(or anyone else familiar with IC circuits)

The LED flasher in my homemade MOB pole was scavenged from a bike light.
See first two pictures he

http://home.maine.rr.com/rlma/MOB.htm

I test it from time to time and it has started coming on as a steady light.
Sometime it will start flashing after about a minute. Usually if I turn it
off and back on again with the magnetic reed switch, it will start flashing.
Once flashing, it works properly until the next time.

Since these are LED's and the circuit board appeared to be all IC's, I don't
think this is the delay you sometimes see when mechanical flashers take a
few seconds to warm up.

The pole is probably just about as good with a steady light but I wouldn't
want a recovery confused by the light switching from steady to flashing. Of
more concern is the possibility that the circuit board is failing and the
light might not go on at all.

Any idea what could make a circuit like this act weird in this inconsistent
way? The reed switch is well separated from the circuit board so that isn't
a factor.

No chance of replacing these components. They were all embalmed in epoxy to
be sure water could never get in. I'll have to build a new pole section
next winter.

--
Roger Long

Reed switches often have short unhappy lives

Brian Whatcott Altus OK

On Sun, 5 Aug 2007 20:38:47 -0400, "Roger Long"
wrote:

(or anyone else familiar with IC circuits)

The LED flasher in my homemade MOB pole was scavenged from a bike light.
See first two pictures he

http://home.maine.rr.com/rlma/MOB.htm

I test it from time to time and it has started coming on as a steady light.
Sometime it will start flashing after about a minute. Usually if I turn it
off and back on again with the magnetic reed switch, it will start flashing.
Once flashing, it works properly until the next time.

Since these are LED's and the circuit board appeared to be all IC's, I don't
think this is the delay you sometimes see when mechanical flashers take a
few seconds to warm up.

The pole is probably just about as good with a steady light but I wouldn't
want a recovery confused by the light switching from steady to flashing. Of
more concern is the possibility that the circuit board is failing and the
light might not go on at all.

Any idea what could make a circuit like this act weird in this inconsistent
way? The reed switch is well separated from the circuit board so that isn't
a factor.

No chance of replacing these components. They were all embalmed in epoxy to
be sure water could never get in. I'll have to build a new pole section
next winter.

"Brian Whatcott" wrote

Reed switches often have short unhappy lives

Good to know but the switch works fine. It's what happens next that is odd.

--
Roger Long

"Roger Long" wrote in news:46b66da5$0$8977
:

http://home.maine.rr.com/rlma/MOB.htm

"I also found this nifty little light. Press or hold the switch and it
strobes. Press it again and it can be used as a steady flashlight. Twelve
bucks. The tiny little module (see insert) will weigh almost nothing on
the pole and is all ready to be wired to batteries in the bottom and a
switch. The light module weights less than the quarter next to it and is
bright enough to light up a room."

I copied this paragraph from your webpage......

Notice how you have to HOLD the switch CLOSED to get it to go into
strobing mode? What happens when the switch gets a little old and
holding it down results in a noisy connection is the IC keeps seeing
intermittent presses, not a steady-state hold down for X seconds. One
little intermittent interruption when your finger moves is enough to
reset the timer that's trying to time down into strobe mode. It's a
stupid design with such cheap switches.

Clip a jumper wire across the reed switch and see if it doesn't start
strobing reliably. If so, it's the reed, even though it will read fine
on an ohmmeter which cannot see the pulsing probably going on.

I bought a whole bunch of these little 3 really bright yellow LED
"Emergency Flashers" from a flea market dealer. He wanted $1 each, but
sold me the whole box for $10, what was left. I've given all of them
away except what I wanted to keep. The button in it is unreliable in
this same way. It's a rubber switch pressing on the circuit board pads.
Its sequence is a little different than yours. One press gets a "steady
on" light, which is an optical illusion because the LEDs are actually
switching on and off at a very fast rate, 50% duty cycle, to save
batteries. Press it again and the flash rate goes to about 2 per second
at full power. Press it again and SOMETIMES the damned things will turn
off....SOMETIMES...(c;

The correct term for your observation is called "contact bounce" and has
plagued data engineers since before ICs were invented. Most ICs that are
forced to interface with humans have a timer circuit at each interface
point that takes input, then refuses to take more input until the timer
has timed out, several milliseconds, to keep the cheap, noisy switch from
giving it 500 button pushes with each finger push. It's called a
"debouncer", obviously. Your little light's debouncer isn't timed long
enough, now that the switch is crap.

ICs are way too fast for their own good....(c;



Larry
--
Democrats are raising taxes on oil companies by $16,000,000,000.
Oil companies don't pay taxes, just like every other company.
Consumers pay all taxes, corporate and individual.
What's the price of a gallon of regular going to go to to pay $16B more?

Did you try replacing the batteries? Initially cold and low voltage, flow a
bit of current and they get a wee bit warmer and a wee bit higher voltage.
Just a thought.
And why smoke alarms always give low voltage warnings in the early
morinings.



