- All Forums
- EV Racing
- Electric Rides
- Electric Technology
- Community Marketplace
- V Forum
- Recent posts
- My ...
- Sign in
Installing LEDs in the Vectrix spotlight
There are currently 0 users online.
The specifications look very confusing to me:
As far as I am aware, Watts = Volts x Amps.
12V x 3.0A (3.2A -0.2A) = 36 Watts
12V x 3.4A (3.2A +0.2A) = 40.8 Watts
So how can they justify a figure of 126 Watts per bulb?
The title states an even higher figure of "2 x 252W" instead of 2 x 126W.
After carefully studying the pictures, it appears that the bulb has two sets of 6 LED chips for main/high beam and a separate set of 6 LED chips located closer to the base for dip/low beam.
This would mean that the main/high beam would draw twice the current of the dip/low beam, so if you're still using the original ICM firmware, it will probably still trigger the flashing bulb warning light.
I also found what appears to be the same bulb units but with additional specification:
Unfortunately there is no mention of how much current is being drawn by each bulb in either high or low.
If it is a genuine Philips bulb, it should be of reasonable quality, but I still don't understand why they would describe it as a 126W bulb when it obviously isn't, or it wouldn't be as stated: "ENERGY SAVING UP TO 85%".
I'm also not convinced that the 12,000rpm cooling fan will manage a ">50000hrs lifetime".
what about the claim of 12600lm/bulb?
I have a bulb with this description and was wondering how they'd compare "80W 8000LM Kit Car CREE LED Headlight Bulbs H4 HB2 9003 H13 9004 9007 Hi/Lo Beam" similar to this http://www.ebay.com.au/itm/80W-8000LM-Kit-Car-CREE-LED-Headlight-Bulbs-H4-HB2-9003-Hi-Lo-Beam-6000K-Pair-/281724136831?hash=item...
Bikemad has already done a fairly good analysis. However, I'll answer your last question.
The absolute best Philips CSP LEDs (assuming their claim of using Philips is even true), is 125 lumens per watt.
12,600 lumens would mean you would need to draw 100w, measured post driver. At 12.8v, this would be approximately 7.9A post driver or 10-16A at the H4 plug. Simply not possible when they say they draw 3.2A draw
Taking measurement the flip side, again, the absolute, absolute best Philips can offer in the most ideal laboratory condition for their CSP chips is 245 lumens per diode. That thing has 18 diodes. Assuming all of them can light at the same time (and we've already seen that some drivers won't drive all of them at the same time), then the absolute best - assuming no shadowing, no under-driving, no defects in the reflector, etc... Is 4410 lumens. This is a very, very, unrealistically optimistic figure.
These guys are so far out of the ball park, none of their claims can be trusted. It's like someone putting up a Honda Cub 50cc, and claiming it can do 0-100 in 0.3 seconds, and the rocker cover is plated in 24 carat gold. You know the first can't be even vaguely true, so why would you believe anything else they say?
Oh. and if you think a 15krpm fan can be "mute", watch this video: https://www.youtube.com/watch?v=KSUoyeVJdMQ
Notice the voltage drop from the current draw on the multimeter at the end.
The seller's a fraud. Stay well clear from him. If you're just after the brightest H4, Cree XHP50 seems to be the only option. Philips Lumileds, CSPs, Seoul Y?? all are just too under powered to pack enough to break much more than 1000 lumens per bulb. There are a few claiming to use Cree XHP70s, but a close up on the photos make it clear that the chips in the photos look nothing like the chips on the Cree website. Again, probably fraud and steer clear. If anyone knows of a genuine XHP70 based H4, please let me know. I'd be keen to try it.
Edit: Missed that last bit of comparison. The other item you linked claims 4 Cree XT-E LEDs per bulb. Each XT-E LED gives out 150-180 lumens. So your "8000 lumen" kit, would put out around 600-700 lumens per globe - only marginally brighter than the OEM 35W globe which should be around 500 lumens. Even the "25,200" lumen kit looks like a floodlamp by comparison :P.
