dynamo light cutout

james said:

I have new shimano nexus dynahub with the switch unit that comes with
it and a B&M lumotec + with standlight
(http://www.sjscycles.com/store/item544.htm)and d toplight + rear
(http://www.sjscycles.com/store/item554.htm). The lights are wired in
parallel using heavier-than-usual-for-dynamos speaker wire with proper
soldered ends and spades etc. The light cuts out after anything from
a few seconds to a few minutes. The standlights remain on (if they
had time to charge up). If I stop the front wheel for a few seconds
and then set off again the lights work again. I cannot determine if
the length of time they work for correlates with anything (such as
length of time that the dynamo stops for) or is random. The dynamo
needs to stop for several seconds (eg braking and lifting the front
wheel whilst in motion doesn't do it) It seems as though slowing to
below the speed required to fire the light up (ie <2mph) does the same
thing but I cannot be certain of this. Switching the switch off and
on (without stopping the front wheel doesn't seem to do anything. I
*think* the key is stopping the elctrical output from the dynamo (as
opposed to some sort of mehanical effect from stoppingthe wheel) which
leads me to suspect something to do with the zener diode protection or
somesuch. IRRC both the switch unit and headlamp have zener diodes.
I have never knowingly shorted any terminals. I have had the unit up
to 40mph (although not with lights on) but dont often go much above 25
mph (peak speeds). Can anyone suggest a likely hypothesis for these
syptoms or has anyone had similar eperiences. ISTR Guy has bad words
for B&M lamps but that was mostly to do with terminal connections and
I don't *think* that is the issue here

I wonder if the xener diode has a bad connection. They get hot in use
and the connection could be breaking. Could you try another lamp?

james said:

I have new shimano nexus dynahub with the switch unit that comes with
it and a B&M lumotec + with standlight

I have the version of the light without a standlight.

Quoted message said:

(http://www.sjscycles.com/store/item544.htm)and d toplight + rear
(http://www.sjscycles.com/store/item554.htm). The lights are wired in
parallel using heavier-than-usual-for-dynamos speaker wire with proper
soldered ends and spades etc. The light cuts out after anything from
a few seconds to a few minutes.

Could the effect be related to heating? For example, if the light went out
after you were going fast and you only stop for a short time, it may only
work for 2 seconds. But if the light goes out whilst you are going slowly,
then you stop for several minutes - it will work for several more minutes
before going out again?

Quoted message said:

The standlights remain on (if they had time to charge up).

They are probably a separate circuit (powered by the dynamo, of course) so
I doubt they can tell us that much about how the rest of the light works.

Quoted message said:

If I stop the front wheel for a few seconds
and then set off again the lights work again. I cannot determine if
the length of time they work for correlates with anything (such as
length of time that the dynamo stops for) or is random. The dynamo
needs to stop for several seconds (eg braking and lifting the front
wheel whilst in motion doesn't do it) It seems as though slowing to
below the speed required to fire the light up (ie <2mph) does the same
thing but I cannot be certain of this. Switching the switch off and
on (without stopping the front wheel doesn't seem to do anything. I
*think* the key is stopping the elctrical output from the dynamo (as
opposed to some sort of mehanical effect from stoppingthe wheel) which
leads me to suspect something to do with the zener diode protection or
somesuch.

Could be...

Quoted message said:

IRRC both the switch unit and headlamp have zener diodes.

Quoted message said:

I have never knowingly shorted any terminals.

If you had, this would probably damage the dynamo rather than the lamp.

Quoted message said:

I have had the unit up to 40mph (although not with lights on) but dont
often go much above 25 mph (peak speeds).

At least one dynamo - the expensive and sophisticated Lightspin - is
destroyed whenever the rider exceeds a certain, rather low speed (3X mph
IIRAC). Those continentals don't have steep hills like we do, so their
bike lights aren't always designed to withstand the speeds that we can attain.

