Grumpy

The acoustic frequency (actually signature) the Knock Sensor is looking for is based on cylinder bore. The balance and reciprocating items inside the block other then those mandated by cylinder bore, have nothing to do with false knock. You can trigger the sensor, falsely, but it's not a function of internal part weights/ balance. You can have solid lifters generate a acoustic signature similiar enough to trigger the sensor, as well as piston slap, etc., etc., but the harmonics of balance ain't your problem.

89 Iroc Z

I much rather use an ostrich then program through the ALDL. Programing through the ALDL is not an advantage nowadays with emulators.

The 7730 is not "slow", it is also by no means lagging behind the engine or something. It is more then fast enough to handle everything in the $8d code fine. It doesn't take much at all in terms of processing power to handle a engine. So making the argument that the LS1 PCM is faster is somehow an advantage is bogus.

The added processing power and added memory of the LS1 PCM is not going to give you any advantage. It would be an advantage if the 730 didn't have enough memory or processing power to do what is needed but that is not the case. The reason the LS1 PCM has a faster processor and more memory is that it has to be able to handle the huge bulky code that the LS1 PCM has. The code is much larger because it has to handle a transmission, process the crank and cam sensor for SFI and CNP, take into account a lot more emissions related stuff that OBDI never had to worry about, a much more advanced knock detection system, error handling and recovery if all the added stuff has problems, fly by wire (on the corvettes), and lots more. I don't view the LS1 PCM as a good choice for a ECM swap into a gen I SBC.

I think the best cost effect choice for a Gen I SBC would be a 730 with expanded VE and spark tables and a few other tweaks to the code. I am not one for making things over complicated.

If you are looking for the latest and greatest a aftermarket ECM would be a much much better choice then a LS1 PCM and I am willing to bet that it wouldn't even cost much more then all the work needed to run a LS1 PCM.

11sORbust

I could do it for $500 or less. My personal experience is much different than yours, I guess. In my mind, the LS1 PCM is WAY better than a 730. The 730 has a very sloppy fuel delivery setup (batch fire) and the ignition is just a modified HEI mechanical distributor. There are no moving parts in the LS1 ignition system(unless you consider the crank sensor and reluctor wheel part of the ignition). The other points that you think are bogus could be usefull, if writing new routines to the code. I'm not trying to sell you on the LS1 PCM....it just seems logical (for me) to use it over anything else.

Plus, I'm not a big fan of realtime emulation. The way I tune requires me to look over the log data several times. It seems like a better approach to really analyze the data. My tuning is done by myself, so I can't emulate and drive. Call me crazy...

11sORbust

Bore size, heh? That is interesting. I thought that the weight and composition of the reciprocating/rotating mass has a role in engine harmonics. What is your source of information? I would love to read up on it!

I do know that it doesn't take much to trigger false knocks, once you have them(false knocks) it's basically a useless system.

Grumpy

Porting the 8D code to run in the 2240 gets you everything but the CNP. While nice individual cylinder countrol sounds nice, one test only found 15 on a 1,000 HP engine. Hardly worth the effort, IMO.

CNP while neat, basically just removes the dissy cap air gap from the system. Using an eDist with the 2240, would leave you with parts that are all available in today's market place, to do CNP, if you just have to have that.

Add a good CD to an HEI system, and that's hard to beat.

I've run CNP, and to date, my CD/DIS outdoes the CNP. Having multispark til red line is something not to be ignored, and CNP only gives a single strike. Thou MSD, is supposed to be working on a multistrike CNP, but you can count on that being $$$. One quote was $65 per coil.

11sORbust

I'm going to get my 406 running good at the track(about 450hp n/a & 580hp on nitrous), get the best MPG possible AND run it through the emissions dyno. Then install a LS1 PCM and see what happens. If anyone cares, I'll post up the results. Depending on several things, the engine management will be converted to LS1 specs this spring. I really think the car will run better. Guess I'm the only one though...

89 Iroc Z

The added memory and processor power in the LS1 PCM will not be useful for the end user like us. It is only helpful to the designers who made the code. I don't think anyone is going to be changing or adding to LS1 code for a very long time, or ever. The newer PCM's are written in higher lever languages and compiled from there so if you decompile the code and try to follow what's going it is a complete mess. Not to mention that the code much, much larger that alone makes it quite a task.

The 730 was written in assemble so when decompiled it is much smoother flowing and easier to do something with. Also the 730 has much less code which is another plus. The code for the 730 can be edited any modified by the end user enormously easier then the LS1 code would be to change.

Why go through all that hassle? For what, CNP and sequential? The CNP can be done without a LS1 PCM and for cheaper with the EDIST. And the sequential doesn't add horsepower, the LS1 PCM goes to batch at WOT.

So there is only ONE benefit after going through the huge trouble of getting everything wired up, and changing everything in the calibration to get it to run good, you may get a little more fuel economy from going to sequential.

If you really want CNP get an edist.

11sORbust

I'm going to get my 406 running good at the track(about 450hp n/a & 580hp on nitrous), get the best MPG possible AND run it through the emissions dyno. Then install a LS1 PCM and see what happens. If anyone cares, I'll post up the results. Depending on several things, the engine management will be converted to LS1 specs this spring. I really think the car will run better. Guess I'm the only one though...

