I didn't get a response from Niteflux, despite the fact that I know
they looked at the images of the exploded light.
Verbatim letter to them:
Date: Sun, 25 Sep 2005 00:16:14 +1000 (EST)
Hi again,
Lastnight while riding home, the new 5W globe (bought 14th August) in
the same housing and mount as in the fault reported below,
spontaneously exploded. It exploded after going over a small bump
(unevenness in the road, not a tram track or pothole) at the bottom of
Camberwell road, Camberwell (Melbourne), south-east bound. I had
barely noticed the bump, and it was certainly at most only a catalyst
for the destruction of the bulb, and not the cause. I stopped and
checked the bulb before putting in the 15W, and there was no foreign
objects in the bulb or housing. Overheating wouldn't have been an
issue, since the bike was being ridden downhill, and the air
temperature was only 16 degrees. Issues with the battery would be
very unlikely -- there were no further problems upon replacing the 5W
bulb with my spare 15W bulb, and NiMH batteries don't create voltage
surges. Inspection (shown in photographs linked herein) shows that
the bulb disintigrated, taking out the front glass cover with it.
As to the condition of the lights, and whether that may have a bearing
on why the bulb exploded -- I've had the lights since early May. I've
had the 5W on the bike continously, for about 2500km, with about 700km
on the new bulb. That's about 30 hours of road time, and about 13
hours of actual operation. The damage sustained to the 5W housing
during the previous fall only affects the mounting area - there is no
deformity near the bulb.
These two shots illustrate the damage:
http://pizza.homelinux.org/~tconnors/photos/misc/broken_niteflux/dscf0869_Med.jpg.1.html
http://pizza.homelinux.org/~tconnors/photos/misc/broken_niteflux/dscf0887_Med.jpg.9.html
Do you know the likely cause of the disintergration? While the MBTF
(expected life-time) of the bulb is not in the hundreds of hours, 13
hours is a very short time (compared to the figure I remember reading
somewhere on niteflux.com.au), and I've never encountered such a
drastic failure mode -- disintegration of the bulb and protective
covering. It's quite dissapointing, given the cost to replace these
bulbs!
--
Adrian Konings said:Hi Tim
Thanks for your comments and support of the NiteFlux product.
We did do extensive testing of the bracket over 8 month period before
production never encounter this problem, the only time we could get it to
fail in this way was when it was only half done up.The rubber griping pad at the bottom of the half cup part of the bracket
prevents the bracket from slipping if the bracket is correctly tensioned.We take all customer comments under consideration and if we can improve the
product we will.If you continue to have loosening problems let us know and we will look at
the unit and see if there is a problem with it.Best Regards Adrian.Konings
Product Manager
The Bicycle Authority Pty Ltd-----Original Message-----
From: Customer Self Service [[email hidden]]
Sent: Monday, 15 August 2005 10:27 AM
To: [email hidden]
Subject: Web Site Contact FormWeb Site Contact Form - Other
User Name:
Tim ConnorsUser e-mail Address:
[email protected]Company Name:
Message:
I have a niteflux 5/15 system which I love, but I would like to suggest an
addition to any new light systems manufactured: A locking mechanism on the
mount.I left very early yesterday morning, with the light securely in place. After
a couple of hours of riding 40km/h on a rough road, the vibrations managed
to work the light loose, and it met the tarmac at 50 km/h. Naturally, the
globe was blown, but everything else remained intact. This has happened to
me once before, and a friend of mine in the touring club once, and we've
both only had our lights for 1-3 months. The globes are costly to replace
before their time ought to be up, and I don't really want this to happen too
often!POWERED by Customer Self Service, a Commerce Vision Pty. Ltd.
<http://www.commercevision.com.au> product.
--
TimC
Perl 6 will give you the big knob. -- Larry Wall
--
TimC
Error: Furry Pointer Exception