Open-circuit dynamos can produce quite high voltages at these speeds, and I
wouldn't be surprised if they could roast a small varistor or zener diode
placed accross its output. You might put such a component in front of the
switch (i.e. not turned off when the light is turned off) in order to
protect it from the sparks the dynamo would otherwise make if you tried to switch
off the lamp whilst you were moving.

Quoted message said:

Can anyone suggest a likely hypothesis for these
syptoms or has anyone had similar eperiences.

If I short out my Schmidt dynohub, it starts to produce a moderate amount
of resistance as I try to turn the wheel - it's a small recumbent wheel so
this may be easier to detect. On the other hand, if it's open circuit then
it just keeps spinning.

I would suggest you put your bike upside down, and spin the wheel until
the lamp goes out. If the wheel suddenly starts to slow down rapidly,
this would suggest that the problem was a closed-circuit fault - perhaps a
failed Zener diode. On the other hand, if the wheel just keeps spinning
then you probably have a good old fashioned loose connection.

AC

james said:

I have new shimano nexus dynahub with the switch unit that comes with
it and a B&M lumotec + with standlight
[...] The light cuts out after anything from
a few seconds to a few minutes. The standlights remain on (if they
had time to charge up). If I stop the front wheel for a few seconds
and then set off again the lights work again.

Round Lumotec? This is normal :-(

You need to spring the internal contacts a bit, and make sure you don't
adjust the light up and down by pushing on the reflector. The oval Lumotec
is somewhat better, but will still do this. The SON E6 does not do this,
but has no standlight.

--
Guy
--
May contain traces of irony. Contents liable to settle after posting.
http://www.chapmancentral.co.uk

Just zis Guy said:

james said:

I have new shimano nexus dynahub with the switch unit that comes with
it and a B&M lumotec + with standlight
[...] The light cuts out after anything from
a few seconds to a few minutes. The standlights remain on (if they
had time to charge up). If I stop the front wheel for a few seconds
and then set off again the lights work again.

Round Lumotec? This is normal :-(

You need to spring the internal contacts a bit, and make sure you don't
adjust the light up and down by pushing on the reflector. The oval Lumotec
is somewhat better, but will still do this. The SON E6 does not do this,
but has no standlight.

The switching off is caused by the shimano switch which is brimm filled
with clever electronics you could do without. It probably detects a
loose contact and manages to turn this into a real problem, hiding the
cause in the process
--
---
Marten Gerritsen

INFOapestaartjeM-GINEERINGpuntNL
www.m-gineering.nl

anonymous coward <[email hidden]> wrote in message news:<[email hidden]>...

Quoted message said:

At least one dynamo - the expensive and sophisticated Lightspin - is
destroyed whenever the rider exceeds a certain, rather low speed (3X mph
IIRAC). Those continentals don't have steep hills like we do, so their
bike lights aren't always designed to withstand the speeds that we can attain.

I bug to defer: I max'd out last week at 81kph going down a
continontil hill, with Nexus + BuM (mit Standlicht) blazing away
happily - my commute in Luxembourg has 1000m of ups&downs, and the
dynamo is always on, I've changed the bulb once in
the last year.

- Rory

m-gineering said:

The switching off is caused by the shimano switch which is brimm
filled with clever electronics you could do without. It probably
detects a loose contact and manages to turn this into a real problem,
hiding the cause in the process

It also happens on my SON-equipped recumbent. Honest.

Either way, the problem is in the lamp -or at least it is if my experience
is anything to go by.

--
Guy
--
May contain traces of irony. Contents liable to settle after posting.
http://www.chapmancentral.co.uk

various people wrote various things...

I don't think theissue is overheating as there doesn't seem to be any
correlation between speed, time to cutout, time paused (as long as it
is longer than about a second) and time to next cutout following the
pause. After spinning teh front wheel by hand until the light cut out
the resistance to turning *seemed* to reduce suggesting an open
circuit (I say seemed, becuase the difference in resistance is most
easily observed as the wheel stops but this is also about the time
that the light comes back to life. I will investigate the contacts in
the front lamp and try swapping it out (although my other lamp has a
light sensitive switch in the lamp head so that might comlicate
things). MG, do you have experience of teh switch unit being a common
failure point with shimano units? I might also see if I can bypass
the switch altogether

thanks everyone
james

"Just zis Guy, you know?" <[email hidden]> wrote in message news:<[email hidden]>...

