Model Engine Maker

Supporting => Engine Ancillaries => Topic started by: Quadra on March 09, 2014, 10:19:17 AM

Title: Reed switch ignition
Post by: Quadra on March 09, 2014, 10:19:17 AM
Has anyone tried one of those very small reed switches and small magnet instead of the usual contact breaker, Hall effect, induction pickups etc ? I was thinking of using a reed switch as it could be made very compact  but I am concerned about accuracy of timing and reliability.
Title: Re: Reed switch ignition
Post by: stevehuckss396 on March 09, 2014, 12:49:39 PM
Has anyone tried one of those very small reed switches and small magnet instead of the usual contact breaker, Hall effect, induction pickups etc ? I was thinking of using a reed switch as it could be made very compact  but I am concerned about accuracy of timing and reliability.

I think you will find that the reed switch being a mechanical switch will have a slow switching time. In a slower running engine like a hit/miss it may work fine but in a 4 or 8 cylinder might fall short. There is no doubt it will work in a circuit as a replacement for a hall switch.

You have my interest peaked. I'm going to check it out and see what modern reed switches have to offer.
Title: Re: Reed switch ignition
Post by: Quadra on March 09, 2014, 01:55:11 PM
My aim was to use them on a twin cylinder engine of modest RPM.
Title: Re: Reed switch ignition
Post by: stevehuckss396 on March 09, 2014, 02:14:01 PM
From everything I read you should have no trouble running a 2 cylinder. They are much faster than I thought. I checked out some of the faster switches and found they should run an 8 cylinder to 5000 RPM.
Title: Re: Reed switch ignition
Post by: cfellows on March 09, 2014, 03:20:59 PM
The other issue you may have is bounce.  Mechanical switches don't close and open cleanly, but rather tend to vibrate or rebound slightly when they first make contact.  For some electronic ignitions, this causes the spark plug to fire when the switch makes contact as well as when it opens. 

Chuck
Title: Re: Reed switch ignition
Post by: Don1966 on March 09, 2014, 03:32:17 PM
I would agree with Chuck about switch bounce, but most read switch are short on how much current they are good for. You may find them to fail pretty quick if an electronic circuit is not incorporated into the circuit. So be careful how you use them.

Don
Title: Re: Reed switch ignition
Post by: gerritv on March 09, 2014, 07:29:00 PM
The other issue you may have is bounce.  Mechanical switches don't close and open cleanly, but rather tend to vibrate or rebound slightly when they first make contact.  For some electronic ignitions, this causes the spark plug to fire when the switch makes contact as well as when it opens. 

Chuck

Old school debounce circuit should fix the bounce. http://hackaday.com/2014/03/07/solving-endstop-woes-with-a-simple-analog-filter/
Or a schmitt trigger chip. Interface that to a driver for the rest of the ignition.
Reed switches were used a lot in the old tab equipment (pre-1945 mechanical computer stuff). And even into the 60's and 70's for high speed card sorters for triggering on card rows while the card was moving at 800+ cards/minute.
Title: Re: Reed switch ignition
Post by: gerritv on March 09, 2014, 07:38:35 PM
I would agree with Chuck about switch bounce, but most read switch are short on how much current they are good for. You may find them to fail pretty quick if an electronic circuit is not incorporated into the circuit. So be careful how you use them.

Don
Valid concerns but you can get some high current rating reed switches. And the mercury wetted ones reduce arcing even further: http://www.mouser.com/Electromechanical/Switches/Magnetic-Reed-Switches/_/N-5g2l/

Gerrit
Title: Re: Reed switch ignition
Post by: zeeprogrammer on March 09, 2014, 07:50:10 PM
I probably wouldn't go with just a Schmitt trigger. It will clean up some noise but any spike that exceeds the thresholds will still be seen - you get multiple nice edges. You'd also have to consider the supply the chip would need. Should be able to clean it up with just a couple of resistors and a capacitor.
Title: Re: Reed switch ignition
Post by: Don1966 on March 09, 2014, 07:57:25 PM
I would agree with Chuck about switch bounce, but most read switch are short on how much current they are good for. You may find them to fail pretty quick if an electronic circuit is not incorporated into the circuit. So be careful how you use them.

