Chinese ignition module

[minh-thanh]

Previously, I have used reed switch with 4 stroke engine, run time is about 30 minutes, they are fine .
Longer runtime I haven't tried so I don't know.
I haven't tried it with a 2-stroke engine so I don't know .
In my experience, the reed switch is fine with the engine at moderate speed .
With high speed, the reed switch does not have enough time to react to the magnetic field of the magnet
( of course my observations ... and of course no data to prove )

 

[Maris Dislers]

Would this ignition system work on a regular old-time model airplane engine with regular contact breaker points and running at up to 10000 RPM?

 

[TonyM]

I am in the process of making a 1/4 scale 'magneto' housing the Battery, Coil, Condensor, and all electronics. I am using a reed switch which I expect to be a serviceable component. KSK-1A35-1520 available from Digikey, Mouser or Farnell

 

[74Sprint]

If you use the KSK-1A35-1520 then you can expect a service life time of straight running time of avg. 100,000,000 operations which at 5,000 RPM is:
For 4 stroke only using the cam is half the below calcs, except the life time is double.
For 2 stroke:
5,000 / 60 = 83.34 per second or 0.08334ms.
5,000 RPM = 5,000 per minute.
5,000 x 60 = 300,000 per hour.
300,000 x 24 = 7,200,000 per day.
or
100,000,000 rated cycles / 7,200,000 = 13.89 days before it will fail, not bad.

My biggest concern is it can only handle 20G of shock before it opens erratically which, is about you flicking it good with your finger nail. My other concerns is the rating to open and close is kind of strong.

Also:
"Mechanical shock as the result of dropping the reed sensor typically from a distance of greater than 12” may change it’s magnetic sensitivity and/or destroy the sensor." ( from datasheet) I assume this would also include a cylinder firing or worse engine knock.

Operate and release times are good.
"Operating Time (max.)Incl. Bounce; Measured with w/ Nominal Voltage 0.5ms." Ah yes the dreaded switch bounce, not good.

"Release Time (max.)Measured with no Coil Excitation 0.1ms." Can't directly drive a coil, well it could but the release time would slow down. 0.1ms is really good but, it should be consistent.

I'm not against using a reed switch for they have a place. But if I have to use a magnet I would go with a Hall-Effect. The circuitry required is almost the same with the 2 but, the Hall out performs the reed. Then there is still the need to control the dwell. As far as switching a total time of 0.6 and a coil charge time of say 3ms and discharge time of 2ms still is good for 10,000 RPM so it comes down to life expectancy and degradation. Oh the reed switch must be mounted correctly to operate within the stated parameters.

Just my 2 cents worth.
Ray

 

[willray]

I'm not against using a reed switch for they have a place. But if I have to use a magnet I would go with a Hall-Effect.

I should say, nothing in what I've been arguing should be construed as a good reason to use a reed instead of a Hall or optical sensor in an ignition system. I only brought it up because my initial gut reaction was the same as Bluejets - no way a dinky little reed can do that! - but thinking about it more, it seems like most of the "obvious" reasons it wouldn't work, really aren't significant impediments. I find it interesting and noteworthy to observe when our gut leads us to an "obvious" but not necessarily true conclusion.

I do remain surprised that optical approaches don't see more use. Dwell should be so much easier to adjust with an optical system.

 

[TonyM]

74Sprint. Thanks for your comments. As its a magneto I would not expect it to go above 1000 - 1500 RPM. So lets say 40-50 days of running. 1000 days @ 1 hour a day. Most models will never achieve that so longevity is not a problem. Physical size is important. As it will be encapsulated with no moving parts touching it I didn't worry about 20G But I do take your point regarding the shock element as I intend it to be operated as a flick type magneto.

 

[willray]

I didn't worry about 20G

I expect that is 20G inline with the opening direction of the reeds. Shock loads transverse to the opening direction probably can be significantly higher - at least in terms of sporadic opening. Whether the reed can survive significantly higher shock loads without breaking is a different matter.

Also - and keep in mind that this isn't a reason to specifically select a reed over a Hall sensor - the compression stroke and coasting up to TDC should be a fairly uneventful period for an engine. If you're triggering on points (reed/Hall/whatever) opening, all that you really care about is the timing of the first opening.

Firing, shock loads, and other chaos happen after that. If the contacts bounce and you get spurious sparks during the power stroke, or any time up until you've wrapped all the way around to intake, it really doesn't matter.

 

[TonyM]

I realise that I have hijacked this thread a bit although my comment was about the reedswitch use. I am going to start a new thread on the magneto

 

[deverett]

After testing the ignition module again and finding it still didn't work, I cut off the sensor and put a test meter across the two wires. It was open circuit until a magnet was passed close to the sensor, then the meter showed 0 ohms. With a small hacksaw, I cut into the sensor housing and inside was a reed switch.

You were right, Tomlinson. Thanks to you and Zephyrin for dispelling my initial thoughts. It is indeed an internal component that has failed so the whole lot has now been consigned to the scrap bin.

Dave
The Emerald Isle

 

[Gedeon Spilett]

the electronics inside these little ignition modules is very close to the circuit of some electronic gas lighter, and as shown by Jan Ridders on his site, the conversion into a usable ignition is easy...
http://www.ridders.nu/Webpaginas/pagina_blokker_vonkcircuit/blokker_crcuit_frameset.htmand also on the videos of "mylittlediesels", who explain neatly how to make a ignition for little 4stoke engine...

some components may be salvaged, ie the ht coil, and you may have a go at making a new one...