"Roger Long" wrote in message
...
(or anyone else familiar with IC circuits)

The LED flasher in my homemade MOB pole was scavenged from a bike light.
See first two pictures he

http://home.maine.rr.com/rlma/MOB.htm

I test it from time to time and it has started coming on as a steady
light. Sometime it will start flashing after about a minute. Usually if I
turn it off and back on again with the magnetic reed switch, it will start
flashing. Once flashing, it works properly until the next time.

Since these are LED's and the circuit board appeared to be all IC's, I
don't think this is the delay you sometimes see when mechanical flashers
take a few seconds to warm up.

The pole is probably just about as good with a steady light but I wouldn't
want a recovery confused by the light switching from steady to flashing.
Of more concern is the possibility that the circuit board is failing and
the light might not go on at all.

Any idea what could make a circuit like this act weird in this
inconsistent way? The reed switch is well separated from the circuit
board so that isn't a factor.

No chance of replacing these components. They were all embalmed in epoxy
to be sure water could never get in. I'll have to build a new pole
section next winter.

--
Roger Long

Thanks Larry. That makes sense.

I discovered when I put the thing together that just turning power on and
off to the circuit board would result in its always going into flash mode so
I just cut the wire to the original switch. A chattering reed switch could
certainly be doing this though.

Brian's observation about reed switches jibes with my other reed switch
experience. I installed an AquaLarm raw water sensor in the boat three
years ago. All was fine the first season and it did save me once from
overheating my exhaust hose and or engine. Last year, however, it began
sounding every time I started the engine and not going off until I went
below and moved the adjustment back and forth or banged on it. Big pain.
It's been better this year but is starting to do it again. This unit has a
magnetic reed switch in it.

Too bad. The ideas of a switch hermetically sealed from any moisture is
conceptually attractive but I'm not sure thay belong on boats in critical
applications any more.

--
Roger Long

"Roger Long" wrote in
:

Too bad. The ideas of a switch hermetically sealed from any moisture
is conceptually attractive but I'm not sure thay belong on boats in
critical applications any more.



Other than it all being a cheap piece of boat crap, that aside, reed
switches have, after being left on (magnetized) for a time, the
perpensity to become magnetic, themselves. The reeds are not supposed to
be magnetized, but they get that way. Once this happens, they become
either always on by their own magnetism, or so sensitive any magnetic
field stronger than the background earth magnetism will switch them on,
rendering them pretty much useless.

The other problem is the glass case. You wouldn't think it could bend,
being brittle glass, but it can! It only has to flex a tiny bit to make
the reeds touch "on" again. Sometimes the heat on one side of the glass
is more than the other and it'll bend them on without breaking them.

I've made a fairly good living on reed switches because Hammond Organ
used reeds actuated with pieces of refridgerator magnet rubber for pedal
switches for years. All this information comes from observing this
design defect and replacing reeds soldered into a really flimsy, cheap
circuit board wrapped in a curve around wood screwed down with sheet
metal screws as cheaply as possible. Hammond would have never used them.
Electromusic in Chicago has no scruples to cheapness under the Hammond
name. They made them this way.

Well, I just had a panic call from a pudgy little Phillipino girl who
plays the organ at her church. When they turned the organ off last
night, it stayed on and smoke came rolling out from behind the power
switch. The power switch has a paper bypass capacitor to keep it from
making a pop in the audio amp and I suspect it shorted and all the organ
current went through its short causing a little fire. I got an old
Hammond to fix at 1 then I'll meet her after work at 6 to have a look...
It's gonna be a profitable afternoon! SEND MORE THUNDER STORMS!!

Larry
--
Democrats are raising taxes on oil companies by $16,000,000,000.
Oil companies don't pay taxes, just like every other company.
Consumers pay all taxes, corporate and individual.
What's the price of a gallon of regular going to go to to pay $16B more?

So Larry, what would you recommend for this MOB application?

I have one of those plastic mounted magnetic switches used on windows in
burglar alarm set ups epoxied into the pole. The magnet is glued into a
Velcro strip with a line to the boat. When the MOB pole goes overboard, the
magnet stays behind and the light goes on. Perfect - except for reed
switches being unreliable.

What would be flush, watertight, and reliable?

--
Roger Long

On Mon, 6 Aug 2007 16:55:40 -0400, "Roger Long"
wrote:

So Larry, what would you recommend for this MOB application?

I have one of those plastic mounted magnetic switches used on windows in
burglar alarm set ups epoxied into the pole. The magnet is glued into a
Velcro strip with a line to the boat. When the MOB pole goes overboard, the
magnet stays behind and the light goes on. Perfect - except for reed
switches being unreliable.

What would be flush, watertight, and reliable?

There is a reed switch - a mercury wetted reed switch - which is less
bad than most - though environmentally suspect.

Brian W

"Roger Long" wrote in news:46b78aba$0$30688
:

What would be flush, watertight, and reliable?


http://www.nitro-pak.com/product_inf...roducts_id=802
&osCsid=810cd61405cfedde7f1534bb270970af

Not sure about "flush"....???

I just hate to think someone's life isn't worth more than a strobe from the
flea market when this one is only $20. There's one on my Sospenders with
the 10 year Lithium D cell. I go all out when I'm drowning....(c; Why not
use the same one on the pole?

Larry
--