Don
Valid concerns but you can get some high current rating reed switches. And the mercury wetted ones reduce arcing even further: http://www.mouser.com/Electromechanical/Switches/Magnetic-Reed-Switches/_/N-5g2l/

Gerrit
I agree with you Carl and when you deal with ignition coils you have and inductive load not resistive as most of the switches are rated for resistive loads. Not the same rating and the inrush current helps to destroy it. Like I said a electronic circuit should be used.

Don
Title: Re: Reed switch ignition
Post by: Tonyr on March 09, 2014, 07:59:02 PM
Hello Quadra.
This is something I am trying out myself.
I want to trigger Jan Ridders ignition with a reed switch.
I have made one ignition and I have it working manually by passing a magnet close to the reed switch.
Later this week I want to mount the magnet on a disc and see how repeatable it is and how fast it will work reliably. If it works I can see how long the reed switch lasts.
I have a selection of magnets and reed swiches to try.
If it works it is a cheap and simple system, if not I will have to go back to a traditional system.
Nothing to lose by trying.
Tony.
Title: Re: Reed switch ignition
Post by: PStechPaul on March 09, 2014, 09:49:09 PM
You might try an inductive sensor which does not require a magnet. I got one of these and it seems to be well made and operates nicely, for about $5.
http://www.ebay.com/itm/Inductive-Proximity-Sensor-LJ12A3-4-Z-BX-3Wire-NO-NPN-6-36V-Detection-Switch-AK-/281222464702 (http://www.ebay.com/itm/Inductive-Proximity-Sensor-LJ12A3-4-Z-BX-3Wire-NO-NPN-6-36V-Detection-Switch-AK-/281222464702)
or
http://www.goodluckbuy.com/lj12a3-4-z-bx-inductive-proximity-sensor-switch-npn-dc6-36v-4mm.html (http://www.goodluckbuy.com/lj12a3-4-z-bx-inductive-proximity-sensor-switch-npn-dc6-36v-4mm.html)
http://www.goodluckbuy.com/lj12a3-4-z-ay-pnp-nc-dc-inductive-proximity-switch-sensor-1.html (http://www.goodluckbuy.com/lj12a3-4-z-ay-pnp-nc-dc-inductive-proximity-switch-sensor-1.html)
 
It operates up to 100 Hz which should be good for 6000 RPM (and the second one 150 Hz). But I haven't actually used it for ignition or motor speed/position sensing. I planned to use it for a small vehicle speed sensor.
 
Hall sensors are very cheap:
http://www.ebay.com/itm/Hall-Switch-Sensor-Motor-Speed-Test-Smart-Car-Accessories-Module-For-Arduino-/271234601242 (http://www.ebay.com/itm/Hall-Switch-Sensor-Motor-Speed-Test-Smart-Car-Accessories-Module-For-Arduino-/271234601242)
http://www.ebay.com/itm/10pcs-A3144-A3144E-OH3144E-Hall-Effect-Sensor-Switches-TO-92UA-3pin-SIP-/191081231627 (http://www.ebay.com/itm/10pcs-A3144-A3144E-OH3144E-Hall-Effect-Sensor-Switches-TO-92UA-3pin-SIP-/191081231627)
 
I don't think most reed switches will handle the current and voltage for the inductive load of an ignition coil. It would be about 2-5 amps and 400-600V. Much better to use an electronic circuit. And there are a number of spark generator circuits and modules discussed in Brian's thread in HSM: http://bbs.homeshopmachinist.net/threads/61789-A-different-opposed-piston-engine/page54 (http://bbs.homeshopmachinist.net/threads/61789-A-different-opposed-piston-engine/page54). I plan to build a simple circuit and have it at the Cabin Fever expo in April, and I'll post a build thread here with schematics, pictures, and video.
 
SimplePortal 2.3.5 © 2008-2012, SimplePortal