MS master minds needed
 

So as most of you know I'm doing a sr20 swap in my miata and I would love to run ms on it also. After talking to Matt bit he said the cas on the sr20 won't work with the ms. So now I'm wondering could I just do a crank trigger wheel and not use the cas?

I can't imagine why not. With a generic crank trigger wheel (such as a 36-1) you can run a megasquirt on damn near anything that has spark plugs, fuel injectors, and an even number of cylinders.

I'm not all that familiar with the SR20 engine. Is it one of those weirdos that runs a CAS pattern giving something like 473 pulses per revolution on CKP?

Jmscortina has been working on some nissan cas stuff for ms3, you might wanna read about that.

http://www.msextra.com/doc/ms3/nissancas.html

Matt was telling me there working on some firmware to make the cas work and I'm not sure why the cas don't work. I was also told people swap the wheels for the one out of the dsm.

can the nissan cas be modified to produce a known sync pulse?

Here is what Matt said to me on mx5atlanta.com

There currently isn't any firmware for the stock CAS, although there's some experimental code for the MS3 that needs to be tested out. Most installs have used an AEM or DSM trigger disc in the stock CAS

I finally found an image of one, and I'd have to say that the answer sounds like a yes.

Here's the stocker:

http://www.supercars.net/gallery/132464/1497/645195.jpg

360 pulses per revolution on the main channel. Does that poor ECU have time to do anything other than receive primary interrupts?

I see two possible courses of action. You could knock out the majority of the holes in the outer ring, and apply a patch over others, to bring it down to a more reasonable pattern. Four evenly-spaced slots would suffice. Patch over the three small slots on the inner ring and you have a nice, easy 4+1 dual-output CAS.

Might now be much space available to do the patching, though. The optical elements have got to be pretty closely spaced if they're reliably reading that outer pattern.

Or you could do what some enterprising enthusiasts have done and get a new trigger plate. 24+1 seems to be popular, judging from a quick google search:

http://200sx.kicks-ass.net/niclas_ab...950189bed8.jpg

http://www.vems.co.uk/VEMSTrigger/newdisk.jpg

http://www.supercars.net/gallery/132464/1497/646979.jpg

It would appear that AEM maketh one, too:

http://www.supercars.net/gallery/132464/1497/899154.jpg

As does VEMS:

http://www.vems.co.uk/VEMSInstalls/N...sertIsland.jpg


This is what it would sound like if If Leon Redbone Suffered a Major Head Injury.

Joe you must have the holy Grail of ems info. This is great info you just sent me.

I must ask what did you type into google.

Thanks Joe I never knew how to use google, just amazing lol. Anyways I am liking this ideal you have here.


" You could knock out the majority of the holes in the outer ring, and apply a patch over others, to bring it down to a more reasonable pattern. Four evenly-spaced slots would suffice. Patch over the three small slots on the inner ring and you have a nice, easy 4+1 dual-output CAS." Joe

So all I do is make 4 good size slots that are the same size and evenly spaced to make it work?

http://img447.imageshack.us/img447/6876/cas3oy.jpg



Thinking this might help.

Hehe.

Actually, I just went to google images and typed in SR20 CAS. The first page was filled with images of non-standard CAS wheels, so I'm guessing this is a pretty common issue.


That, and also block off all but one of the four CMP slots. If it were my car, I'd also want to add some material between gaps in the outer ring, to make sure that the CPU has time to detect the transition to "off" before the next transition to "on". Whether this is a real problem or just an academic issue I can't say.

The only other concern is how to go about blocking the slots. From the image miatamike203 posted, you can get a feel for how tight the fit is in the gap between the LED and the phototransisor. Anything that adds significant thickness to the disc may prevent it from passing through this channel. You're not going to just glob some JB-weld on there. Foil tape might work.

I was thinking of filling the holes with bondo then sanding down the high spots.

Oh other question is what hole out of the 4 holes should I keep? I'm sorry this is the point where I'm lost on everything and how it would be set on ms.

