flip/flop latch for ignition + kill sw

Hi, sorry for asking a repeated question, but it seems that there a many ways to achieve this result.
I am want to use a single + pulse  input to enable/disable the ignition of a motorcycle, but also want a kill (NC to GND) switch to turn off the ignition and reset the state to off, so I cannot use a latching relay?

I believe this configuration would do this, all I need to do is add a relay in place of the push button.
https://www.emil.matei.ro/onof2.php
This looks right. Connect coils in series and enable switching through current flow taking path of least resistance.

Can anyone suggest a better solution and what I should do to make it a real world solution, I want to protect the circuit and prevent power drain.

Is timing an issue and would I need to use relays with built in resistors to ensure? 

I think mechanical relays are best as they do not have current draw when in the off state and solid state relays could get hot and drain the motorcycle battery…or is that negligible?

Many thanks!

Are you using the standard kill switch at all (ie, handlebar spark kill)?
But that's probably academic - the usual kill is a short across the ignitor or timing sensor (points, hall-effect sensor, reluctor pickup etc). That's using the NC contact of an SPDT relay (aka 5-pin), hence no current drain when off.

The problem is how do you want it activated?
I'd assume you'd want it killed when IGN is ON until some switch/circuit is activated hence energising the relay and opening the short. (For a momentary push button etc, my latching relay circuit using a normal relay should work - albeit in this case modified for an SPDT relay. (See here or here.)

If it's to be a "set to kill" prior to IGN OFF, then IMO a latching relay which is then unlatched with IGN ON and a mom PB etc.

The "short across ignition pulse" method (whether sensor or ignitor) has the advantage of NO cuts to existing circuits, and no ignition misses (kills) during severe bumps (or cornering LOL) or relay/circuit failure.


I too prefer mechanical relay over solid state. If nothing else, normal transients/spikes won't kill a mechanical relay.

Hi, no your right, the Kill is a simple switch that opens the ground connection for the coil. For some reason I thought it would be a good idea to have the kill switch turn off everything.

I am going to rewire the motorcycle and have an RFID that I was going to use, it will generate a positive pulse. I don’t care about an accessory line, I just want the relay to toggle states when it receives a pulse from the RFID circuit.

Perhaps I haven’t thought about this enough, and there isn’t much value in associating the kill switch with the ignition switch… But I do want to associate the circuit with the coil as the motorcycle is a kick start so it would be possible to start the bike when the “ignition is off”.

I wanted to put the entire electrics into an off state until activated by the RFID circuit, I was also thinking about triggering a timer off of the brake switch to activate the RFID antenna for 15 seconds.

You’ve made an interesting point about the bumps, so I should look at keeping the coil energised to keep it off and then cut power to turn it on? But am I then making it easier to “hot wire” the bike as they could cut the power to the relay and the ignition coil will be closed? Or am I plain dumb?

So your circuit will stay latched after a single pulse. I would like a flip flop circuit like:

This posted by "hotwaterwizard" looks good, "Pulse Toggle Relay"

https://www.the12volt.com/installbay/forum_posts.asp?tid=106633&tpn=1&PN=1

I think this will be the way to go, I will have the output switch a SPDT power relay to activate the battery circuit (typical ignition) and the kill ground circuit, right before the NC kill switch.  I could use the GROUND before the KILL to deactivate the RFID/ignition circuit, does that sound fesible?

I am thinking I'll need three diodes right before the coils as the load line is multi-directional.

So I am also thinking about the vibration issue, what is the best way around this? NC switch for the KILL line that is constantly energised? I hate the idea of using energy to keep the contact open...

Is there a good book or software application that I could use to workout how to do this stuff? I'm getting it, but slowly.... :-0 

Again, are you talking about the bike's kill switch?
If so, they open the IgCoil (or Ign) power.
Kill switches do not open IgCoil grounds because they have none - grounding is thru the points or ignitor - EXCEPT for certain CDI etc systems.