Thanks for that interesting post, Sunder. That clarifies it a lot.
I have to tell you, though, that the 8000 lumen globe I installed is actually significantly brighter than the original halogen globe, regardless of how many lumens it is compared to what they claim.
I'm going to guess one of two things happened then:
1. You went from a warm colour to a cool white (Cool white lights are perceived as brighter at the same lumen)
2. Your halogen was a bit old. So you might have been comparing a bulb that was originally 500 lumens, but at the time of swapping, had faded to 400 lumens... So a jump to 700 would have seemed big.
Hopefully you don't see this as a put down of your globes... If you're happy with it, you don't need anything brighter - in fact, too bright lights are not just bad for other drivers, they can make your peripheral vision worse. I'm just saying that science doesn't lie - if what we're doing here can be called anything approximating science - so much guessing, but since the advertiser's claims are an order of magnitude off, even the guesses can "prove" what they're claiming is not realistic.
hahaha....no, I don't see it as a put down of my globes. It is whiter and it's definitely brighter...but whether that's because the old globe had faded or not, I cannot say.
Incidentally, you hit on something very valid there. I have a car that I installed trucklites on. The beam is so bright and intense that peripheral vision is almost non existent! Pulling into a driveway at night, for example, is actually really difficult as you can't see it because it's 'in the dark', relatively speaking.
Turns out there's two different blink patterns:
1 for under current - It just blinks consistently
1 for over current - it blinks some sort of 2 - 1 - on solid pattern.
On the bright side (no pun intended), that second set of bulbs I bought is the real deal. Draws 4a on low beam and just over 5a on high beam. Unfortunately, that's enough to make the Vectrix complain. Fitting a resistor in series is much harder than fitting in parallel, so I'm not sure what I am going to do.
Well. This is getting more and more fun. I put a load generator (Arachnid Re:load) and found the following:
Low Beam: Blinks under 2.2A and over 3.5A
High Beam: Blinks under 2.2A and over 3.7A
BUT - curiouser and curiouser, making up a resistor that falls exactly into that range seems to still have a warning.
In fact, hooking up the lower current LED, then hooking up the load simulator, will still not get it into a stable state.
I have read that some Canbus computers use high frequency AC to test "resistance" - effectively testing for impedance instead.
I'm going to hook up a capacitor to it next weekend, and see if I can trick it that way.
I'm not sure if other people have gotten lucky with the brands they bought, and they have a low enough impedance that just getting the resistance down to the right level stops the warning, but I can't believe nobody else has had this issue before :(
I suspect most owners will have opted for the ICM firmware cure instead:
The ICM firmware that has been modified by DIY and Framuga also provides additional useful features like instant horn and a basic cruise control (throttle hold) function.
My hesitation to go down that route is that I don't know what state the bike is in at the moment. For example, I know the BMS and the charger are totally aftermarket. Pretty sure I also don't have any cooling fans, or perhaps not temp sensors on the battery. If I write a firmware version that is relying on any of those sensors, I could completely bork the bike. So I am somewhat committed to go the CANbus fooling route.
Of course, right now, I'm going the "Black tape over the flashing light" route. That'll work until my next road-worthy check. I might just have to swap out the light back to a Halogen for the road-worthy inspection, and swap the LEDs back in after.
Try this new ICM SW:
then scroll down and click on VX1 Full Package SW to download zip.
Run the new configurator (Rev2) after installing DE-ICMR6.hex. In the ICM box there are 3 columns of values. The first column is the nominal measured hi/lo beam current (in A/D counts). 2nd column are custom values you can set as the nominal. 3rd column is values the SW is currently using.
So basically, turn on your LED lights and see what the nominal current is. Copy these to the custom column, then set check box. SW will use these new values as nominal and set limits accordingly.
Thanks, but does that firmware assume an otherwise stock Vectrix? I'm not sure it's a risk I want to take. My understanding is that the stock firmware relies on the stock charger being there, as well as temperature sensors on the battery.
Its just ICM SW so it doesnt affect any charging and discharging.