2nd installation : No spark

alexander · Post by **alexander** » Wed Apr 09, 2008 10:19 am

hi patriq,
i would be more than happy to send you a slice of the gearstock. i have one sitting on the desk in front of me.
first, however, can you tell me the diameter of the shaft of your distributor? the piece i have already cut off, has been bored to approximately 8mm, so your shaft would need to be bigger than than, to allow for it to be bored a bit more. if for some reason your distributor shaft is smaller than 8mm, i will have another piece cut off. either way, PM me your address and i will send it asap.

re the sensor, i would happily supply my spare, but i am about to use it to convert my jeep to a megajolt too! so sorry to say you will have to find that yourself. FC series RX7 crank angle sensor. it looked like a distributor body, but with a steel plate bolted to the top of it instead of a distributor cap.

a bit of technical stuff: i dismantled the distributor completely, and took the shaft and gear stock to an engine reconditioner. i asked them to cut a slice off, and bore it for a light interference fit on the shaft, and left that to their discretion. after the idiot snapped my distributor shaft trying to hammer it on (eek!), i got another distributor and used a hydraulic press to fit. whoever bores the thing just needs to make sure the hole is dead centre as there isnt much tolerance for runout. i have the sensor no more than 1mm from the teeth.

to remove the teeth, i used a cheap dremel copy, with a tiny sanding disc ie about 10mm in diameter. it was surprisingly easy! i used 2 or maybe 3 little sanding discs to remove two teeth.

photos below show: my wheel, and the trigger wheels from the mazda part; a photo of the mazda CAS - those part numbers will guarantee you get the right part; my wheel press fit on my dist shaft; the mazda sensor mounted on a plate i hand made, and ready to be fixed into the distributor with the original base plate screws.

so, measure the diameter of your distributor shaft, and send me that with your address and i will get a piece of gear into the post. lastly, if for some reason you cant find the mazda part in norway, i am sure i could track down another one at a motor wrecker here in australia. after all, i am sure i will find a use for another spare sensor

regards
alexander.

alexander · Post by **alexander** » Wed Apr 09, 2008 10:27 am

and the other photo i mentioned..

Patriq · Post by **Patriq** » Wed Apr 09, 2008 11:07 am

That's very kind of you !

I have a dremel, (actually bought exclusively to slightly widen a hole on the panel to fit a new cigarett lighter on the car I'm jolt'in) - so than that's the route I'll go too.
I'll have a look at my dizzy and pm you later. Just give me all the costs you have and I'll transfer money to you. Maybe paypal ?
I'm pretty sure I'll be able to source a sensor.

Regards,
/Patriq

Patriq · Post by **Patriq** » Wed Apr 09, 2008 11:30 am

How on earth did you by the way figure out that this was the sensor to use ?!

alexander · Post by **alexander** » Wed Apr 09, 2008 1:46 pm

Patriq wrote:How on earth did you by the way figure out that this was the sensor to use ?!

the question of alternative trigger wheels came up on the megasquirt discussion forum on Yahoo. someone on that group said that he had made this setup, and was kind enough to tell me the exact parts he used. how HE knew, i dont know.

alexander.

Patriq · Post by **Patriq** » Mon Apr 21, 2008 7:24 am

Progress !

i figured, that if the voltage increased that much just by moving a hex bolt such a small amount closer to the sensor when spinning it on a drill - then maybe the sensor was to far away from the trigger wheel after all.
So I modified it slightly to let it come even closer.

Bingo - spark !

(and by the way - I had soldered the cable correct the first time..... the polarity was correct when the same colour on the sensor cable and the EDIS-module cable was matched. But maybe on the installtion page it should include the same small images as on msextra.com, that shows what is ve+ on the standard Ford sensor and on the connector to it. Then it is very easy to see if the cable has been soldered correctly)

Well, so now I have spark from the EDIS unit - triggered by the standard Ford sensor to the 72-2 trigger wheel I made. But....
It only gives spark on average on every third event. I'm suspecting this is due to the teeth being to small for the standard Ford sensor. So next step is to try with the same type of sensor that Alexander has been so kind to send me !
First I was thinking that when cranking the speed of the wheel didn't generate strong enough signal. But then I realized, I was thinking wrong.