After looking at the diy KA wright up it looks like the ka uses the same cas as the sr.

http://www.diyautotune.com/tech_arti...ssan_240sx.htm

I also could do what aem did with the 24+1 by taking out 15 slots then skipping 15 slots. It would give me the same out comes as aem has. Just would need to fill the 15 slots I left in place. Should work right?

Wish I saw this thread earlier...I probably could have helped you out.
You can change the wheels out pretty easily as people mentioned. We ran MS on my friends older VE30E Maxima that had a similar setup. The inner and outer holes use different sensors so you can do pretty much whatever you want when it comes to a trigger wheel.

I was about to start getting into MS right before I crashed my second 240.

So after spending a good amount of time looking into running ms on my sr I decided to go with the crank trigger wheel. I do have a few questions tho. What is the best cps to use? Next is should I set up the ms2 for edis or dis? The sr uses 4 cops and an ignitor so I'm guessing dis for wastespark? Then I'll just need to figure out the needed coil dwell.

Maybe this is just my ignorance showing through, but that article doesn't make any sense to me.

Basically, they are saying to leave the wheel completely unmodified, ignore the outer ring completely, and trigger on the trailing edges of the inner ring (which are evenly spaced.)

In that configuration, the only thing that the ECU would be able to infer from any given pulse is that "something is ready to happen." In other words, it has no way of distinguishing one cylinder from another. Forget about wasted spark, you'd have to use a distributor.

Did these engine have both distributors AND cam-angle sensors?



That's basically what I'm describing. You could go all the way down to four slots if you wanted to, but doing the +15-15 routine and winding up with a 24 hole disc (equivalent to a 12 tooth crankwheel) would be better.

You still would need to do the same routine on the inner ring, leaving only one hole intact, as you need some kind of reference to judge absolute cam position.

Well, actually I take that back. Since this wheel is turning at cam speed, you could do a "missing" tooth and still be able to run sequential, but I think it'd be better to run twin-signal. Less ambiguous.



Ok, wait. Are you the same guy who started this thread? :D

A crankwheel, if properly done, is always superior. More precise, since even timing chains have some slop. If you go the crankwheel route, you will still need a 1-pulse-per-rev cam reference if you want to run full sequential. (You could still wire for semi-sequential fuel and wasted-spark ignition if you needed to, but the cam sensor will allow you to avoid this.)

Avoid EDIS. It will give your dog herpes.

The best CPS to use is the one that fits. Here is one example of a completely custom crankwheel setup: https://www.miataturbo.net/megasquirt-18/wheel-timing-success-last-20241/ (Other have done theirs differently. I pass no judgement here. I will add that on mine, it took quite a bit of fiddling to get it all just right.)

Thanks for the link very helpful.

Yes. More common than you may think. The write up for using the 4 tooth wheel assumes a factory distributor motor.

MS3 is expected to work with the Nissan 360 tooth wheel, I just don't know anyone who has tried it.

So i just got home and i went and looks over your post joe. So now i do have a few more questions. The first one is about setting up the ms2 im guess to use the wright up diy has for the ms2 for dis ignition systems. Then the next one is do i need the two bip373 as i have an ignitor for the stock coils?

Ok, that's the preface to the question. What's the question?


No, no, no. You need to completely ignore that link you posted to the DIY site unless you actually plan on using a distributor. That's what the bip373 component is for; it's a high-current switch intended to directly drive one ignition coil without an "igniter". That writeup, and everything in it, is totally irrelevant to running an engine with distributorless ignition, be it wasted-spark or COP.

For a situation like yours, where you are using DIS coils and a factory or aftermarket igniter (be it external or integrated into the coils) you'll build your ignition drivers just like you would for a Miata.

But more importantly, you'll need a trigger system capable of giving the MS enough information to produce discrete output pulses for multiple ignition coils. To run wasted-spark, a missing-tooth crankwheel is sufficient. To run fully sequential, you'll need either a crankwheel plus a one-pulse-per-rev camwheel, or just a camwheel with either a single missing tooth pattern or two outputs, one with one (or more) pulses per crankshaft half-rotation and the other with a single pulse per cam revolution.