Quoted message said:

m-gineering said:

The switching off is caused by the shimano switch which is brimm
filled with clever electronics you could do without>

It also happens on my SON-equipped recumbent. Honest.
Either way, the problem is in the lamp... [in my] ...experience

I sprung the centre contact forwards this morning and managed 12 laps
of the garage without a cutout. It isn't conclusive (I will run the
light home tonight) but it looks hopefull although I am still puzzled
by the apparent pattern I identified earlier

best wishes
james

in message <[email hidden]>, james

(') said:

various people wrote various things...

I don't think theissue is overheating as there doesn't seem to be any
correlation between speed, time to cutout, time paused (as long as it
is longer than about a second) and time to next cutout following the
pause. After spinning teh front wheel by hand until the light cut out
the resistance to turning *seemed* to reduce suggesting an open
circuit (I say seemed, becuase the difference in resistance is most
easily observed as the wheel stops but this is also about the time
that the light comes back to life. I will investigate the contacts in
the front lamp and try swapping it out (although my other lamp has a
light sensitive switch in the lamp head so that might comlicate
things). MG, do you have experience of teh switch unit being a common
failure point with shimano units? I might also see if I can bypass
the switch altogether

I know this is awfully scientific, but have you put a multimeter across
the bulb? What happened to the voltage? What happened to the
resistance? FWIW mys experience of the Shimano magic switch is very
good.

--
[email hidden] (Simon Brooke) http://www.jasmine.org.uk/~simon/

;; Perl ... is the Brittney Spears of programming - easily accessible
;; but, in the final analysis, empty of any significant thought
;; Frank Adrian on Slashdot, 21st July 2003

Simon Brooke said:

in message <[email hidden]>, james

(') said:

various people wrote various things...

I don't think theissue is overheating as there doesn't seem to be any
correlation between speed, time to cutout, time paused (as long as it
is longer than about a second) and time to next cutout following the
pause. After spinning teh front wheel by hand until the light cut out
the resistance to turning *seemed* to reduce suggesting an open
circuit (I say seemed, becuase the difference in resistance is most
easily observed as the wheel stops but this is also about the time
that the light comes back to life. I will investigate the contacts in
the front lamp and try swapping it out (although my other lamp has a
light sensitive switch in the lamp head so that might comlicate
things). MG, do you have experience of teh switch unit being a common
failure point with shimano units? I might also see if I can bypass
the switch altogether

I know this is awfully scientific, but have you put a multimeter across
the bulb? What happened to the voltage? What happened to the
resistance? FWIW mys experience of the Shimano magic switch is very
good.

From the bible on dynamo's (beleuchtung.pdf @ enhydralutris.de):

An einem gebrauchten SW-NX30 gemessen ergeben sich folgende Erkenntnisse:
• An einem Frequenzgenerator mit Verstärker (kein Fahrraddynamo!) bricht
ab einer Eingangsspannung von 6,99 Veff (bei 50–1000 Hz) die
Ausgangsspannung, dann 6,22 Veff an 12 , zusammen und erholt sich erst,
nachdem die Eingangsspannung über ca. 1 s fast auf 0 V reduziert wird.

ie after an overvoltagesituation you have to stop for a second before
power can be resumed
/MArten

Quoted message said:

From the bible on dynamo's (beleuchtung.pdf @ enhydralutris.de):

An einem gebrauchten SW-NX30 gemessen ergeben sich folgende Erkenntnisse:
An einem Frequenzgenerator mit Verstarker (kein Fahrraddynamo!)
bricht ab einer Eingangsspannung von 6,99 Veff (bei 50–1000 Hz) die
Ausgangsspannung, dann 6,22 Veff an 12 , zusammen und erholt sich erst,
nachdem die Eingangsspannung uber ca. 1 s fast auf 0 V reduziert wird.

ie after an overvoltage situation you have to stop for a second before
power can be resumed

I've been pretending to myself I can make sense of that, but who am I
kidding?