RednGold86Z

iTrader: (1)

Does an LS1 use flat response or resonant knock sensors? Resonant sensors are selected by bore, according to (i think it was that) Wells catalog. But, it was something like a list of 5 ranges of bores, so it wasn't all that critical. Flat response sees everything, and the computer sorts out the knock.

Also found several instances in papers where it says the "knock resonant frequency is strongly related to the cylinder bore's diameter..."

boring, but does mention that bore relationship
http://cafee.wvu.edu/Assets/Document...s%20Engine.pdf

Read this one for fun
http://yarchive.net/car/knock_sensors.html

???

iTrader: (5)

i know this thread is old and got very offtopic in the end, but i'm looking into running the n* setup with a 730 and found this in a search.


i'm still not understanding why a lot of what i read says a cam sensor isn't needed. where it would make life easier, why would it work. i understand wasted spark fires2 plugs at i time, so it doesn't need to know cly6 from 1 but it still needs to know say cly #3 from #1 as in when to fire it. how can it know that without a cam sensor?


and the pic of the caddy crank wheel you posted it weird looking with to sensors. can it somehow do both the crank sensor and cam out of the same wheel?

this pic here.
https://www.thirdgen.org/forums/atta...c-100_1110.jpg





this is for a odd, one off setup of trying to run code59 on a 93 lt1 by putting a 730ecm on it and not having to drill a hole and mod intake for a dizzy, which they say can't really be done to a 93 intake anyways because the fuel rail cross over is in the back on them.


here's another pic i found

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


also i've read that i will need to make changes in the bin to the referance angle. does anyone have a link to what all i will need to change timing wise. and understanding why would help alot too lol

jwscab

The 730 ecm is not designed to run in a sequential fuel mode, all the injectors are tied together (batch). The northstar ecm had individual cylinder fuel control, and 8 injector drivers. So it could use the cam sensor. the 730 would not know what to do with the cam sensor. it's also not used for ignition control, so it's a pointless effort.

the beautiful thing about the northstar pack is that it looks like a regular distributor to the ecm, with the exception of it's reference angle and end limits(code changes). there is an article on diy-efi.org or gm-ecm.org wiki that talks about the reference angle and what you need to set it to. if you are really interested, I can dig up information i had from that long ago.

here is the link on diy
http://www.diy-efi.org/twiki/bin/view/Gmecm/DISdist

???

iTrader: (5)

I understand the batch fueling apart. what I'm missing is, ok. on the lt5 crank wheel you have 9 teeth, 8 are evenly spaced for trigering spark. the 9th is at 10* before top dead center. so #1 plug gets fired when it should. on the northstar setup. without the cam sensor, is there a odd tooth that it uses for this? I figured it used the cam sensor. I've read a little about the code changes. are these code changes I can make with turnerpro. or do I need something on the code level. cause I'm not there yet, but I believe 59 is being used on atleast one gn so it should be the same type of changes right? just maybe different numbers? I'll read your link when I get home tonight. thanks for the info and I'll save you the time of digging up the old notes till I know forsure I can pull this off.

jwscab

the Lt5 is a different animal, the timing generation is very different. I don't pretend to know much about it, except it does use a much simpler crank signal.

the northstar crank signal is very complicated. it's not at all like the simpler v6 or ford 36-1 stuff. it has a very special pattern that is decoded in the ignition module. the two sensors determine position, and supposedly, the engine would run if any of the three sensors was to fail(2 crank, 1 cam). I have some northstar documentation that I designed the wheel from. it basically shows you in degrees where each blip on the sensor would be. from that, i determined the slot position and width based on the sensors. there was some documentation depicting the spacing on the two sensors, so that part was easy.

as far as code is concerned, its a few values changes, that could be done in tunerpro i think, I don't know if the fields are easily visible in a standard definition, but they are constants, so they might just need to be defined.

just let me know if you need more info. I can give you details regarding the trigger wheel, and the northstar documents. i think they are on gmecm however, somewhere, you'd have to dig.

???

iTrader: (5)

ok, just got done with your diy link. the code changes make sense now. of course i'd have to figure out what the referance angle for a v8 would be. but since the n* stuff is so different. blah.


so now i'm left to figure out if i can get a crank trigger made up. and the n* stuff has 2 crank sensors? i didn't know that. i just figured you had the crank and cam sensor aimed at the same wheel lol.

i'll search around emecm but if you have the wheel details handy. it would help alot.

RBob

iTrader: (1)

Looking at the N* wheel it is probably a fast start setup. Whithin 90 degrees of crank rotation the ignition module knows which cylinder pair are coming to TDC. Fire off the correct coil and the engine is running.

The LT5 setup is simplier with the one odd tooth out. But can require a full revolution before firing a coil.

RBob.

jwscab

i'll see if this attachment works

???

iTrader: (5)

yep that worked..


ok 2 questions, from the pic you posted on your bench. it looked like you made different size teeth for the 8 sycn notches over the reg ones. or is it just 32 evenly spaced notches, about a 1/4 in wide. or does the with matter too. like you need to know the size of the oem crank mounted wheel,(which i had found a pic of and can;t find again for the life of me. but it looked small, like 4in) and blow it up to the size we need and keep the teeth with same? what about thickness.