And as with cars, engine kill and battery isolators (ie, system inhibitors) are two totally separate things - despite some combining both which means the alternator is still connected to the battery which many regard as not isolating the battery from the rest of the vehicle. (But those people are silly enough to allow such remote & cabin isolators to be on the hot side (+12V) instead of GND where is should be for the safety reasons they claim to be so concerned about.)


Hotwaterwizard has been excellent for latching and other relay circuits.

I'll review in more detail later.
I know you wont be using the bike's kill switch (obviously) but wondering if it was that circuit you were looking at interrupting. (Though I still suggest the non-insertion add-on shorting method - with some provisos - eg NOT a long short that can allow high RPM spark.)

Hi thanks for the quick reply, so:

oldspark wrote:

---

Again, are you talking about the bike's kill switch?

---

Yes, the bike is a 1976 Suzuki GT500 (CDI) and I believe the kill switch is NC and connects the ground circuit of the coil.  So I would need something heavy duty to keep this closed when the ignition is ON. I am now thinking that a SS switch might be best.

oldspark wrote:

---

And as with cars, engine kill and battery isolators (ie, system inhibitors) are two totally separate things - despite some combining both which means the alternator is still connected to the battery which many regard as not isolating the battery from the rest of the vehicle. (But those people are silly enough to allow such remote & cabin isolators to be on the hot side (+12V) instead of GND where is should be for the safety reasons they claim to be so concerned about.)

---

So the idea was two (three for the RFID) isolated circuits activated by the latch. I could use the RFID circuit load to 1) activate the "battery circuit" via a 30A EM relay and 2) "engine kill circuit" via another switch. Question is would a SS switch be better for 2? I dont want this to cut off over a large bump, but is it a good idea to mix a "high resistance" switch for the spark? I dont want to affect the spark quality.  


So thanks again for your time. It sounds like you really know your stuff. I am just building my first motorcycle and am a newbe on all fronts....heaven help me when its done. 6months consumed and should be done in a couple of weeks! :-)

Hmmm.... My best mate had a GT550 (2-stroke triple) of similar vintage and that was definitely points ignition with the usual points (only) grounding if the IgCoils.
Later he did replace that with a CDI or similar. I remember it well because it was a wasted/batch spark - ie, one IgCoil firing all cylinders, and proved that you can fire any splug and it won't detonate unless compression etc is correct - even if there is a "combustable" air-fuel mixture present.
Not that I can remember if it was a 120° or 180° degree triple (but I think 120°). And he did the same to his later Laverda Jota triple for which both 120° & 180° versions existed (he had the rarer (earlier?) version), but that's a 4 stroke...


But you don't want to keep that NC kill circuit closed when ignition is on, you want it open when ign is inhibited. (Otherwise you have no OEM kill switch function, and they are there for a reasons - as well as a parking convenience LOL.) [POST EDIT - that is probably what you meant. I'm clarifying your writing ELSE my misunderstanding or misinterpretation.]
Hence an in line NO relay contact which closes with IGN on and the "un-disable" activated.

But that is an insertion, and any relay bounce or circuit failure will be an issue.

I still suggest shorting of the sensors.
If the CDI is integral with the sensors, then that can't be done, but usually sensors are separate.
Most sensors are grounding types - eg points, most Hall-Effect and opticals - but even if +ve outputs, it's a "mere" case of shorting sensor outputs to their supply. IE - the equivalent to shorting points etc outputs to GND.    

If integral CDI and sensors, I'd be inclined to short the CDI output to GND (equivalent to shorting points & ignitor outputs to GND).
Because CDIs are a capacitive discharge into the IgCoil, shorting to GND should not be an issue**. But make sure that short is solid - I know of instances where too long wiring (or high resistance) means that the spark pulse still gets thru (like when a bike was redlining with my bare leg inconveniently cushioning its exhaust header from scratching the bitumen underneath!) - not that I ever tried starting with such a "failed" system engaged. (That's more of a warning for shorting-type kill switches, but motorcycles are usually NC power break circuits to ensure engine stoppage.)

Again, I'd avoid SS relays. I don't see the need, and I don't like their possible resistance or spike issues - nor their expense.
And either way, since this is a "passive when armed" circuit (no power consumed), mechanical relays are perfectly suited.
But an SS relay could be added to parallel an in-line +12V connecting relay (if you choose that method) to overcome relay bounce.


