11sORbust

I'm working on something different. It might be possible to retrofit a 730 equiped car with a DIS(using all OEM parts). Not wanting to get into the details right now. I just need to test it. IF it works then I'll submit a tech article to TGO on how to kill the distributor and install coil packs on any sbc. Before I test everything out, I'll need to know a few things. Maybe someone here can help.

The dist reference pulse on our thidgen cars is what I'm interested in. This is what I have learned so far...

For every two revolutions of the crankshaft, each gang of four injectors opens and closes twice, alternating left bank (four cylinders), right bank (four cylinders). Large and small cap ignition modules are the exact same, just "packaged" different. The pickup coil uses a magnetic field to trigger the ignition, as the timer core teeth move away from the pickup tooth, the magnetic field becomes weaker. As the air gap gets larger, the pickup coil signals the ignition module to turn the primary current off. This induces a high voltage current in the secondary coil, and the distributor fires one spark plug.

The questions,


1) Does the distributor pickup coil trigger the ecm/injectors on each of the 8 "trigger teeth"? The diagram I have fires one batch every 180* turn of the timer core/reluctor. If that's true then how does it work exactly?
2) How does the ECM know which gang of injectors to fire for each DRP?
3) What type of signal runs from the ignition module's signal converter to pin BD8 (REF hi).
4) What does the ECM do with BD8's signal? This IS the DRP, right?

I'm thinking you could answer this with a general explaination.

RednGold86Z

iTrader: (1)

As far as I know, all TPI's were Batch fired, not bank, that means all injectors on at a single time, all off together.

The REF signal is 5 volts as far as I remember.

Ref signal (and ECU) has no idea which cylinder it's on. Only the distributor "knows."

//You can do V8 DIS with 4 coil packs if you wish to save money or even waste fire 8 Coil on Plugs if you don't have a cam reference. You'd have to build a circuit to "distribute" the signals, and you'd have to find either a crank or cam single pulse for a reference (crank pulse to the circuit with waste fire will start that much quicker).

I personally see no point in doing this without using the benefit of a 60-2 tooth crank wheel and an aftermarket ECU to do the timing. It's hard to beat 60-2 accuracy, especially when comparing to a distributor, which is gear driven at the end of a cam, which is usually chain driven to the crank, and only has 8 ref signals in 2 revolutions.

11sORbust

So you saying that both injector banks fire fr each DRP?

Not worried about the hardware end. That's already mapped out with all OEM parts. My focus now is on the ECM side...

JP86SS

iTrader: (1)

I only see where the "crank" puse width is multiplied but nothing on the actual running.
I'm also interested if someone could clarify when the DRP count triggers the injectors to fire. Is there just a count of 8 or 16 pulses and they fire?
Haven't gotten that far in my commenting yet to figure it out.

TheGreatJ

I believe the injector firing is not part of the code. My interpretation of what I've read and seen, is that the code just sends a PW or DC value to the injector driver circuitry, and the DRP signal tells the injector driver to fire directly, without going through the processor.

edit: Looking at the '7749 schematic here it looks like U2 (PWM output driver?) gets the DRP signal and PW data and controls the injector driver.

Grumpy

As long as crank sensor, and cam sensor develope GM style *triggers*, then whatever module you're using will work. If you stick to a GM module then you don't have to worry about nuttin, hardware wise. Then all you need to do is use the appropriate bias's, max, and min advances from whatever code you sourced the module from. The GM sensors typically, use reluctor, or Hall effect sensors. The Reluctor type are A/C generators, and the signals vary as RPM increases. If you use a reluctor type sensor, then you need to verify the calculated timing matches what you see at the crank, at all RPM levels.

The injectors fire all at the same time, on every 4th DRP, ie once per crankshaft revolution.

11sORbust

Yep, the hardware side is all lined out. So, are you sure that both inj drivers fire for every 4th DRP? This is looking too easy! Can somebody send me a link of a SBC retrofitted with DIS (ALL OEM parts)? I heard there is someone out there. Still have yet to see it. Not talking about using an Edist or aftermarket parts. Talking about over the counter factory parts. The system I'm going to use will be more acurate than the lt1.

-=Jeff=-

I cannot wait to hear more about your set-up

JP84Z430HP

I don't have it in use, but I have the ground work down for such a thing. The only thing that will not be OEM is the trigger wheel that goes on the crank, a bracket to hold the crank sensor, and the wiring will have to be fabbed up. I'm still trying to get the dimensions of a northstar crank trigger wheel to complete the drawings.

I have a couple pics of what I'm working on here: http://www.jpteck.com/efi/northstar

I plan to use the Northstar system from cadillac, although, the LT5 (Vette ZR1) would work pretty much the same with the exception of the trigger wheel. The LT5 wheel is simpler, but the parts are going to be more costly, due to the fact that there are MANY more northstars out there then LT5's.

