Topic: Transistorized ignition, points style. First ones are ready!

[bryanj]

Many years ago i had a bought system on a car (ford) which incorporated a switch to convert direct back to points and a small led to do the timing. outlasted 5 cars as i remember and once fitted never checked the points/timing again, count me in for a couple of sets as long as customs dont get me!

[eldar]

I would be interested to hear the results of this. It may be something I might like. Another alternative to getting a dyna.

[Raul CB750K1]

Hey Hondaman, you crearly know what you talk about. Thanks for the explanation.

Nevertheless, I have to decline your offer. And not because I don't trust you can do it, because by your words it's clear you know about electronics. But being an engineer myself, I feel somewhat stupid letting somebody do what I thought about doing. I have had bad experiences in the past, building things that you could buy cheaper and better, but after all, nothing beats the feeling of achievement when something you design and build works...

Anyway, this is far below in my "to do" list, so it will still take some months before I start working on it.


Raul

[Kevin400F]

Do you absolutely need to invert the points?  Can't you use the 1-4 point set to drive the 2-3 coil and vice versa?  I think their opening and closing events are exactly 180 degrees out of phase.   Simpler?

Kevin

[Raul CB750K1]

Do you absolutely need to invert the points?  Can't you use the 1-4 point set to drive the 2-3 coil and vice versa?  I think their opening and closing events are exactly 180 degrees out of phase.   Simpler?

Kevin

What he means is that, as the points are connected permanently to ground, when they close is when the transistor opens, so it's necessary to insert another stage to invert the logic and then the final transistor opens when the points open and vice versa.

At least that's what I understood. Hondaman please correct me if I'm wrong.


Raúl

[Jim F]

Terry would kill me if I pulled off his ARD (and everyone else that has helped) to give this a try but for laughs and giggles I might get one for the future.
Like Terry I am a gadget guy too
Jim

[Frank T]

I would be interested in purchasing one also.
Thanks,
Frank
Decatur, AL

[Kevin400F]

Raul & Hondaman, if I understand the logic, if no inverter logic is used the main transistor would open when the points close, and the coil fires when the main transistor opens.  Is it not true that the 2-3 point set closes at the exact point the 1-4 set opens?  If so, couldn't you connect the main transistor firing the 1-4 coil to the 2-3 point set, and avoid using an additional component to invert the logic?

Kevin

[HondaMan]

Raul has it right: the problem is just that the transistor itself needs "reverse logic" to drive the coil in a normal Kettering way. For example, some electronic devices (like GM's HEI) fire the coil by passing current through it, like a CDI does. This is reversed from a Kettering, which charges the coil with current and then "drops it" to get the spark from the collapsing magnetic field. There are pros and cons to the different systems, but the main thing here is: Honda's stock coils won't fire by hitting them with current: you have to do it like the points did. CDI and HEI systems require specially-wound coils to "turn it around", so to speak.

Besides, part of the plan is to be able to finish the bike trip after someone steals it off of YOUR bike, like mine was... 

I don't mind showing the schematic and parts list stuff: maybe it will become "The SOHC4.us Ignition" !   
But, AFTER the testing is done - I don't want someone yelling at me because I didn't test something and they had to push their bike home...   

[Jonesy]

But, AFTER the testing is done - I don't want someone yelling at me because I didn't test something and they had to push their bike home...

But, I thought the nice thing about your system was that if it failed, you can can just reconnect the points to get home... Did I miss something? 

(Sorry, just had to bug ya!)

[HondaMan]

Yep, yep...but, if other's memory is like mine is getting, someone might forget that they can switch it back!

 

[Sam Green Racing]

snap

[Terry in Australia]

snap

Osteoporosis Sammy? Be careful mate, you're not getting any younger!

[Kevin400F]

Hondaman, what is your opinion of the circuit at the URL below?  It appears to be a bit more complicated that what you're planning.  Are all the extra components warranted?

http://www.cs.berkeley.edu/~wkahan/TransIgn.pdf

[Clutch Cargo]

Hondaman,

I'm in the electronic component business. Let me know what you are looking for - I may be able to order samples.

