MS extra siamese sequential tune (msq) to start with
Turbo Mini Forums

in case its of any use whatsoever will ,

on my bike i had to create 2 seperate ignition systems , i used a magnetic pickup ,and a 4 pin gm module for each system ,the module creating a bigger signal for the cop .

Medusa + injection = too much torque for the dyno ..https://youtu.be/qg5o0_tJxYM

I have just finished testing the distributor on the bench. It must have something to do with the internal sensitivity being compromised by reducing the number of static blades to one. Normally four rotating blades are passing the four static ones at the same moment giving a probably a stronger signal. I tried a normal coil as well as a resister of 470 ohm as load and both are giving the same results, a very clean pulse of about 7 msec below 600 distr. Rpm and above 700 rpm I only see spikes of a few usec. I wil try to solve the problem by making both the static and the rotating blade four times as big.

Will

Will,

May be worth just buying a Hall switch or Opto-switch and modifying the dizzy to suit ???

Both are known to work, your proposal may not....

No offence intended, experimentation is what this is all about.....

Schrödinger's cat - so which one am I ???

So far for using the electronic distributor as a cam sensor. Just snapping of three of the sensor blades and enlarging the left over one does not function at higher RPM either. It seemed,at first, to be a quick and easy way to produce a cam signal. I will follow the advice and build an opto switch inside the distributor, I have an old graycode shaft encoder laying around wich has an opto switch inside.
Has anybody ever tried modifying the mechanical fuel pump to serve as a cam positioning sensor?? It is nicely hidden behind the engine. I do not like the sight of an old stripped distributor at the front of the engine very much.

Regards Will

Edited by Rod S on 25th Dec, 2009.

Schrödinger's cat - so which one am I ???

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

I'm just wondering if the fuel pump lobe or the arm in a modified fuel pump is going to give a clean signal due to the sinusoidal nature of the movement.

It's not like a vane passing a hall or opto switch with a distinct on/off. It could be a bit noisy.

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

The lobe itself (with say a Hall switch looking at it) would probably not be a clean switch for the reasons you say but we've already agreed absolute positional accuracy isn't too important.
The arm in a modified pump could be as you could put the switch point a long way away from the point of the lobe (say half travel once it starts to move). That would give far longer on to off time compared to our current Hall/opto arrangements but as we have the choice of rising or falling edge in the software it shouldn't matter so long as we pick the correct one.

With the actual cam lobe, what about drilling/tapping it and screwing/tack welding in a square headed steel bolt and machining it to a radius off the cam centreline, or even just milling the pump lobe to a very short "square" tip (if you could break through the surface hardness).

EDIT - or even just use a plain VR sensor against the lobe, just because Rover used a fancy one running through a slot in the lobe, doesn't mean a normal sensor won't work.... the VR waveform off a toothed wheel is distinctly sinusoidal (look at my plot previous page) even though the teeth on the wheel are absolutely square. I'm sure a suitable conditioning circuit, say Jean's but with slightly different value components, would work fine on the pump lobe profile at low speeds.

EDIT 2 - thinking about it a bit more, it probably wouldn't even need a VR conditioner circuit as it would only be looking for the lobe peak once per (cam) rev, just a simple trigger circuit to square out the simple sine wave rather than find the missing tooth in a complex sine wave.

EDIT 3 - something as simple as this ???

http://www.electronics-lab.com/projects/os...ors_timers/003/

Only 8 components.

Edited by Rod S on 27th Dec, 2009.

Schrödinger's cat - so which one am I ???

I've just dismantled on old mech fuel pump.

The cam lobe moves the fuel pump arm about 5 to 6mm. The arm has an approximate ratio of 1:2 meaning that the useable end of the arm only moves around 2 to 3mm.

I don't think that there is enough movement for my Hall sensor :(

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

Schrödinger's cat - so which one am I ???

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

That looks good if it will go through the hole and can be mounted accurately for cam/sensor gap.