HTH
JP

11sORbust

in 6 months it'll be done. *IF* it works then I'll post up what I did...

89 Iroc Z

I have a Northstar crank sitting here that I am going to CAD out the trigger wheel. The problem is I don't have the large measuring calipers needed to get the radius of the wheel so I am not able to make an accurate enough drawing. I have had the crank for a few mounts now waiting to be drawn. I will come up with something to get this draw out.

-=Jeff=-

Cool...

11sORbust

I have a few ideas on the reluctor wheel. Need to try em out before passing the info along. Your problem would be fixed really quick if you knew someone with a CNC mill.

Could someone link me to info on the northstar ignition?

I'm still wondering if there are any 3rdgen out there with DIS (using factory parts). :shrug:

Grumpy

http://www.jpteck.com/efi/northstar

11sORbust

I didn't find anything anbout the northstart ignition but the title.
:cornfused:

JP84Z430HP

http://www.jpteck.com/efi

I have a Zip file there that has some info on it..... Maybe I should put it with the rest of the N* stuff....

11sORbust

Wow, that northstar stuff is WAY to complexed for me......WOW.


To recap, both injector drivers fire once every other revoution of the distributor's timer core? That's it?

Grumpy

The 730 only has one injector driver.

The injectors fire once every crankshaft revolution, and that's every 4th DRP (when in Synch mode, 8D code).

11sORbust

Once every 4, doh! I got it now. Thanks for the info.....and looking over the diagram I made.

11sORbust

Once every 4, doh! I got it now. Thanks for the info.....and for looking over the diagram I made.

PhLaXuS

Did this project ever come to be?

JP84Z430HP

Not in my case. I haven't dropped the project, I still plan to complete it, I have just had a huge lack of time and a lack of funds. I have other things I need to do to the car first. I am not giving up this project though!

JP

Z69

It's still being worked on.....

11sORbust

It's going to happen. Just might switch over to a caddy ignition module, kind of like the deltech setup. The ZR1 module is over $300 and will not be found in the junkyard. Have also thought about using two 4 cylinder modules.

..starting to think it would be best to switch my 3rd gen over to 4th gen electronics. Start out with the 94-95 LT1 PCM. Then it seems like a simple conversion box could be made from the opti to the DIS module. Get that working good and then try an LS1 PCM. THe PCM connector pins look the same for both LT1 and LS1. Maybe one could get by with a repin and connector swap(for the most part). LS1s doesn't need the crank sensor to run. Just make a reluctor wheel that'll fit where the distributor is at for the cam sensor, mount the pickup and create a housing. Could possibly use an LT1 oil pump driveshaft. Then we are talking CNP and cheap parts...

Buying my first house had a huge imact on funding for this project. That is the primary reason it's not finished.

PhLaXuS

If I get my raise as expected, I'll be building a engine and I like this idea a lot! I also really trust AC Delco parts for functionality and ESPECIALLY reliability. later model style Delco electronics would be a good move by me.

88 350 tpi formula

iTrader: (9)

what is wrong with using the msd crank trigger? or what about the vortec timming cover and pickup?

PhLaXuS

I personally do not have anything against those, but I specifically trust AC Delco.

88 350 tpi formula

iTrader: (9)

thats fine.

a/c delco does not make the replacment timming cover and sensor ect? (I can't recall what was all inside the timming area) must be gm only?

Z69

Vortec and MSD are both 4 pulses per rev.
MSD stuff is expensive and costs even more to get to run a multi coil set up. Break something and it's another big expense. The Vortec set is a lot cheaper than the MSD and can be got anywhere. But needs a $300 eDist to run multi coil.
The Northstar ICM used is in the $70 range. The MSD trigger is $250+. The N* trigger set will vary but is comparable to the MSD price. Till you have to replace something.
The N* uses a 32 tooth wheel and at least 2 sensors.
But the 4 coil pack can be had cheap used.
So an N* waste spark set up will end up being almost half the price of a MSD/Edist CNP set up. And everything but the wheel is easy to source. Try to get info on the eDist running with a p4 ecm.

jwscab

I've played around with that exact set-up awhile ago. I had a '730 ecm set-up on a test bench, with a Northstar coil-pack/module, wired into the ignition controls. The module was fired by a simple duplication of the northstar crank trigger, and two northstar sensors. I had the trigger made up of a steel disk, connected to a motor I could control the rpm of. the sensors are spaced apart the right distance, the ecm controlled the spark advance correctly, I just need to figure out where to mount the two sensors in relationship to #1 TDC. Ha, thats about the time my stupid computer crashed, and the damned battery died, and I never got around to firing it back up.

there is no need for the cam sync pulse. not for batch fire anyway. someone could build code to do it sequential, then you would need a cam sync signal, ala vortec distributor and a couple of passive parts. thats advances stuff though, since you would need the inputs for the cam sync.