See, the ECU has to have some way of knowing, at a minumum, when #1 (or any other specific piston) crosses some specific threshold. If you want to think of this in simple terms, imagine that we have a specific event that happens (such as a missing tooth on a crankwheel) when #1 crosses TDC. (This particular configuration is invalid, but this is simplified for educational purposes.) When it sees that unique event happen, it can say to itself "Ok, I know that was #1 that just went by, so the next time I see a pulse, that one will be intended for cylinder #2/3" and thus, it can synchronize itself such that it can know which one of the two ignition channels to fire at any given moment. If we extend this further, and add a unique event once per cam rev, then the ECU knows not only that piston 31 just went through TDC, but more precisely that it went through TDC on the transition from the compression to expansion phases, so we now have enough information to discretely fire one of four ignition outputs channels, as we know which piston is going to be next for an actual combustion event.


On the aforementioned DIY writeup, they don't have this info. They're just getting one pulse every half crank rev that says "fire the coil." There's not enough information to fire one specific coil of a number of coils, so they have to rely upon a distributor to force the spark event to be directed to a particular cylinder.

This is why I said "WTF, do those cars have distributors?" You can't run the engine on evenly-spaced pulses without gaps or some other unique event without a distributor.


Sidebar: Why the heck did we ever start calling these devices "igniters"? They're not igniting anything, they're just switches, same as an injector driver. Am I the only one who finds this term ackward and imprecise?

Ok, I didn't look closely enough at the picture. I thought it had both a distributor and a separate CAS; all of the CAS pictures I found in my search were closed-ended, but in the picture on your webpage, there's clearly a shaft for a rotor to fit onto.

That I've seen before (distributors with built in logic-level pickups.)

I know i dont use the wright up for the CAS, what i was asking is do i use the wright up found on this link here

http://www.diyautotune.com/tech_arti...s_overview.htm

Im asking about the last wright up on the page i wanted to know if thats the one i should be using.

or

should i do it like you said and build it like the miata ignition setup. If i should do it like this what year miata wright up should i fallow?

I also agree with you about the ignitor, i think it should be called coil switch or driver.

The one called "Direct Coil Control Distributorless Ignition on a Megasquirt-II V3.0 or V3.57 Board"?

You could if you wanted to not use the OEM coil driver that you have on hand. That writeup assumes that you are using coils which didn't come with a driver, hence all the BIP373s. It's just more work for no gain.

If you want to use your factory coil driver, then as I said, you will wire as per the usual Miata standard (assuming two-channel wasted spark.)


But before you worry about any of this, you need to figure out what your inputs are going to be. Apologies if you've already done this, but I didn't get the feeling that you'd settled on a solution here. You can either install a missing-tooth crankwheel and call it a day (good enough for two-channel ignition and injection) or you can come up with some kind of hybrid solution involving that CAS of yours, either a replacement disc or some means of filling in some of the holes that are on it so that you wind up with a much lower toothcount on the primary and only one hole, not four, on the secondary.

Has Datsun never put an actual crankshaft sensor on any of these cars?

Im 100% on the trigger wheel i have said fuck the CAS ill just leave in there and not hooked up. i want to just do the crank trigger wheel. I spent 10 hours on my ipad yesterday reading on the trigger wheel. I was confused about the wright up diy had but now that i have you helping me. I am able to get all the question i had answered.

Joe i think im going to do the EDIS with the later e46 m3 coils. A guy found that they will work for a EDIS4 for a inline 4. here is the link also i have a ton of these coils from my old job.

http://www.hbci.com/~tskwiot/2002_MSII.html

NVM some one at my house tossed them and they were new in the box from Bosch. =[

In that case, your life is easy. Set up the two ignition outputs exactly as you would on a Miata, and configure the MS software for generic wheel, missing tooth, no secondary, wasted spark.

