RogerN wrote:

"IanM" wrote in message
...

Wes wrote:


Bruce L. Bergman wrote:



As for pre-installing the Back EMF Snubber on stock products, they
never know what type or voltage relay you're going to use with that
relay socket. But it would be a cool and useful factory option if you
are ordering production quantities of relay sockets.


Good point on the 1N4XXX. Should not have trusted memory. As far as the
prewired diode,
I wanted it in the relay itself. AB makes some dc coil control relays
with an internal
snubbing diode. Wes

The diode is reverse biassed when power is applied. It conducts ald allows
the coil current to recirculate decaying slowly when power is removed. It
*never* sees the full back EMF spike as it shunts it before it builds up.
You'll never see a problem from the back EMF if the diode is rated at
twice the supply voltage so an 1N4001 is fine for a 24V relay. The only
'gotyas' are that the diode *must* be rated to carry the coil current and
as the flux will decay a lot more slowly, the relay *will* be slightly
slower opening.


That's what I've wondered about electronic ignition. Do they use real high
voltage ignition coil switching transistors or would the magnetic field
still collapse fast enough to spark a spark plug with a diode? Or perhaps
they use a condenser (or equivalent) on the solid state components like they
do on ignition points.

RogerN


The usual setup for switched mode power supplies is to use a transistor
rated for a very high voltage for a very short time. Even so they have
to use a snubber consisting of a very fast high voltage diode in series
with a small capacitor. Instead of clamping the spike to the supply
rail, this alows the series combination of the coil and capacitor to
'ring' for half a cycle, ending up with all the remaining energy in the
coil primary that didn't get coupled into the secondary transferred to
the capacitor which is typically chosen so it will reach 1 to 2 times
the supply voltage. There also is a high value bleed resistor (usually
a high power one) that is chosen to effectively discharge the capacitor
for the next pulse.

If you want the transformer to 'ring' for more than 1/2 cycle then the
snubber can be just a capacitor, often with a low value resistor in
series with it so you dont blow the transistor at switch on.

if you fitted a normal anti-parallel diode, the primary current would
rise and fall smoothly in a sawtooth and the output voltage would be
insufficient for a spark plug.

However, ignition circuits frequently used a thyristor and that a whole
other ball game . . .

Unfortunately I cant find my old Heathkit electronic ignition manual to
see what approach they used, but I did find a site with some links to
circuit diagrams that might be of interest :

http://www.molla.org/DIY-CDI/Others/Others.htm