Its pretty cool to watch the sparks running 'round robin at 1000 rpm or so.

the game plan is to mount the trigger wheel between the crank pulley and balancer, and build a sensor bracket for the two sensors.

for code changes, all i did was set the timing table to 10degrees advance across the board, to eliminate timing movement. I never actually got to make sure the timing advance worked, other than disconnecting the set-timing connector and seeing it retard 10 degrees.

I'll have to post some pics when I get around to it.

JP84Z430HP

From the GM documentation that I've read, it will work with just one crank sensor, the second is there to aid in finding where it is in the firing order, sooner, to help it start up faster.

I have CAD drawings partially completed for hte trigger wheel and bracket. I need to know the sizes of the notches of the OE N* crank to complete the trigger wheel, and dimensions of the crank sensors to finish the bracket. If I get closer to being ready to implement this and don't have the info, I'll find a N* core engine to get the info from.

Actually, the crank and one sensor would be all I'd need to finish my drawings, along with time to dedicate to the project. Time is at a premium lately, in fact, I should be asleep right now, preparing for the long day ahead of me!

JP

89 Iroc Z

I have a N* crank sitting here that I bought to CAD the trigger wheel but had a lot trouble getting the cad drawing accurate due other parts getting in the way of measurements. I have not touched it in months. I will try my second approach, take a angle grinder and cut the trigger wheel off the crank to make it easier to CAD out. Now that the project is back on my mind and I have some free time this weekend I will try to get it done.

jwscab

i don't think the notches are all that critical, dimension wise. they just have to be slightly wider than the tip of the reluctor sensor, and the timing of each notch is in the GM document, that is the critical part. It doesn't matter about the diameter, either. The wheel I made is 8" in diamater, so it fits on a harmonic balancer. about 3/16" thick. each notch is about .300" wide, since the reluctor pickups are about 1/4" that's all that's needed. there's nothing magical about the crankshaft and notch width other than making sure it triggers.

JP84Z430HP

I wanted the notch width, just to be sure it triggers properly. The reason I need the diameter is so that I can scale it to the larger wheel that I've drawn

I remember reading where someone was doing the 4cyl./v-6 setup retrofit onto another application, and the notch width WAS important, to some degree, since it took a few tries to get it to trigger..... Again, I just want to make sure! I suppose I could start out a bit narrow, then spin it and see what it does, even if just on a scope.

Anyway, this is still on hold for me, until I get some of the other stuff out of my way...

jwscab

OK, heres some pics, hope this works out.
Pic 1:

jwscab

That was the basic set-up. don't mind the mess, and the extra electric motor, its just ballast for weighing the fixture down. I have it powered from a box fan motor, and a rheostat control to vary rpm. the disc is mild steel, I have #1 cylinder on a spark plug wire so I can put a timing light on it.
this one should be the close up of the wheel and reluctor position:

jwscab

and the last is the sparks while in operation. it might be tought to see, but you can see the blue arcs on the top of the little light bulbs (nice glass insulators, and what I had around, just had to blow them out first so they were open circuits):

Grumpy

Casper's Electronics, sells what you might call *a proper spark gap tool*, for working with DIS systems. Mundelin (sp), IL.

89 Iroc Z

I did some measurements on my N* crank trigger wheel and here is the breakdown of the measurements of the wheel. Each of the 32 notches is 7/30'th of an inch in length and 1/5'th of a inch deep. There are 16 small raised area's which are about the same length as the notches, they are 13/60'th of a inch. There are 16 large raised area's which are about 3 times the length of the notches and measure in at 41/60'th of a inch.

So calculating in a percent of error for machining tolerances, each notch and small raised area are the same length. The large raised area is three times the length of the notches.

The above broken down into degrees:
- 32 notches 3.75 degrees in length
- 16 small raised areas 3.75 degrees in length
- 16 large raised areas 11.25 degrees in length

The thickness of the trigger wheel is 4/15'th of a inch. The diameter of the trigger wheel is about 6 & 15/16 inches. I will get this in CAD asap.

[EDIT]
- After I put the wheel into CAD I will be able to compare it to the crank I have here and see how close the above breakdown of measurements is to the real thing, should be pretty close.
- Sorry about the weird numbers I left my measuring caliper at home and am using 1/60 engineering scale ruler.