2. what about the 24x out. we use the 4x right? any was to use the 24x out but make a much easier crank wheel? i'm thinking no, since it finds the #1 referance by counting the odd number teeth with the 2nd sensor. which i still don't know how it can tell the differance between referance teeth and normal teeth.


i did find this place that makes trigger wheels, but its in the uk. the wheel itself will be the hard part for me. i don't have the tools to make it up myself if the specs are so tight it can't be done with hand tools. i really wish i had a mill lol

http://trigger-wheels.com/store/inde...FYoH3goda20JSw

jwscab

you can do this with hand tools with the exception of the center hole for the crankshaft pulley. although that is probably possible if you have a drill press.

just to let you know, I never got this on a running engine becuase I still need to finish the machine work and just never got around to it. I did try it on an ecm test bench with tunerpro running, so all the sensors and modules were working. the rpm was working properly, so this tells me the loop was working properly, as this is end value determined by the disc, sensors, ignition module, etc.

I took a brand new protractor and a metal scribe and laid out all the angles the best I could on a piece of cardboard, or cardstock, heavy paper. then i transferred this to a 1/4" thick steel disc, 8" in diameter. this was the largest disc I could get cheap, but you might find something a little bigger, I was trying to make sure the disc was slightly larger than an 8" balancer and lower pulley. to get the angles, I just used that chart, and blew it up and printed it. it's a 'unwrapped' 360 degree chart, so it's pretty straightforward.

the gaps were around .3" or so, i made them slightly wider than the sensor end, and they were around the same in depth. I had my dad mill them, but you can just as easily use a drill in a drill press to make a hole, then file or saw the opening open to create the notch.

as you can see in the pictures, i never made a sensor mount, but it would have been made of some aluminum stock, with adjustment for final angle at TDC.

Oh, and as for the 24x out, without the cam sensor, i don't think there is an output. even then, it still needs the specific wheel pattern to work correctly.

???

iTrader: (5)

hmmm ok i really would like to give this a shot. i have to be missing something thou. unwrapped 360 degree chart? im not really sure what that is, unless you mean its just evenly spaced teeth. and if your meaning you printed the wheel from the 3 pics in the zip file you posted up. to me that crank wheel looks to be an evenly spaced 32 tooth wheel, with the shaded teeth just to help understand the how it referances #1 cyl. but in the pic on page one of the wheel you made. it doesn't have evenly spaced teeth. this has confused me from the first time i saw that pic lol. the only crank pic i could find on the web doesn't show teeth like yours.


i've got to be missing something?


i did a quick search on car-parts.com in my area and there seems to be a ton of them listed for 98 and older 4.6's, i believe thats all the same northstart stuff, right?


anyways, here's a video of a running big block chevy i ran acrossed. looks like a tbi motor to me


http://www.youtube.com/watch?v=zokjO2aULIM

http://www.youtube.com/watch?v=0w6Le...eature=related

afgun

The early N* trigger wheel is a complicated beast and it is not all evenly spaced... the diagram is confusing. The wheel is not to scale. As the document states, there are 24 evenly spaced holes or notches. There are 8 additional notches (located in some of the unnotched "high" spaces between the 24 original holes) that allow the N* ignition module to determine crank position. If you look at the second image that jwscab posted, you should be able to see this. 100_1110.jpg from his posts on page 1. The module with coils is fairly easy to source, and they go on eBay from $30 up to around $90. The sensors... haven't done a lot of looking yet. Anyways, the thing you'll have to do is what jp did, and make your own trigger wheel, because the northstar reluctor is cast as part of the crank. The guy who put a wheel on his BBC, I believe took a bad N* crank and had it machined to get the wheel.

I'm going to be doing some work to get my recently acquired junk N* running on a 730 this winter before moving on to try and hack some other ECMs. I'll be following this thread and posting my experiences once I get going...

jwscab

afgun's got it right.

by unwrapped 360 degree chart, I mean the chart on the page is broken into 360 divisions, which is exactly the number of degrees in a circle. figure out where each 'blip' is shown on the chart, and this is where the notch goes, in degrees, on the trigger wheel.

Corner Worker

yes, he is trying some crazy stuff to fix my odd situation. I believe I can fabricate the wheel, or even punch one out on a CNC turret punch that I have access to. Just need a good file that I can give to the machine programmer. This whole idea started from the overall dislike for the 93LT1. Hopefully we can pull off something that can be a cheap fix to a bad design and give the LT1 a little better light.
I picked up a N* coil pack assembly and the two crank sensor wheels the other day from a local JY. I was thinking of intially getting a bad crank and cutting it up to get the wheel, and then just laying out a 3 hole pattern and bolting to the front of my balancer. A simple bracket to hold the sensors and mount the coil packs/ignition module....and the hardware side is done...??? has the task of making it run, will keep you posted.

afgun

I'm also looking for an N* reluctor cut from the crank... I'll take one too please.