Why should this be?

AC

Quoted message said:

/MArten

anonymous coward <[email hidden]> wrote in message news:<[email hidden]>...

Quoted message said:

I've been pretending to myself I can make sense of that, but who am I
kidding?

Me too - but Babelfish did its best with :

Based on a used SW-NX30 the following realizations result: ? At a
frequency generator with amplifier (no bicycle direct current
generator!) if the output voltage breaks starting from an input
voltage of 6,99 Veff (with 50-1000 cycles per second), then 6.22 Veff
at 12, together and recovers only, after the input voltage over
approx.. 1 s is reduced nearly to 0 V.

Quoted message said:

Why should this be?

'Cos the cunning people at Shimano designed it so !

I run my Lumotec from the NX30 without the switch (and this means the
light is on all the time.) It also means that when the bulb goes I
probably put rather large amounts of power through the poor little
Zener diode that is really only there to stop me overvolting the bulb
at high speeds - and if I do it too long the diode is fried.

I guess that the Shimano people try to cope with this situation by
putting in circuitry/software so that the generator won't do anything
if abused !

I'd welcome a better explanation .

Tim

[email hidden] (Tim Henderson) wrote in message news:<[email hidden]>...

Quoted message said:

anonymous coward <[email hidden]> wrote in message news:<[email hidden]>...
and recovers only, after the input voltage over
approx.. 1 s is reduced nearly to 0 V.

Quoted message said:

Why should this be?

'Cos the cunning people at Shimano designed it so !

That makes sense then because I determined that I could "recover" from
the (presumed) overvoltage event by slowing down almost to a
standstill but without actually stopping.

Fiddling the contacts in the front lamp turned out to make no
difference. Swapping the front lamp out also made no difference.
Bypassing the switch unit altogether seems to resolve the issue. So I
suppose the question is does that mean that I am risking frying the
bulb? The cutout still happens at low (10-11mph) speeds so is the
switch circuitry being bit previous or is the generator outputting to
high a voltage? I haven't worked out how to get a multimeter
attached, rotate the wheel fast enough for long enough and still be
able to monitor values without something getting eaten by the spokes.
Since the unit is fairly new I guess I will try talking to sjsc and
see how they feel about a swapout of the switch. I would be happy to
move a to a h'bar mounted dumb switch but I quite like the idea of
more sophisticated overvoltage protection (provided it actually
works...)

best wishes
james

[email hidden] (james) wrote in message news:<[email hidden]>...
The cutout still happens at low (10-11mph) speeds so is the

Quoted message said:

switch circuitry being bit previous or is the generator outputting to
high a voltage?

As I understand it, the voltage is just a function of the speed and
the load attached to it. I would be surprised if the generator
malfunctions to produce a greater voltage than design. But if you put
two 6v 3W bulbs in series , at a decent pedalling speed loads of
people seem to get 12v to power both up satisfactorily.
If you were getting overvoltage at low speeds, are you sure that you
have the right bulb in your lamp ? Have you tried another ?

I haven't worked out how to get a multimeter

Quoted message said:

attached, rotate the wheel fast enough for long enough and still be
able to monitor values without something getting eaten by the spokes.

I was recommended to use a power drill and gently bring the chuck in
contact with the tyre to spin it.

Quoted message said:

Since the unit is fairly new I guess I will try talking to sjsc and
see how they feel about a swapout of the switch. I would be happy to
move a to a h'bar mounted dumb switch but I quite like the idea of
more sophisticated overvoltage protection (provided it actually
works...)

Unsophisticated beefy back to back Zeners should do - though I
splashed out on the Reflectalite packaged dynamo regulator (which is
just the same thing in a little resin filled box). Unfortunately when
I swapped from bottle dynamo to NX30, I didn't put it back in the
circuit - so when the bulb went I suspect the little Zener in the
Lumotec did not like it .

Regards,
Tim