As to knowing my stuff - yeah, I do - except when I get it wrong, or don't know. (Usually by not being up with what has finally hit production. )


** CAVEAT - shorting a CDI to GND should not be an issue, but that may be very CDI dependent. I can see how bad CDI designs could be damaged.

Hi mate, you are quite right. I got that part wrong, you create a GND to the coil to turn off the engine, like the kill switch. I am mixing myself up.

The the problem is, what is the most reliable interface to break my NC switch on the coil GND (kill) so that I can start the engine. I think your suggestion to use a SS relay as a backup could be a good solution.....unless you can recommend a grade of relay that should be able to handle the vibrations.

Since this I've looked up som automotive grade relays and found some that can operate within 4.5 G of vibration...would that be enough? I guess the mounting on the bike would play a big part too.

Are there a set of devices, brands, installation techniques that I could check out?

Pre-submit edit: Rather than bother with all I wrote below, why not simply use an SPDT relay apply a short as I previously described?   


Re the GND to coil to kill the engine, NO....? Motorbike kills that I know of are centre-closed +12V feeds to the ign (IgCoil etc). Either side (kill/stop position) is open circuit.
The "bike" I referred to that used a "points to GND" was a Suzuki step-thru (1970s; 50cc or 70cc 2-stroke) because they had no handlebar kill switch, and they were magneto anyhow. Ign-on essentially opened the magneto/points short.

But you don't want to break a kill switch short - that'd be useless since you just flip the kill switch. You want to parallel the kill switch short.
And if it is the typical center-closed +12V connection that I'd expect on non-stepthru Jap bikes from 1970 onwards, I'd do the theft or start kill by shorting the points etc. Hence no issues with relay bounce.
Relay bounce or failure is only an issue when you insert into an existing circuit, hence why I always avoid it for ignition kill circuits. (And IMO electric starter kills are useless - especially on cars.)


But if inserting, then 12V relays should handle the vibration etc unless they are only getting (say) 10V or 9V etc. But the relay bounce concern is once for extreme competitions - eg rally cars that hit 20kph yumps at 160kph, or speed up on the rough stuff (as I did). Not that I ever noted any relay bounce in my car, but I only used relays for the beams, and those relays had more force than typical 12V relays (the early New Era dual-relay types). The only hazardous bounce I suffered on my car (during MY "normal" road driving) was heads off the roof. When those 2 heads are attached to somewhat bigger than me bikers, that is IMO a hazardous situation.
But relay bounce and relay failure (no matter how rare) as well as connection integrity is a concern for any additional insertion into a critical load. However I would take my chances with a normal electro-mag relay over a solid-state relay. IMO ss relays are not robust or reliable enough for critical loads (like ign etc) &/or they are too expensive, and large, etc. A typical automotive 200V spike won't effect a mech relay whereas SS relays usually rely on spike protection that can age. (Not that they use MOVs, but AC protection that uses MOVs is a classic example - they take a certain total accumulated spike energy and then have no effect. A good lightning strike can kill them fist go.)


As to the 4G handling, is that constant 4G (which is fairly hard to achieve), or a peak G? Like if PC HDDs can tolerate 200G "spikes", I'd expect relays to handle more...

Hey thanks for all the useful advice, I'll the chance and fit the RFID circuit as the wiring will be almost identical to the ignition key switch. I'll use a SPDT (can you get NO and NC in one?) to cut the parallel GND from the coil and the NO to open the fused battery supply to the motorcycle electrics.

I was thinking to then run a dedicated realy control line and then n number fused power lines for the head light, spotlight, USB, tachometer and speedometer. Does that sound like over kill? :-) I'll digitise my circuit on Sunday as I am putting the bike back together tomorrow and the wiring will be next.

Hey thanks again! Where in AU you from? I'm originally from Melbourne.

I'm in Melbourne. Presently in the Greensborough area.


