CA18DET Trigger mod for stock ECU

[tommey]

http://www.vems.hu/wiki/index.php?page= ... sanTrigger
Any clever minds out there who could suggest a solution to my problem? I need to make a trigger that makes the same signal as the stock trigger (CAS), but it will be mounted at the crank wich spins twice as fast as the cams.
Any suggestions?

[sjbsuperman1425]

my suggestion..buy another CAS

[tommey]

my suggestion..buy another CAS

How would that solve anything?
the point is to eliminate the "give" from the timing belt wich gives unstable ignition wich over time leads to broken wrist pins.

[D_Stirls]

I have never heard of timing belt "give" causing issues. Timing belts don't really have any give and the little bit that might occur would result in less much than 1 degree change in timing.
There are a lot of Big power GTR's running in the 7 and 8 second mark that still use the optical CAS driven from the exhaust cam. If there was an issue i would be sure that these guys would be all over it especially since some of their engine cost well over $50,000.

I have heard that some of the wrist pins that come with certain brands of pistons are too thin in some people minds, and what has been happening is once you start leaning on the engine 600+wHP there is flex in the pin which flog out the small end bush.

[sjbsuperman1425]

if you really want to eliminate the CAS, just go Megasquirt or SDS and you're golden.

[float_6969]

I have never heard of timing belt "give" causing issues. Timing belts don't really have any give and the little bit that might occur would result in less much than 1 degree change in timing.
There are a lot of Big power GTR's running in the 7 and 8 second mark that still use the optical CAS driven from the exhaust cam. If there was an issue i would be sure that these guys would be all over it especially since some of their engine cost well over $50,000.

I have heard that some of the wrist pins that come with certain brands of pistons are too thin in some people minds, and what has been happening is once you start leaning on the engine 600+wHP there is flex in the pin which flog out the small end bush.

[r34 gtr]

Its a really simple device, and if you have a little electrical know-how, it wouldn't be too difficult to build something that ran off the crank, but sent the same 0-5v signals to the ECU. basically the device uses two LEDs and two optical sensors and a spinning disc. It works the same way the sensors used in the new steer-by-wire cars work (they are pretty much the same thing). Basically all you need is a 0-5v signal sent to the ECU to tell it each degree of crankshaft rotation, and a 0-5v signal telling the ECU when each cylinder is at the top of its stroke. You can send these signals from any sensor, so long as they are received by the ECU in the same way the original signals were received.

The electronics on a 21yr old car are not scary or complicated. The vacuum systems however, are a dark, dark place.

[D_Stirls]

I have never heard of timing belt "give" causing issues. Timing belts don't really have any give and the little bit that might occur would result in less much than 1 degree change in timing.
There are a lot of Big power GTR's running in the 7 and 8 second mark that still use the optical CAS driven from the exhaust cam. If there was an issue i would be sure that these guys would be all over it especially since some of their engine cost well over $50,000.

I have heard that some of the wrist pins that come with certain brands of pistons are too thin in some people minds, and what has been happening is once you start leaning on the engine 600+wHP there is flex in the pin which flog out the small end bush.

Whys that?

Kier Wilson GTR OS-Giken 3 litre
1100 AWHP


Terry Cook GTR
1171 AWHP


Godzilla Motorsport GTR
1200 AWHP


HKS DRAG GTR
1300 AWHP 7.69sec


If all these cars have no problem with running their timing signal from source based off the timing belt then no CA will either. If your worried then pay some more coin on you timing belt and get one that's better quality.

But really you best option is to run an ECU that can use a crank trigger setup if you must change you ignition signal source.

[tommey]

Its a really simple device, and if you have a little electrical know-how, it wouldn't be too difficult to build something that ran off the crank, but sent the same 0-5v signals to the ECU. basically the device uses two LEDs and two optical sensors and a spinning disc. It works the same way the sensors used in the new steer-by-wire cars work (they are pretty much the same thing). Basically all you need is a 0-5v signal sent to the ECU to tell it each degree of crankshaft rotation, and a 0-5v signal telling the ECU when each cylinder is at the top of its stroke. You can send these signals from any sensor, so long as they are received by the ECU in the same way the original signals were received.

