Originally Posted by HardyWeinberg
Does your cost estimate include battery and charger? I'm thinking it doesn't, and that's steering me toward the cygo dual cross or cateye double/triple-shot, which might be bright enough (I am using 15/20W halo), and don't require me to develop engineering skills. Much as I would like to (I keep getting stuck on the concept of a waterproof container for the LEDs if I were to buy my own, it's really wet here).
According to the review, the wilma6's setting w/ 42 hr battery life is similar to 10 or 15W halo (it's disconcerting to me that the reviewer lumps those 2 together, their difference is apparent to me, an otherwise uneducated buffoon), so I am curious why other LED lights (those cygo and cateyes) don't claim 40 hr battery life at 10/15W equivalent output levels. Is the wilma6 really that much more efficiently designed?
Yes, my cost includes the battery and charger, and it's not an estimate it's what I'm using right now.
You make a good point on the engineering skills, I'm fortunate that I'm qualified in electrical engineering and have access to a good array of r&d tools.
I glad you brought up efficiency because that's what its all about. The efficiency is primarily determined by what LED the manufacturer uses. The LED has to convert electrical energy (watts) into light (lumens) the more lumens it can make out of each watt the more efficient the light. As the LED's P-N junction gets hotter it loses efficiency at a fast rate so cooling of the LED is absolutely critical. In my case I used the CREE XR-E Q4 bin, 88 lumens per watt @350ma (Tj=25degC) rated up to 3 watts, which translates to just about the brightest and most efficient LED on the market today.
I'm not sure which LED the Wilam uses but it's physically impossible for it to be much more efficient than my light. In fact, to get over 800 lumens from 4 LEDs it has to be running each LED in a less efficient operating region, remember that LEDs are more efficient at lower wattages. I'm producing ~450 lumens from 3 LED's whereas the wilma makes ~850 from 4 LEDs.
I'm driving my LEDs with a 1,000ma BuckPuck with infinitely adjustable brightness and over 97% circuit efficiency (I confirmed it with some bench testing) its impossible for the wilma to be much more efficient with commercially available circuit components, they may possibly be working with NASA and sharing communications satellite technology but I somehow doubt it
The Wilma's lenses are polycarbonate and a proprietary design, I'm also using polycarbonate lenses that are made by a company that specialises in polycarbonate lenses. The material has certain physical properties including light conduction (over 90% in a good lens) so the wilmas lense can't phisically be more efficient than that, and in fact without independent testing I suspect it may be a little more lossy than a specialist collimator lens like mine. There are also a lot of benefits to using an individual lens on each LED. I have two 9-degree spots and a 15-degree oval pattern. I might experimentwith other combinatins at some point.
The wilma has a bunch of "microprocessor controlled" features, which is great if that's what you're looking for. I have a number of RISC micro controller units, any one of which I could program with the exact same feature set as the wilma, the chips cost a few dollars each and are available everywhere. They're blindingly fast, extremely efficient, very easy to program, and complete overkill for controlling a bicycle light. They have RAM, ROM, EEPROM, an ADC, a CPU, and more buffered I/O pins than you can poke a soldering iron at
all in one tiny little CMOS package, they can be put into sleep/standby mode in software and their power consumption drops to almost nothing. All they needs is a crystal and a +5vDC regulator chip. I chose not to implement it because I have absolutely no need or desire for the bells & whistles. I did take advantage of the buckpuck's dimmer and use a switch for a low power setting (about 3 watts, approx 15 hours runtime), but I've never actually used it on the road yet.
I haven't even begun to cover the wilma's potential lack of efficient cooling for the LEDs. I know exactly how efficient my design is because I tested it, I see no data available on the wilma so I can't comment on it but that housing is certainly not designed for efficient heat dissipation
from the P-N junction where it really counts. Furtunately I didn't have to concern myself with the marketability aesthetics.
Please don't misunderstand, I'm not trashing the wilma, not at all. Overall it's a good light, just way overpriced in my opinion. If I couldn't design & build my own and the price wasn't an issue I'd buy it. I'm just in a position where I can build something which is functionally equivalent (or better in some aspecs) at a fraction of the cost.
Originally Posted by 2manybikes
This is for those who say they can build a light as bright and cheaper than the Lupine Wilma.
Consider all these features. (from Lupine site). You get what you pay for.
Don't be fooled by marketing hype. Some of those claims are unsubstantiated, and worthless without numbers from independent testing. The rest are dead easy to implement and in no way justify the high cost of the unit. I'm not talking about the quality of the product, just the marketing hype.