But I suggest you HALT and reconsider.
If you are adding the GND to prevent staring, there is NO cutting involved.
And you do NOT touch the existing kill switch circuit.

And why do you want to open the existing battery supply to elsewhere? (And when?)

I have only been discussing an engine kill -cum- start disable function.
I may have mentioned "interrupts" or "isolation switches" but that has only been to point out the difference between an ignition kill and "safety" type battery isolation circuits, and to mock those that combine the 2 into ONE switch without understanding the issues and hazards, or to point out the risks with interruption or breaking circuits.


I'm happy to endorse a staring or ign immobiliser that uses the NO contact of an IGN +12V actuated relay else the NC contact of a de-energised relay (when IGN is off, and until the IGN with the "unlocker" energises that relay).

I am NOT endorsing nor mean to suggest any form of circuit or relay insertion - ie, no cutting of existing wires etc. (Nor have I ever done that for critical circuits except under specific conditions.)


BTW - that's what an SPDT relay is - "change-over" contacts. 87 is NO; 87a is NC (with respect to the "common" 30).
And many buy only SPDT relays since that can be used as mere SPST relays. (Hence one part or spare suits both SPST & SPDT applications.)

Hey thanks for pointing out the obvious on the relays. I suppose I could then go for a DPDT to break the coil ground and the other to open the fused +12v supply to the rest of the bike. I would use the other relay to allow power from the battery to supply current to the lights, tachometer etc...like an acc.  so when the IGN is off, nothing will work, engine or electrics, e.g. light, horn etc.. do you think that is a bad idea?

About the wiring, I think I was paraphrasing you. I would keep the KILL swich between the coil and GND. I would also tap in another wire in parallel from the coil to GND through one of the relay poles.

The other relays were to support the idea of reducing load and isolating devices with their own fused supplies, i.e. headlights on their own fused line a USB connector on its own fuse and possibly through a switch.

You still do not understand - there is NO COIL GROUND to or kill-switch GND to break.

An Ignition Coil has no hard ground connection.
It has +12V connection from the IGN switch (or IGN relay) thru the kill switch to the coil+ and the HT lead to the splugs (else dizzy).
Its 3rd connection is from the points or ignitor which connect the IgCoil to GND - that is its ONLY connection to GND. (To GND charges the coil, breaking GND fires it.)

The only difference with most CDIs is that there is no IgCoil charging time - the "points" cause a capacitor to discharge into the IgCoil. (The IgCoil acts as a pulse transformer. It is not pre-charged as with traditional/normal inductive/Kettering type ignitions.)

But in both types, there is no GND to any kill switch to interrupt...


So if you are talking about GND connections, that has nothing to do with the bike's kill switch, so keep the kill switch out of it.   

And if you are talking about breaking the +12V thru the bike's kill switch or breaking the coil- trigger path, then I won't help any further because of the risks I outlined earlier.   


If you do have a system that does not operate in accordance with the above, then explain how it is wired and how it works and I'll take it from there.

Hi mate,
Sorry, I don’t think I have been clear in my posts. I realise that there shouldn’t be a connection from the coil to GND (for normal operation). Now on my motorcycle, there are two ways to kill the engine, the KILL switch which creates a connection from the coil to GND and the ignition switch which does the same (via a parallel connection) when it is in the OFF position.

You can reference the following wiring diagram:
https://www.suzuki2strokes.com/forum/viewtopic.php?f=29&t=4743&start=8
The ignition switch also creates a connection to supply fused +12v from the battery to the motorcycle electrics when it the accessory or ON position.

So all I want to do, is find a way to substitute the functionality described above with an "RFID switch" and I am happy to skip the accessory mode, and just have ON and OFF.

Now using a latch similar to hotwaterwizard along with the RFID module, I have my "RFID switch". The next challenge was how to perform the two ignition key switch steps.
a) Remove the GDN connection to the coil (i.e. Ignition off)
b) Allow fused 12+ to provide current to the electrics.
Based on our discussion I could do this with a DPDT relay or two SPDT relays, let’s assume I am using two:
Relay 1.
Pin 30 to GND, Pin 87A connected to coil, Pin 87 not connected
Relay 2.
Pin 30 to +12v fused out from battery, Pin87A not connected, Pin 87 connected to motorcycle +12v supply.