The electronics on a 21yr old car are not scary or complicated. The vacuum systems however, are a dark, dark place.

The problem is getting the signal to be correct because the crank spins twice as fast as the camshaft.
And to you people saying its not a problem, you are wrong. over time it is a problem.
Even the ca18det stage 3 chip cars with 540 flywheel hp are struggling with this.

[sjbsuperman1425]

MEGASQUIRT of SDS..PICK ONE! I'm not going to argue with you and these are for sure way to eliminate the CAS. Megasquirt can be had quite cheap too!

[tommey]

I know they are cheap and very good. But they are not an alternative.
I need to use the stock ECU.

[sjbsuperman1425]

why do you NEED to use the stock ECU?

[tommey]

Because it is a streetcar and it is illegal to use anything else. And i love the stock ecu.

[sjbsuperman1425]

If you live in Cali, i can understand..but anywhere else, just "eliminate" it the right way..

[float_6969]

I don't think he's in the USA. I could be wrong though.

As for the CAS, my guess (and this is just a shot in the dark) is that maybe the problem is related to the 8bit ecu, and not the CAS itself. I'm really having a lot of trouble understanding the relationship between HP and the CAS signal? Do you mean the increased revs? Stronger valve springs causing pre-mature belt stretch? If it's the former, then your limitation is the computing ability of the 8 bit ECU you're working with and not the CAS. If it's the latter, then once again, a stronger belt should resolve that issue.

I'm not saying that you're wrong, just that the reason's you've given for the CAS to be an issue don't make sense.

[tommey]

I don't think he's in the USA. I could be wrong though.

As for the CAS, my guess (and this is just a shot in the dark) is that maybe the problem is related to the 8bit ecu, and not the CAS itself. I'm really having a lot of trouble understanding the relationship between HP and the CAS signal? Do you mean the increased revs? Stronger valve springs causing pre-mature belt stretch? If it's the former, then your limitation is the computing ability of the 8 bit ECU you're working with and not the CAS. If it's the latter, then once again, a stronger belt should resolve that issue.

I'm not saying that you're wrong, just that the reason's you've given for the CAS to be an issue don't make sense.

No this is not because of the horsepower, or the revs ( i dont think the cas has any problems with revs). it happens because the belt flexes sometimes under heavy load and for just a brief moment the igintion gets screwed up. You could just run a "safer" map, but thats no fun is it?
I have considered a stronger belt and to tighten it little more than what is "healty", but i am not totally comfortable with that. if i could mount the trigger on the crankshaft i would eliminate the problem totally.
I could laser cut a bigger diameter version of the stock CAS wheel and use the original CAS parts on the crank. but then again it would still go twice as fast giving the wrong signal to the ecu.
Btw, i am from Norway.

[float_6969]

it happens because the belt flexes sometimes under heavy load and for just a brief moment the igintion gets screwed up.

OK, so if the problem is supposedly with the belt flexing, why would you go to all the trouble of eliminating the CAS and not deal with the belt stretching? And if you say a stronger belt doesn't work I'm gonna flip out, LOL.

And when I say a stronger belt, I mean these guys, http://www.powerenterpriseusa.net/produ ... g_belt.htm. THESE BELTS DO NOT STRETCH.

[float_6969]

And something else I just thought of is that if you guys seem to think the belt stretching is the problem, why would you go to all the work to just eliminate the CAS/timing issue, when you're going to have problems with the camshaft timing as well?

And BTW, I'm still having trouble understanding why "heavy load" makes the belt stretch. The load on the belt is constant. The springs that create the resistance to movement don't get harder compress when you have more RPMS or more HP. So, unless the valve springs are changed, the load on the belt is constant.

