7" retrofit (GDAA WRX)

[csjoh]

I think the amber one you've got there should be more than bright enough for the purpose.

[satrya]

I think the amber one you've got there should be more than bright enough for the purpose.

I hope so. It will be for an auxiliary turn signal anyway, so I suppose as long as the intensity is within the allowed range, it doesn't hurt to have it.

[satrya]

Testing amber LED

Single 3W amber LED behind each aux turn signal units lit up. There is ambient daytime light from a nearby window to the right of the picture.


Window blinds closed.


Just like in the white LED case, the additional diffuser helps make the 1" LED look like a bigger light source, on par with the ES300 & Micro DE.

[satrya]

Test amber LED at different angles

Testing the amber LED (for aux turn signal) at different angles. The upper pictures are taken with the camera setting at auto, and the lower pictures are at an underexposed setting (1/200 sec, f/20, ISO200). The underexposed picture help differentiate the reduction in intensity at the more extreme angles. Angles described below are from the vantage point of the light source, just like the headlight output photometric testing convention. The angles are approximate; I didn't measure them precisely.

Center & ~5 degrees up ( V, 5 U ). The 5 degree angle is to account for the fact that the housing will be placed very low on the ground.


At various angles towards the right side of the vehicle, right to left: 12:30 o'clock position ( 5U 10-15R ), 1 o'clock position ( 5U 25-30R ), 1:30 o'clock position( 5U 40-45R )


From the left side of the vehicle, left to right: 11:30 o'clock ( 5U 10-15L ), 11 o'clock ( 5U 25-30L ), 10:30 o'clock ( 5U 40-45L )


At the most extreme right & left angles tested ( 40-45R & 40-45L ), the LED from the opposite side of the vehicle would have been obstructed by the bumper cover &/ fog trim cover, so in effect, only the unit closest to that side (i.e. left unit when seen from 40-45L ) will be seen as emitting the light.

For a turn signal, this should be ok. For a hazard signal, it might look like a turn signal instead. I think that should be ok, as oem turn/hazard signals have that ambiguity as well.

[Fstrsn]

Yup, definitely liking the amber more and more.

As far as visibility, I think you'll be fine. These are auxiliary after all.

[csjoh]

How does it look with the yellow fogs lit up as well?

[satrya]

Yup, definitely liking the amber more and more.

As far as visibility, I think you'll be fine. These are auxiliary after all.

Perhaps.

• 2 pairs of biXenons (EvoX-R & Mini D2S 2.0, with 3 way switch): check
• 2 pairs of turn signals (incandescent amber in the headlights, LED amber in the "fogs"): check
• 2 pairs of halogen fogs (~3000K ES300 & ~3700K Micro DE, with independent switches): check
• 2 pairs of DRL (LED in the "fogs" and ... ): uh oh; only 1 pair!

How does it look with the yellow fogs lit up as well?

Good question. I'll have to figure out a bench test method for that. I just re soldered the amber LED into the wiring + connector designated for it. Now, all 4 units (aux turn, DRL, ES300, & Micro DE) don't have exposed leads I can easily connect to a power source. The respective pairs of the connectors are already wired in my vehicle. I may need to remove the headlights to temporarily pull some of the wiring out.

[satrya]

Comparing amber LEDs with a 3W vs 9W claim

I got a hold of another set of amber LEDs with a 9W claim instead of a 3W claim. The larger one below is the old one, at 3W, which has identical packaging as white LEDs I compared a few posts up. The smaller one is rated (actually, "claimed", since I'm not verifying the current draw for any of these) 9W. The older (& larger) 3W version I got either has a more translucent phosphor layer or has a thinner phosphor layer, in the sense that I can spot the 3 blue led emitter dots. In contrast, the 3 blue led dots are not as easily visible on the newer ( & smaller) 9W version.


Top view of the newer & smaller (9W) on the left, and the older & larger (3W) on the right. The thickness of the black enclosure is about the same, and the diameter of the threaded shaft of the new one is a bit thicker by about 1/32 inch. The new one uses a higher thread pitch, which is good I suppose.


One concern I have is that even if the new & smaller one has the same power rating as the larger one, the smaller amount of metal means there may be a higher heat concentration on the LED board. If the new one is 3 times the old one as claimed, then this heat issue might reduce the long term reliability of the LED.

[satrya]

Head-to-head test, "3W" vs "9W" amber LEDs

The smaller sized, "9W" on the left, the larger sized, "3W" on the right. The pictures below are taken at a progressively underexposed camera settings.

At 1/100sec, f/6.3, ISO200


At 1/200sec, f/22, ISO200


At 1/4000sec, f/22, ISO200


The last picture reveals the LED dots at the center. The "9W" is more noticeable from farther distances.

[Fstrsn]

But will the 9w be easily transferrable to the current assembly or will it require you to recreate that assembly with the new smaller diameter?

[satrya]

But will the 9w be easily transferrable to the current assembly or will it require you to recreate that assembly with the new smaller diameter?

Yes. A very good question indeed. I will have to enlarge the mounting hole to account for the slightly larger diameter. The hole in question is buried deep in the frame of the housing. Given the additional heat stress of the "9W", and given that the "3W" seem a bit brighter than the main turn signal bulbs already, I'm not inclined to change anything.

I suppose I posted the pictures above for whoever else that might run into these 2 LED variants and might consider which one to use. If I want to use the "9W", I would mount it in a good amount of properly vented metal to increase both the convective cooling and the thermal capacitance of the area near the LED board.

The difference in intensity is quite noticeable.

[Fstrsn]

Yes. A very good question indeed. I will have to enlarge the mounting hole to account for the slightly larger diameter. The hole in question is buried deep in the frame of the housing. Given the additional heat stress of the "9W", and given that the "3W" seem a bit brighter than the main turn signal bulbs already, I'm not inclined to change anything.

I suppose I posted the pictures above for whoever else that might run into these 2 LED variants and might consider which one to use. If I want to use the "9W", I would mount it in a good amount of properly vented metal to increase both the convective cooling and the thermal capacitance of the area near the LED board.

The difference in intensity is quite noticeable.

Mmmm, given how much brighter it is I thought you were going to explore switching over to the 9W, but if the 3W is already better than the OE unit, and it's not going to be a "painless" process I'd leave it also.

[satrya]

Mmmm, given how much brighter it is I thought you were going to explore switching over to the 9W, but if the 3W is already better than the OE unit, and it's not going to be a "painless" process I'd leave it also.

Indeed.

If the duration of this thread is any indication, I'm not in a hurry to finish the retrofit just for the sake of the result. That's why even though I recently made the switch from the 3W white LED to the 3W amber LED (of identical construction), I still bought and compared the 9W amber LED I ran across.

But I have to draw the line somewhere. I might use it for my aux rear facing turn signal (on the rear wing riser); it currently uses old technology LEDs (amber emitter instead of blue emitter with phosphor-based amber colorshift) and is not visible in daytime at all:

[TRSAndrew]

This is the craziest fog light retrofit ever

[csjoh]

Indeed