Well, if anyone has comments on my thinking - please comment.

alexander · Post by **alexander** » Mon Apr 21, 2008 8:10 am

patriq.. got your PM; thank you.

as a matter of interest, what is the gap now, between the sensor and the teeth?
i initially had intermittent spark when bench testing, and also found that it became regular by moving the sensor closer. ultimately in my setup, the sensor is <1mm from the (very little) teeth. so note that when the other part arrives in the mail!
(when it does, the green wire is the +ve). are you able to move your sensor even closer? after all, it only needs to far enough away to not touch, and if that is 0.1mm then that is fine. ultimately the proximity is limited by runout in the wheel or play in the shaft.

i think this has come up before, maybe in this thread (without going back over it all), but the output voltage IS proportional to the speed at which the teeth pass by the sensor. so if the signal produced is only just enough to trigger the EDIS, then it is indeed quite possible for it to work at engine speed, and not a cranking speed.

Brent: are you reading this? i think you made reference somewhere to benchtesting with a square wave generator instead of a triggerwheel/vr sensor. did i understand that correctly? IF the EDIS can be triggered with a square wave, that should mean that Hall type sensors would also work instead. they have many advantages ie i think there are a lot more available new from suppliers; their output is independant of speed; there are very small ones available (eg google or ebay part number UGN3503U); hall sensors are much less susceptible to EMI. could be interesting...

regards
alexander.

MartinM · Post by **MartinM** » Mon Apr 21, 2008 8:27 am

There's no reason why a square wave (with a missing pulse every 36th time) wouldn't work - as long as it has a -ve voltage component to it i.e not just a 0v to +v range.

See the following, where it's already been proven...
http://www.heggs.co.uk/vrssim/index.htm

alexander · Post by **alexander** » Mon Apr 21, 2008 8:47 am

martin.....good information!

Patriq · Post by **Patriq** » Mon Apr 21, 2008 8:59 am

alexander wrote:patriq.. got your PM; thank you.

as a matter of interest, what is the gap now, between the sensor and the teeth?

It was something like 1.5mm I guess. I haven't measured - but I would guess it is 0.1-0.3mm now? I can try - but I don't think I'll get it much closer than another 0.1mm. There's fairly little play between the wheel and the sensor. So the wheel is nicely centered on the shaft.

i initially had intermittent spark when bench testing, and also found that it became regular by moving the sensor closer. ultimately in my setup, the sensor is <1mm from the (very little) teeth. so note that when the other part arrives in the mail!
(when it does, the green wire is the +ve). are you able to move your sensor even closer? after all, it only needs to far enough away to not touch, and if that is 0.1mm then that is fine. ultimately the proximity is limited by runout in the wheel or play in the shaft.

Ok, good to know which is +ve !

i think this has come up before, maybe in this thread (without going back over it all), but the output voltage IS proportional to the speed at which the teeth pass by the sensor. so if the signal produced is only just enough to trigger the EDIS, then it is indeed quite possible for it to work at engine speed, and not a cranking speed.

Hm, yes - I was aware of that the voltage changed with speed - but signal strength ?
Could this mean that I one could have problems with cranking - that is starting ? So you can't get the darned thing to start - but had it started it would have run nicely !?

Brent: are you reading this? i think you made reference somewhere to benchtesting with a square wave generator instead of a triggerwheel/vr sensor. did i understand that correctly? IF the EDIS can be triggered with a square wave, that should mean that Hall type sensors would also work instead. they have many advantages ie i think there are a lot more available new from suppliers; their output is independant of speed; there are very small ones available (eg google or ebay part number UGN3503U); hall sensors are much less susceptible to EMI. could be interesting...

regards
alexander.

alexander · Post by **alexander** » Mon Apr 21, 2008 10:20 am

hmmm .. despite what i said, i my intuition says that <0.5mm is close enough!

and.. signal strength IS voltage.

regards
alexander.

MartinM · Post by **MartinM** » Mon Apr 21, 2008 10:25 am

...here are very small ones available (eg google or ebay part number UGN3503U); ...

