brighter light, with excellent beam pattern and double the light output.

But what about in the real world?

Well, I left a single bulb in - fear not, I drove back in daylight - and waited for night to fall, before sneaking off and finding somewhere I could test.


Again, taking photo's of light doesn't really come out well. However, the output was noticeably better. Very white, very crisp. From a pure aesthetics point of view, as I run LED DRL's, the front end now looks normal, as the light colour of the headlamp matches the DRL!


Would I recommend? Well, I'd beg anyone who's fitting any kind of uprated lamp to get the alignment checked before and AFTER fitting. Your lack of light could simply be that your headlamps are misaligned, dirty or need replacing. Then, there's the cost. Headlamps last a while, and at 3-7 times the price, they'll need to last an awful long time before I break even.

Finally, there's the MOT dilemma, which states vehicles presented with converted halogen headlamp units first used on or after 1 April 1986 will be failed. However, it also states that "If a complete headlamp unit has been replaced with a unit that was manufactured with HID or LED light sources, it must not be failed for ‘Light source and lamp not compatible’ but it must meet all other requirements detailed in section 4 of this manual specific to the type fitted at the time of test." I'll be chatting to a few testers to see how they interpret this, because I could argue that it is a replacement unit, not bulb. They're not permitted to disassemble, so have to take a customers word for it; and because the pattern and alignment is correct, it would be a pass.

[Jay-Kay-Em]

As usual, great write up and the identification of no doubt a great product in a sea of Chinese cr4p.

The MoT thing...

I could argue that it is a replacement unit, not bulb.

I think it would take quite a naïve MoT tester to accept that, given it has a chuffing great H4 on the glass...



The "H" meaning a halogen lamp of course.

Personally, I think MoT failure item "4.1.4(c) Light source and lamp not compatible" is quite robust.

Like you say, its down to the tester on the day.

The death traps I examine that were allegedly MoT'd within weeks proves its all a nonsense anyway!

Philips are quite clever; note the "squiggly" equals sign...



It's a misnomer you can get an LED H4 because, by definition, the H4 regulations apply to halogen incandescent. You are buying something that simply fits a H4 socket.

A "replacement unit" would mean one of the LED dedicated units that have been type approved to be so (allegedly)...



Note the circled "E" at the bottom.

These units, despite being "approved", present their own problems. They look awful and have no serviceable parts - so when they become full of condensation and a single LED fails, your £100 becomes junk.

I have a examined a few now and let them go - getting popular with some MX5 and Defender owners.

I get the argument that if anything passes all items contained within Section 4, then who cares? Whilst that is true, an MoT beam setter is a quite a primitive tool and only tests straight on. It's the un-tested periphery edges of a beam that can cause misery to oncoming motorists which is why the mis-matching of LED in Halogen units needs to be regulated. That, and they are too bright when it all goes wrong.

On the subject of being too bright, note that a static beam height check is on a flat level ground. What about cresting a hill? The poor sod coming up the hill the other way is blinded!

For that reason, I think MoT failure item 4.1.4(c) should be enforced where possible.

You knew all this was coming

[WRDendy]

Like you say, its down to the tester on the day.

I think that's a good way to sum up LED headlights in general. I have LED H7s in the high beam position on a couple of my other cars, but they happen to have separate bulbs for high and low beam. Both of them have gone through many MOTs with that configuration, presumably (and I defer to the wisdom of those here who have actually performed MOT tests) because patterns and cut-offs on high beam are somewhat less of an issue.

I have tried H7s as main beams and was never able to get a good pattern, presumably because it's near enough impossible to replicate a point source illumination like a filament with one or more COB LEDs, which by definition have some kind of thickness. Because of that the geometry is always going to be slightly off as regards reflectors and lenses, as per Jeff's paragraph below the beam pattern diagram.

This gets me wondering - is an H4 actually a better candidate for LED low beam performance because of the filament layout?



We see here the low beam filament behind a reflector, and then below how that translates to high and low beam patterns:





(I borrowed these from some people who have an entire forum dedicated to a specific model of Mitsubishi cars - what a bunch of losers )

We may also consider the following bulbs, which the manufacture boldly claim to be E-marked (i.e. legal):


(link)

Note how the LED is positioned facing into the reflector, which itself then forms a projection system. Based on this I postulate wildly that by the time the light hits the reflector on the back of the lamp housing it's probably behaving in much the same way as it would be had it come from a filament, since the source is a reflector rather than a point.

High beam I imagine will still be a bit off since it relies on a pair of chips either side of a board with a big heatsink, not a point source, but as mentioned earlier that's probably less of an issue as far as beam pattern is concerned.

Of course, none of this gets away from the fact that it's an LED retrofit in a housing originally designed for halogens, so once again

its down to the tester on the day.

With all that being said, many thanks to Jeff not only for taking the risk on some rather expensive bulbs, but for doing a comprehensive test!

[Alan 480]

I admire the thought & effort, but given the limited miles the 480 now covers I'll just stick with the existing bulbs as I think i have only ever replaced ONE headlight bulb over the last ten or fifteen years,

maybe been lucky but it was the 'high bright' DRL that used to get chewed through when ran it every day......

and yes i agree that there are far too many of those SUV with beam set 'too high', maybe the internal thumb wheel that is used to lower for when towing has been turned too far UP? (i'm giving them some lee-way )

[triumphtoledo]

Another issue involves the bulbs themselves. Replacement bulbs must be e-marked. LED conversion bulbs (i.e. those with a halogen base and LED top) cannot e e-marked.

Therefore, they are for off-road use only. Using them on the road renders the car unroadworthy.

I had to do loads of research on this topic for not just car magazines but also a court case.

R