Mike

email -  mdurkin@windstream.net

[Raul CB750K1]

Raul & Hondaman, if I understand the logic, if no inverter logic is used the main transistor would open when the points close, and the coil fires when the main transistor opens.  Is it not true that the 2-3 point set closes at the exact point the 1-4 set opens?  If so, couldn't you connect the main transistor firing the 1-4 coil to the 2-3 point set, and avoid using an additional component to invert the logic?

Kevin

No, that is not true. The only thing you can guarantee is that the points open at the "F" mark, and that is when you want it to happen so the coil will generate the spark. Once the coil fires, you really don't care about when they close, because the only thing you want is that it will close soon enough to energize the coil before the next point opening. The ratio between points open and points closed is called "dwell" if I'm not wrong. You want the points closed long enough for the coil to energize, but short enough so not to overheating the coil.


Imagine you have piston #1 at the "F" mark. In that given moment, points for cylinder #1 are starting to open. That should be the moment, according to your theory, that the points for pistons #3 and #4 should close, so the transistor will invert the logic, open and fire cylinder #1. But you don't know where exactly the point closes. True, instead of adjusting the points at the "F" mark, we could make new marks when the points close, but in my opinion that's more complicated than adding a driving transistor, in Darlington configuration.

Somebody correct me if I'm wrong...

Raul

[OldSchool_IsCool]

HondaMan,

I've been reading the thread about the stolen bike in Atlanta, and it got me thinking (a dangerous thing!).  Wouldn't it be nice if there was a hidden push button switch  that had to be depressed to activate the coils at start-up.  Once running, the switch need not be pressed any longer.  The goal being theft prevention.  Sure, a toggle switch could be added, but the one time you forget to flip it off after your ride is the time some punk swipes your bike! 

I scratched out a diagram with a normally open push-button and a double-throw relay.  Then it hit me, one big bump and the relay would let loose and there goes your coil circuit and your power!

I'm curious if your solid-state circuit could also have an interlock circuit to allow for theft prevention.

[eldar]

I see one issue with something like this when applied to theft. Most people push the bikes away before trying to start them. Sounds likea good idea but hard to say if it would have much impact and may actually make the system more prone to falure.
But that does not mean that a person should not try it, cause it probably would be a good addition if it could be done with no bad effects.

[Kevin400F]

Raul, I took a 12V test light and checked the points open/close states on one of my spare engines.  The 1-4 point set opens at the same time the 2-3 set closes, then 180 degrees of crank rotation later the 1-4 point set closes at the same time the 2-3 set opens.  So, tell me again the definition of "inverted switching logic"?   If a circuit was designed to fire the 1-4 coil when the 2-3 point set closes, you still use look for the 1-4 timing mark if your timing light is in a #1 or #4 plug wire, you just move/adjust the 2-3 points to set the 1-4 timing. 

I'm not trying to beat a dead horse here, just hoping to better understand our options.

Kevin

[kaysystems]

HondaMan...I'd like one too please.

Thanks

David

[Raul CB750K1]

Raul, I took a 12V test light and checked the points open/close states on one of my spare engines.  The 1-4 point set opens at the same time the 2-3 set closes, then 180 degrees of crank rotation later the 1-4 point set closes at the same time the 2-3 set opens.  So, tell me again the definition of "inverted switching logic"?   If a circuit was designed to fire the 1-4 coil when the 2-3 point set closes, you still use look for the 1-4 timing mark if your timing light is in a #1 or #4 plug wire, you just move/adjust the 2-3 points to set the 1-4 timing. 

I'm not trying to beat a dead horse here, just hoping to better understand our options.