I didn't realise how small! I've just read the Allegro data sheet and the schematics of having a magnet on the back, with the trigger wheel in close proximity look fine....until you realise they're not to scale, with the sensor being less than 0.2" square and having three leads coming out of it across a total distance of 0.1"

Patriq · Post by **Patriq** » Mon Apr 21, 2008 11:01 am

alexander wrote:hmmm .. despite what i said, i my intuition says that <0.5mm is close enough!

and.. signal strength IS voltage.

regards
alexander.

Hi

Yes - voltage is the potential, right ? So yes - it's strength. But - I was more in terms of readability for the EDIS unit. Does the signal then get stronger and more readable when rpm increases ?
I thought that amount of electricity - that's ampere right?, was the same. And that the ampere was the same, regardless of rpm - but that as rpm increased, at a certain point the EDIS unit wouldn't anymore be able to distinguish between the teeth.

The voltage in this circumstance was for me just a sign of a waveform coming from the sensor. And that the voltage off course would increase with increased rpm. But not that the signal in a sense would be stronger just because rpm increase - but rather that the signal strenght was more relaying on the distance between the sensor and the teeth.
But I guess I'm wrong here ?

(Do I make myself transparent....

)

Post by **brentp** » Mon Apr 21, 2008 3:57 pm

alexander wrote:patriq.. got your PM; thank you.

Brent: are you reading this? i think you made reference somewhere to benchtesting with a square wave generator instead of a triggerwheel/vr sensor. did i understand that correctly? IF the EDIS can be triggered with a square wave, that should mean that Hall type sensors would also work instead. they have many advantages ie i think there are a lot more available new from suppliers; their output is independant of speed; there are very small ones available (eg google or ebay part number UGN3503U); hall sensors are much less susceptible to EMI. could be interesting...

Yes, as Martin pointed out, the EDIS module would require the -ve part of the signal, but it should work with a simple interface circuit. Our current test bench has a simulator running from a spare controller, so the EDIS appears to be happy receiving a square wave.

alexander · Post by **alexander** » Mon Apr 21, 2008 5:21 pm

Patriq wrote:
Hi

Yes - voltage is the potential, right ? So yes - it's strength. But - I was more in terms of readability for the EDIS unit. Does the signal then get stronger and more readable when rpm increases ?
I thought that amount of electricity - that's ampere right?, was the same. And that the ampere was the same, regardless of rpm - but that as rpm increased, at a certain point the EDIS unit wouldn't anymore be able to distinguish between the teeth.

The voltage in this circumstance was for me just a sign of a waveform coming from the sensor. And that the voltage off course would increase with increased rpm. But not that the signal in a sense would be stronger just because rpm increase - but rather that the signal strenght was more relaying on the distance between the sensor and the teeth.
But I guess I'm wrong here ?

(Do I make myself transparent.... )

ok patriq, let's get technical here

there are 3 things to consider: voltage, current, and frequency.
textbook definitions are that (as you note) voltage is 'potential difference', also referred to as electrical pressure. yet another way of expressing it is the force acting on electrons to make them flow. in our case, the voltage induced in the sensor increases with wheel speed (to some finite point, no doubt).

current is the amount of electricity flowing, and if you look at its definition in terms of physics, it is actually defined by the number of electrons flowing past a point, per unit of time. the current flowing at any given moment, is proportional to the voltage, so, no, current is not constant: it varies increases, decreases and reverses direction along with the change in the voltage.

signal strength is not a distinct thing in itself. in our case, if you are talking about signal strength, you are talking about the voltage. you cannot consider signal strength as something different to the above. i think you are using the term "signal strength" to mean the quality of the waveform. that will break down at some upper point, but see my next paragraph.

i think that the upper limit of operation with this sort of system is set by the frequency. i expect the sensor, the EDIS and the MJLJ can do their thing at much higher frequencies than can be produced by a rotating engine. see here: http://trigger-wheels.com/store/index1.html the trigger-wheels.com guys show a perfectly clean VR signal produced at 26,000 rpm. i think the real limit is set by the coils. it definitely takes a significant amount of time to charge the coils, so as the frequency of the signal increases, there is simply not enough time to charge them to a point where they can fire a spark plug when they discharge.

anyone should feel free to correct or add to the above.

regards
alexander.