Knowing precisely how to set the missing-tooth parameters can be a tad confusing- specifically setting up the mechanical relationship between the missing tooth, the VR sensor, and the crankshaft. To keep things simple (assuming you use a 36-1 pattern) you should strive to get everything aligned such that the sensor is aligned with the ninth tooth "after" the gap when the crankshaft is positioned at exactly #1 TDC. That way, you can just copy the "normal" parameters and not have to fiddle with the wheel spreadsheet.

See here:

http://msextra.com/doc/ms2extra/Igni.../36-1setup.GIF

Full details: http://msextra.com/doc/ms2extra/MS2-...Miss_Tooth.htm (Don't worry about the lower half of that page which deals with internal wiring. The stock VR sensor decoder sucks, and when it comes time to build, I'll show you the MAX9924, which sucks far less.)


Dog herpes.

Seriously, I have no problem with most of the supporting hardware used in the EDIS system- the crankwheel, the sensor, even the coils themselves (though you can do better on the coils.) But the actual EDIS module itself... It's not that there's anything wrong with it per-se, it's just needlessly complicated and forces the MS to operate in a mode which is not native to its design.

Yes the 36-1 wheel is the one i have been looking into. So what i need to do for wheel setup is set the motor to TDC then mount the wheel so the air gap 180 degree from TDC then mount the VR sensor over the 9TH tooth. Then use DIY settings for the MS. I just mentioned the EDIS again back there as i want to try and take in the 2 ways i could do the trigger wheel simply.

Or does it not matter were the missing tooth gets place. Just as long as the crank pulley is set to TDC and the VR sensor is 9 teeth after the missing tooth. Like in the image joe posted for me. I think that would work as even the the missing tooth is any were from TDC as long as its 9 teeth after it will still read right. From what im guessing the only thing it would change it where the VR sensor is mounted.

Am i right?

you should be able to tell it when the engine is at top dead relative to the dead spot on the crank trigger.

AEM ems has an offset feature which is used to sync engine and computer timing.

edit: why not use the factory sr20 ecu harness and a nissan ECU of some sort...be it power FC or similar? there are only 5 wires on the sr harness that go to the inside of the car... most of that is for the cluster... then there is only a few others that consist of power etc.

The location of the VR sensor relative to the missing tooth @ TDC really only matters if you're running an EDIS module. Otherwise, the trigger offset is adjustable in software.

okay thats what i was thinking ben.

Joe i wanted to know if you could give me the pros and cons for the edis and they way you said you would do it. If you have the time to type it out.

I have search and have some basics on the two i under stand the EDIS more tho. I also know that if i was to do the EDIS with 4 coils for wastespark or with a coil like diy IGN-4. A guy found the EDIS at 12v has a 6AMP and 4.3MS dwell for one coil. So if i was to run wastespark i would need a coil that charges at 6AMPS and 2.0MS dwell. Now for a coilpack like the IGN-4 i would need it to just operate like the factory EDIS coil 6AMPS 4.3MS dwell.

Now i am trying to find the specs for the stock S13 SR20DET cops. I might also look into the CBR or GSX coils as that is what people upgrade to with the SR motor. Those coils from what i am told are just weapon x coils but with another brand name on them.

Quick question the diagram calls for a 25 microfarad capacitor any reason why its needed?
http://www.megamanual.com/ms2/edis4ew.gif

Also what voltage capacitor should i use?

http://www.eui.eu/Personal/Researche...is4_ms_eng.gif

here is another image the one from mega manual is not working. If you look next to the coil it calls for a 25microfarad condenser. Is it really needed, will it make my MS2 blow up? The diagrams on megamanual for the edis call for a 25microfarad but dont say why.

So look what DIY just came out with after i went wit the EDIS system.


http://www.diyautotune.com/catalog/n...4de-p-472.html

http://www.diyautotune.com/catalog/n...4de-p-472.html