Kevin

If it works the way you say, then definitely you can use points 2-3 to trigger cylinders 1-4 and vice versa. My point was that you can't tell in advance that a breaker closes EXACTLY when the other one opens. The cam opens the breaker, but once the cam has rotated "far away", the breaker closes by the action of the leaf spring and then stays this way until the cam opens it again.

Needless to say, you have to make sure that the breakers are open EXACTLY 50% of the time. If the breakers are open 20% of the time, and closed the remaining 80% (let's take it as an example), you would be inverting the dwell too. That is, using the opposite breaker your coil would be energized 20% of the time, and de-energized 80% of the time. Nevertheless, if you are positive that one opens exactly when the other one closes, it could only mean that the dwell is exactly 50%.

You have made the static test, but it would be necessary to check that the advanced timing mechanism respects the 180º separation between opening-and-closing. I guess it would be as simple as to put the crankshaft in the "F" mark and adjust the opposite breaker to close in that moment, and do the same with the other cylinder. Then, start the engine and use the strobe light to make sure it works all the way up.


Raul


Raul

[Kevin400F]

Excellent point, Raul.  Although the points open/close events are exactly 180 degrees apart, the opening event is a "positive" action...the points are pushed open by the points cam, whereas the closing event depends on spring action.  In the OEM setup, if you get point "float" as the spring-loaded points are trying to keep up with an 11,000 rpm engine, the worst that happens is you sacrifice a little coil charge time.  If you were dependent on points closing to fire the ignition, there is the potential the timing would be retarded.   

Thanks, Kevin

[Raul CB750K1]

Kevin, I'm with you when it comes to questioning. There is nothing wrong with making oneself questions like: Would this work? If not, why not? Would there be a simpler/cheaper way? etc.

In this case, there are some peculiarities that lead me to think it will not work exactly the same. The advantage of using the opposite points are just to save a transistor. Disadvantages could be much worst, from retarding the ignition to running erratically. In any case, the excess in fuel consumption for a retarded ignition exceeds by far the cost of a transistor, so I would go with the "fully semiconductor inverted logic"


But, and this goes to Hondaman, instead of using a driving transistor, I would use a Schmidt hex inverter, like the CD40106. Can be powered directly to 12 volts, has open collector -so it can drive directly the final stage transistor, and having hysteresis, you avoid the possible rebounds of the points. And having six inverter in the same chip, you can put three of them in parallel to increase output current and reliability....


Raul

[750essess]

how bout somethin' like this http://www.st.com/stonline/books/pdf/docs/1073.pdf. looks to me like you would just need a 5v reg and run your points or pick up sensor to this. It should be noted that I just looked at this quickly and may have missed something lol

[HondaMan]

But, and this goes to Hondaman, instead of using a driving transistor, I would use a Schmidt hex inverter, like the CD40106. Can be powered directly to 12 volts, has open collector -so it can drive directly the final stage transistor, and having hysteresis, you avoid the possible rebounds of the points. And having six inverter in the same chip, you can put three of them in parallel to increase output current and reliability....

In the past, I've found that CMOS and HV ignition systems do not go well together, for a number of reasons. Heat, for one, causes CMOS to radically change its conductive rates (to very high output currents). Cold, below 0 degrees F, causes it to conduct with very high power supply currents, because CMOS, by its nature, connects Vss to Vdd momentarily during the switching time. When the CMOS gate gets cold, it switches slowly and the resulting surge currents thru the body of the switch slowly destroy the PN junctions until the device heats up to at least 10 degrees F. Motorola has some nice tutorials about this in their CMOS and COSMOS handbooks (maybe online, too, I haven't looked...), which generally say, "heat it up first or switch at low frequencies when cold". The CMOS glass inputs also erode when subjected to HV plasma that can travel across the circuit if, for example, the end user had a poor ground. Since this sort of thing is beyond the Designer's control (that's me in this case), it is better to go with a base technology that won't suffer from possible mis-installations. The "old fashoined" biploar transistor is about a rugged as a CB750 bottom end, and that's hard to beat.