Chinese ignition module

[deverett]

I have one of the Chinese ignition modules that was working fine until the earth wire became detached.

Reading various places about these modules failing, it is probably the Hall sensor that has fried.

I'm considering getting a replacement Hall sensor to try, but the ones I have seen have 3 wires whereas these Chinese systems only have 2 wires going to the Hall sensor. So can anyone tell me what type of sensor to look for?

My electronic ability is zero and I don't have the confidence to try and muddle my way through making a system from scratch.

Dave
The Emerald Isle

 

[minh-thanh]

deverett !

Hall sensor usually has 3 pins: Power ( 5 V ) , GND , Signal ( I have not seen the 2-pins type )
I think - maybe - your igniter uses 2 wires for the Hall sensor : Power , Signal , and the GND of the sensor will into the engine body .

 

[stevehuckss396]

Do a search for "2 wire hall sensor" in google and you will find plenty of potential replacements. They are 3 lead devices with a signal lead and 2 grounds. Basically the sensor shorts the signal lead to ground. The big suppliers like digikey have them.

 

[deverett]

Ah, Steve. I should have guessed that Google would be my friend!

Dave
The Emerald Isle

 

[Tomlinson]

I could be wrong, but I think you will find that the sensor on that device (2 wire) is a reed switch

 

[bluejets]

I could be wrong, but I think you will find that the sensor on that device (2 wire) is a reed switch

I doubt that......

 

[Gedeon Spilett]

Yes, the 2 wires "sensor" is just a plain reed switch...you can easily find the same at RSpare.
but I doubt that it is the failed component, alas, the cooked one is probably embedded in the putty...

 

[74Sprint]

2 wire Hall sensors use pwr for one lead and the other/s is used for ground/signal. They are usually packaged with 3 leads, pwr, gnd, gnd and they interrupt the ground so you need additional external hardware to use them, usually current sensing. So one could tie the 2 grounds together to make it a true 2 lead Hall.

2 wire Hall

Ray

 

[Tomlinson]

Regarding the Reed Switch, It's worth checking that the switch is not remaining closed after a spell of use. I had this problem with the same device. The unit worked fine for a short time then failed to generate a spark. I checked the switch and found it short circuit. Strangely, after leaving it for an hour or so it began to work again. Almost as if the switch was becoming magnetised and holding closed. I overcame the problem by substituting normal contact points for the reed switch.

 

[Gedeon Spilett]

it is this kind of reed contact
https://fr.farnell.com/hamlin/59140-1-t-02-f/sensor-magnetic-spst/dp/1839038
it do not depends on the orientation of the magnetic pole to be activated, as is the Hall sensor, but the position of the magnet can be critical for a precise triggering...and a ferrous part in proximity may hinder it.
on a home made ignition module with a reed switch for a mini engine model , closing the bonnet on the engine instantly stopped the ignition...I had to change the position of the switch, after long tests...

 

[deverett]

it is this kind of reed contact
https://fr.farnell.com/hamlin/59140-1-t-02-f/sensor-magnetic-spst/dp/1839038

That looks like the one fitted. Good find on that, so many thanks.
It's worth paying the small cost to see if that was the problem.

Dave
The Emerald Isle

 

[bluejets]

Might look like the same housing but I doubt very much it would be a reed switch.
Specs say best operating time (apart from life span) is around 3 milliseconds.
Reed switches would never stand up to the demand of an engine ignition system.
They're made for such things as door and window sensors for alarm systems.
An engine idling at 1000 rpm covers over 16 revs per second (thats close to 6,000 degrees)
You do the maths......

 

[willray]

Reed switches would never stand up to the demand of an engine ignition system.

I've got to say, my initial gut reaction was the same, but... if it's a low-current signal? Points handle vastly more current, and generally in a much more hostile environment than the inside of a reed switch glass envelope, and they seem to have worked well enough to get us down the road reliably for a while.

 

[bluejets]

Point being, it's a mechanical leafspring that bends to open and bends to close.
How long would that last on an engine, before even considering the opening and closing times required.
As pointed out the best that reed could handle is 3 milliseconds, hardly cover the idle let alone anything else.

 

[TonyM]

Although that example is 3 miliseconds there are plenty of reeds that can cycle in less time. Even so, with 6 000 milliseconds in 1 minute thats 2 000 RPM on a two stroke and 4 000 RPM on a four stroke. Life expectancy at full power throughput can be as little as 8 hours of run time but at low load the lifetime can be 8 000 hours. It's a matter of specifying the right switch for the job.

 

[Gedeon Spilett]

I did little ignition modules according to the site of Jan Ridders with modified electronic gas lighter, and they worked a treat, with reed switch or home made switch that imitates conventional contact breaker.
and I'm not alone, see for instance "mylittlediesel" with several videos on this topic.


one of my own ignition block
https://photos.app.goo.gl/nVmavjqMDR1HmX959

you may see on this picture the reed contact and the magnet on the camshaft.
I do have videos of my loco running on my Channel,

 

[willray]

Although that example is 3 miliseconds there are plenty of reeds that can cycle in less time. Even so, with 6 000 milliseconds in 1 minute thats 2 000 RPM on a two stroke and 4 000 RPM on a four stroke. Life expectancy at full power throughput can be as little as 8 hours of run time but at low load the lifetime can be 8 000 hours. It's a matter of specifying the right switch for the job.

There are not 6,000 milliseconds in a minute... There are 60,000. I don't know what Bluejets' concern is about the timing. 3ms is enough for over 300 open/close cycles per second.

And the reed spring is under far less stress than points springs or the valve springs, so I'm not seeing the issue there either.

 

[TonyM]

There are not 6,000 milliseconds in a minute... There are 60,000.

Oops - Of course there are. Point made though.

 

[bluejets]

Point being the point open during a 10 degree or less period.
Do the maths.

 

[willray]

Point being the point open during a 10 degree or less period.

Maybe I'm just being dense, but -- how is the normal-system points-open time relevant? The math says at 1000 RPM (single cylinder) you've got about 60ms before the points need to open again, so there's more than enough time for them to close 3ms later, rather than 10 degrees later.