If I had found a cap the size listed on the diagram I would have breadboarded the circuit. And if I hadn't collapsed with the heat here, and the cocktail, I would have tried with the 3300 microfarad cap I do have.

Maybe in the cool of the morning

Alistair

> On Aug 1, 2017, at 4:51 PM, David Beierl <dbeierl@ATTGLOBAL.NET> wrote:
> 
> I perhaps should add that LEDs by their nature require a current-limited
> supply, either by the internal resistance of the battery/power supply, an
> in-line resistor, or active electronic limiting.  Like any diode, once they
> reach the voltage where they begin to conduct (~1.8V to 4+V for various
> types) they will pass drastically increasing current for small increases in
> drive voltage.
> 
> I just tested three random 5 mm LEDs from a Chinese bulk supplier -- one
> red, one blue, and one white (which is actually a blue LED with  a yellow
> phosphor covering the chip).  LEDs like these are typically rated for 20
> mA, and after my test these had permanently changed characteristics from
> overdriving.  I tested at current levels of .06, 0.6, 6, 20, 60, and 120
> mA.  Voltages were as follows:
> 
> Red: 1.6617, 1.7765, 1.9425, 2.0769, 2.1300, 2.0100.  These values were
> higher than I expected, typically red LEDs run around 1.8v at a few mA.
> Note the falloff at the highest current, as the LED was being damaged.
> 
> Blue: 2.4917, 2.6055, 2.9003, 3.2820, 3.6830, 3.9600.  A second sample from
> the same bag was closer to 2.1V at .06 mA (not tested further).
> 
> White: 2.5639, 2.6904, 3.0303, 3.3745, 3.6960, 3.9125.
> 
> I would expect higher-powered LEDs to have a more rapid current increase
> per voltage change, but I haven't tested any as all the ones I have are
> installed with driver circuits.
> 
> My test rig was a variable 0-15V supply in series with a 2K ten-turn
> wirewound potentiometer.  I varied both voltage and resistance as
> convenient to get the desired currents.
> 
> Yrs,
> d
> 
>> On Mon, Jul 31, 2017 at 11:17 PM, Neil N <musomuso@gmail.com> wrote:
>> 
>> Ironically I was outside taking apart, putting together, taking apart
>> a fog light switch. Who knew a later style dual bulb fog switch would
>> be in my '81 air cooled? Long story. Anyhow....
>> 
>> Great to see the replies. I will read them!
>> 
>>> On 7/31/17, David Beierl <dbeierl@attglobal.net> wrote:
>>> 
>>> With the given values and a two-watt festoon bulb, the cap has enough
>>> energy to supply the lamp for a third of a second at full brightness.  In
>>> practice it would fade down over a longer time, maybe a second?
>>> 
>>> Yrs,
>>> d
>>> 
>> 
>> --
>> Neil n
>> 
>> Blog: Vanagons, Westfalia, general <http://tubaneil.blogspot.ca>
>> 
>> 1988 Westy Images <https://picasaweb.google.com/musomuso/New1988Westy>
>> 
>> 1981 Westfalia "Jaco" Images, technical <http://tubaneil.googlepages.com/>
>> 
>> Vanagon-Bus VAG Gas Engine Swap Group <http://tinyurl.com/khalbay>
>>