[EDIT 2]
I put the northstar crank trigger wheel in CAD back in 2005 but never updated this post. Here is a link to the CAD drawing: http://www.lukeskaff.com/car/northstar/design.html

jwscab

that will be good to see a real measured/drawn cad representation. The critical dimensions are the crank angles, which GM spelled out in its documentation. even then, there is probably a degree or two of slop in construction, because the module simply 'counts' the pulses to determine firing points.

I do have to ask exactly what kind of ruler you were using though? I'm guessing is a draftsmen's scale, due to the odd divsions/inch.

Grumpy, thanks for the info for the DIS 'gap tool' I didn't find it on Casper's website, but its somethng to look into. I had some trouble initially with the first 'spark gaps' I made, which carbon tracked fairly quickly, which made me think I broke the ignition!

now to build a guinea pig to try this out on.........

89 Iroc Z

I used a 1/60 scale engineering ruler since I left my measuring caliper at home.

11sORbust

I think my idea to use a LS1 PCM is the best. The ignition module is built in and only requires a cam sensor. This would take alot less work than a crank trigger system. Just think about what's possible with that code. Nothing is better than timing control for each cylinder.....

89 Iroc Z

Running a LS1 PCM will take a lot more work, not less. It will also cost much more. The LS1 has a crank trigger in addition to a cam sensor so you are still going to need a crank trigger. You will have to rewire the harness and add some sensors. The LS1 has 2 knock sensors so I am not sure what you would do about that. The LS1 PCM also is sequential so you will have to rewire all the injectors to the PCM. You will also need a VSS conversion box unless you can account for it in the calibration. If you are running an automatic transmission you will need something to control the TCC because the automatic LS1 PCM calibration is for an electronic transmission.

Tuning software for the LS1 PCM is $350 last time I checked. Then you have another minimum of $150 for coils. Plus a bunch of little things here and there. With all the added stuff it will add up very quickly. In terms of cost and ease of installation it doesn't make sense, you are better off using a FAST eDIST on your P4 ecm.

With the Northstar setup you need the ignition module, coils, trigger wheel, and a position sensor. I got my ignition module and coils for $25, there are not hard to find in junk yards everywhere.

Grumpy

Nope, SEFI, with adjustible injector timing, is.

JP84Z430HP

And my new idea, now that I know you only NEED 1 sensor, is back to putting the trigger wheel under the timing cover, as in the 350 Vortec setup, but with the trigger wheel from the N* setup scaled down to fit under the cover. The only issue I have right now (and I haven't spent any time looking as of yet) is figuring out a sensor that will fit the hole that is not a hall effect, as the Vortec sensor is. Give a few days, adn I may have something!

11sORbust

The northstar idea is not bad, don't get me wrong. You do have a few things confused though. You do not need the crank sensor for an LS1 PCM to start and run fine. It's just for misfire detection and a few other small things. It's not that important, trust me. The knock sensors is a no-brainer...delete em. I don't run with knock sensors. My setup has a totally different harmonic signature, so they are useless anyways. The VSS shouldn't be a problem, just use a 4l60E one(or something like that). I know the LS1's vss output to the gauge is a direct connect. TCC control can be handled by the PCM. You are right about the software, it cost some $$$$ . I'm lucky enough to own OBDII Tunercat already though. So my main cost would be the PCM, connector C1 - C2 w/ 6"+ of wire from them and some small stuff. Coils can be had for under $150, if you look hard enough. I could mend my 85 harness to the PCM connectors. THat is the easy part. THe hardest part would be to mount the cam sensor. Have not thought about that yet, maybe just gut my distributor. I really think it would work...

Grumpy

I worry most when I hear that.

While it's your option to delete them, there's plenty of fast cars using them as a front line of defense in engine protection. Try the GM C5 team as one, then Mark Jackson's 9 sec GN, as just two off the top of my head examples of people who's opinon I trust, as examples of why to bother working out the KS details.

JP84Z430HP

Any more questions?

11sORbust

I understand your point, BUT the sensor is designed around a narrow frequency range. My 406 sbc is externally balanced and any knock sensor will register false knocks. The knock sensor is "out of range" once engine harmonics have been changed. I don't have the money or knowledge to design a KS for my application. Even if I did, changing something else could produce a false knock too(like if I switched to solid motor mounts).That is why I said they are useless....for my application. Sorry that I didn't clarify.

Tend to forget this IS the internet and misinformation spreads like wildfire. One has to be very clear when typing a reply..

11sORbust

Yeah I realized after the post that I had it backwards. The LS1 PCM doesn't require a cam sensor to operate. Still seems like a gravy job...to use an ls1 reluctor wheel and crank trigger. Those parts should be very reasonable used. Yeah, it's some work to convert the entire harness for an LS1 PCM. IMO, it's worth it though.
SFI
CNP
Programming through the ALDL
Much faster processor(I think it has two)
More memory
Availability throught the coming years
etc..