[r34 gtr]

The fact that the crank spins twice as slow as the camshaft is completely irrelevant. You are just going to trick the ECU anyway. Just make a trigger wheel with 180 slits in it, and two bigger holes. Then make you a little board that sends the proper 360 degree signal to the ECU every two crank revolutions, and have it fire the #1 and #3 cylinders the first revolution, and #4 and #2 cylinders the second revolution. It IS easy. I'm discussing this with my boss right now (PhD in optical engineering) and he was like "oh yeah, Ive build a thousand of those things! Want one?"

[tommey]

And something else I just thought of is that if you guys seem to think the belt stretching is the problem, why would you go to all the work to just eliminate the CAS/timing issue, when you're going to have problems with the camshaft timing as well?

And BTW, I'm still having trouble understanding why "heavy load" makes the belt stretch. The load on the belt is constant. The springs that create the resistance to movement don't get harder compress when you have more RPMS or more HP. So, unless the valve springs are changed, the load on the belt is constant.

Its not a big deal if the timing is a little bit off, that wont kill the engine.
What i meant about heavy load is that the belt stretches more when the speed(rpm) changes fast and the belt can`t keep up.
Yes i have been thinking of kevlar belt, that would probably do the trick. But i want to make sure there is no way the ignition can be off.

[tommey]

The fact that the crank spins twice as slow as the camshaft is completely irrelevant. You are just going to trick the ECU anyway. Just make a trigger wheel with 180 slits in it, and two bigger holes. Then make you a little board that sends the proper 360 degree signal to the ECU every two crank revolutions, and have it fire the #1 and #3 cylinders the first revolution, and #4 and #2 cylinders the second revolution. It IS easy. I'm discussing this with my boss right now (PhD in optical engineering) and he was like "oh yeah, Ive build a thousand of those things! Want one?"

Yes it is that "little board" i need help making/ getting. I am not good with electronics.
It would solve most of this problem. no problem making the triggerwheel or sensor brackets. just the electronics that "compensate" for the double speed.

[float_6969]

If you're just wanting to be different, then I completely understand, but if there is a product readily available (the kevlar belt) that WILL resolve your issues, why would you consider any other option, except to be different?

[tommey]

I`m not wanting to be different I talked to a engine builder today and he said that a kevlar belt will NOT solve much of the problem.

[D_Stirls]

Here's another one that doesn't have a problem with an optic CAS signal driven from the timing belt.

Jun DRAG S13
840ps

http://www.syj.jp/~null/0-400/pic.html


[youtube]http://www.youtube.com/watch?v=aP7zP1idmt4[/youtube]

[tommey]

Omg.. it`s not hard to make big numbers with the CAS, i am talking about durability!

[r34 gtr]

Tommey,

I agree the CAS can wear and become unreliable. It can also be expensive to replace. However, if you want to get rid of it, you are either going to have to build the contraption I mentioned earlier, or use a different engine management system that will let you use a 36-1 trigger wheel on the crank. I cannot help you build the system I mentioned earlier, as I simply do not have the time. However, it really isn't too complicated, and you could probably do it yourself with a little research and spare time.

Edit: Also that video is awesome! Check out all the nitrous on that little CA18!

[tineira]

You can't simulate the stock CAS signal from the crank. You can't synchronize the engine correctly as you can't tell TDC on cylinder 1 from TDC on cylinder 4, as you can from the cam.

Also what makes you think that cambelt stretches? I think it's more probable that you are using the timing light incorrectly. You will have a hard time measuring ignition drift from belt stretching.

[tommey]

This is a very common problem, its not recomended to stick to the CAS.
I went Vi-pec v44 plugin card with 24 teeth on the crank and 1 for sync on the camshaft.
Even Ray Hall (one of the Vi-pec dudes) says the CAS is inferior.

[boost_boy]

This is a very common problem, its not recomended to stick to the CAS.
I went Vi-pec v44 plugin card with 24 teeth on the crank and 1 for sync on the camshaft.
Even Ray Hall (one of the Vi-pec dudes) says the CAS is inferior.

So is your new set-up operational? Any pics for the CA family to see?

[tommey]

Not operational at the moment, but i have a pic of the modified crankpulley