My questions
a) Would Relay 1 be a safe addition to the circuit? I was worried that it might detach due to vibration.

b) Is the latch in hotwaterwizards design good for this application, should I add any other protection (diodes, resistors)? This relay circuit would also be subject to mechanical failure. But then again, what kind of relays are used by modern vehicles and motorcycles?

Hope I have been a bit clearer. Thanks again for your time.

Ok, so you have my step-thru's type of system. I would not have expected that on a normal bike.

Your relay connections above are fine; both relays energised by IGN and the RFID.
Relay bounce has no effect on the points/ign short.

I'm not sure that the +12V relay is needed nor desirable.
Ign is shorted until the RFID is used, and that should only be with the IGN key.
So why cut the +12V unless you want the RFID to bypass the IGN switch which means the RFID receiver is draining the battery?

I figure IGN on which powers the receiver.
The RFID then pulses the latch to energise the NC spark shorting relay.   
I reckon I'd use my single relay latching circuit - maybe by using a DPDT relay to do both (latching and shorting).
IGN off kills the +12V to the relay coil, hence breaking the latch and reapplying the short until IGN-on and RFID repeats.

But I am a bit on the tired side...

oldspark wrote:

---

I'm not sure that the +12V relay is needed nor desirable.
Ign is shorted until the RFID is used, and that should only be with the IGN key.
So why cut the +12V unless you want the RFID to bypass the IGN switch which means the RFID receiver is draining the battery?

---

So I was thinking it would be safer to have my electrics OFF when I am not arround. Stops people truning on my lights etc...or letting them think they could start the bike.  I Thought about having a single connection off the battery connected to a switch, either dedicated or off the front brake that will activate a timer to provide power to the RFID antenna. Thinking a simple push button switch would be best.

oldspark wrote:

---

I figure IGN on which powers the receiver.
The RFID then pulses the latch to energise the NC spark shorting relay.
I reckon I'd use my single relay latching circuit - maybe by using a DPDT relay to do both (latching and shorting).
IGN off kills the +12V to the relay coil, hence breaking the latch and reapplying the short until IGN-on and RFID repeats.

---

Cool, using one DPDT relay makes the most sense, I just thought I would be best to use a "high power" relay to run the +12v BAT to the rest of the bike. I have upgraded my alternator and will be putting some serious lights on the bike.

So the idea was that I could use the RFID reader to turn off the IGN, this is why I was looking at the "flip/flop" relay design. You're circuit is a permanent latch held until supply voltage is lost, additional pulses from the RFID circuit will not alter the state.

I am in two minds about doing this just yet as I have spent the best part of 4 months re-building this bike from the ground up...so I am sure I will have other issues to deal with...I just would have liked to do away with the key and have a cleaner looking bike...

Thanks again for your thoughts.

So my take away is that the use of "automotive" relays should be fine and failures  from viabration should not be a concern, I am not doing stunt riding or racing.  The circuit provided by yourself will work as will hotwaterwizard's latch. I guess the only thing would be to put a diode accross the coil.

But your electrics are off when you are not around the bike. Removal of the IgKey cuts power.

Without any IgKey we have that issue if the relay bounces or fails, or the battery/supply goes too low.    
Also that the RFID is always powered, but that depends on its drain versus battery capacity and desired battery life.
And being stranded if you have a flat battery - unless a push start and a permanent magnet type alternator else enough residual magnetism in a wound rotor.
[ PS - Sorry - I am repeating earlier stuff that you have acknowledged. ]


You are quite right about the DPDT rating - high current for the alternator & lights.
I was assuming a 30A relay which should be more than enough for your alternator, though I also assumed that since the kill switch was a grounding type, the alternator was a typical "matching" dual-output - one dedicated to lights and the other for the battery and bike (ign, instruments, horn, flashers, stop, etc) and probably a stator = permanent magnet rotor type.


I wouldn't think a diode across relay coils would be necessary, but better safe than sorry.