And if the waveform is still too sinusoidal, any of the conditioning circuits should work to square it off.

EDIT - ignore that last comment, it's a Hall switch so it will either switch or it won't, the conditioning is only required for a VR - end EDIT

Just out of interest, I've just looked at the datasheet for the opto-switch I use and it's optical apperture is only 0.05" (1.3mm) so may be just possible to work on the arm if set VERY accurately.

But I think something inside looking direct at the cam, rather than any mechanical linkage, has to be better.

Edited by Rod S on 31st Dec, 2009.

Schrödinger's cat - so which one am I ???

Actually that one will not go through the hole.

This would be easier to mount:

http://www.hamlin.com/specsheets/55075%20IssueAF.pdf

although it a bit short.

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

Looks better.

And the spec sheet says a 10bit A-D converter built in so it should switch cleanly even on a curved edge.

You should easily be able to fabricate an extension tube to the flange but beware of the thread size, it's not the normal M12 ISO and not even the normal M12 fine.....

Schrödinger's cat - so which one am I ???

I was thinking off welding one of the nuts to a bit of stainless tube, then weld the tube at an angle to a plate to cover the pump mounting hole.

Otherwise I would have to get a M12*1 tap, but then I could make it in aluminium.

I think I'll order a couple of sensors. Digikey have some in stock.

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

The only thing that concerns me is the lack of adjustment.

I'm hoping that some adjustment of the gap and selecting either leading or trailing edge will allow some movement of the timing point.

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

Go for it...

The only thing with welding one of the supplied nuts onto a tube is it might shrink a bit so still need a tap running through after welding.

I certainly had to after welding lambda bungs on (both stainless and mild steel) so I bought an M18 X 1.5 tap even though it has only been used four times so far just to clean up threads after welding......

M12 X 1 is a strange size though.

Schrödinger's cat - so which one am I ???

I can get a set of M12 * 1.0mm taps from RDG tools for about £7, so I think I'll do an alloy mount.

I've just had a gander at the cam. The fuel pump lobe peak is about 180 cam degrees after No. 1 inlet starts to open (approx. 30 degrees BTDC).

I think that will put the sensor switching points well out of the way ?????

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

Schrödinger's cat - so which one am I ???

Just thought, the other thing to bear in mind is that the hall switch will not switch at the peak of the cam lobe, but some point before depending on the gap, so it's the actual switching point you need to keep away from the missing tooth.

Schrödinger's cat - so which one am I ???

I'm pretty sure that if I set it at falling edge, then it will be BTDC.

It depends how large the gap between the sensor and the lobe gets before it switches.

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

I have just finished building a new cam sensor using an optical switch inside the distr. I will still try the fuel pump lobe when I’ve got the right parts for it later on.
The afr signal of cyl 1_4 (green) is more unstable than the signal of cyl 2_3 (red) this is so throughout the entire RPM range. At low RPM i.e.<2000 cyl 1-4 goes lean now and then but I suppose that has to do with my crank. At this moment injection timing is set at 50 degrees fixed.
I have added a few plots at different RPM’s. Next I try and see what moving of the injction timing does before I start with the tables.

regards Will

Attachments:

10.41kb

9.73kb

9.96kb

10.83kb

Will,

You seem to have a good setting there for high MAP values.

I would expect a lower timing figure will improve the idle.

However, once you get under load and the fuel pulses double in length, you will need to set up the tables.

EDIT: Also, you do not want to take too much notice of readings below 85 Deg C coolant temperature as it has a dramatic affect on the mixture distribution.

Edited by Paul S on 31st Dec, 2009.

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

Will,

Excellent news.

Can you get a scale on the y axis of the graphs to help reading them ???

Otherwise, I'm pleased you went the Opto route in the dizzy or my cat would have been very disappointed !!!

But if the fuel pump hole is free it makes sense to put a Hall or VR sensor there if it can be made to work.

Schrödinger's cat - so which one am I ???