afgun

Look at lukeskaff.com; they've got a .pdf as well as .dwg file

???

iTrader: (5)

i'll ask my friend which one he used. i'm heading over there later tonight.

junkcltr

iTrader: (1)

Post #143 says:
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. Sorry about the weird numbers I left my measuring caliper at home and am using 1/60 engineering scale engineering ruler.

The PDF file at that site has a diameter of 6 & 5/16.

Which is correct?

junkcltr

iTrader: (1)

It sounded like he cut it out by hand. Using sawzall, cut-off wheel, etc?

I wonder how much it affects the balance at the crank snout. It sounded liked he recessed the balancer. Did he have it machine 1/4" shorter to make up for the 1/4" wheel?

EDIT: I looked at the files and they give the dimensions. I think I might know someone with a Northstar crank to verify. My balancer is 8" so I will have to resize anyway. Using the values in this thead it comes out to about 1* for a 1/16".

junkcltr

iTrader: (1)

The other thing that probably needs to change in the BIN is the spark latency table. Since the various HEI modules have a different latency then the DIS module is probably different.

This only bin changes I see that need to be done so far for the $8D is:

junkcltr

iTrader: (1)

Here is what I found. If you look at the timing waveform at:
http://www.lukeskaff.com/car/northstar/theory.html
you will see that the falling edge of the 4x signal is 70* before the next TDC. That 70* is the 10*_base_timing + 60*_ref_angle

Since the $8D doesn't have the typical ref_angle constant then it must be included in the base_value which is the initial timing constant.

address L8025: goes from a value of 6* to a value of 10*+60* = 70*

Suppose the desired timing is 59* BTDC at the crank after all the table lookups. Suppose the DIS has a ref_angle of 60* and an initial/base advance of 10*. The total initial/base is 10+60 = 70*. This base/initial angle of 70* is seen in the timing waveforms.

Then 59* - 70* = -11* is computed in the ECM. The ECM would then do L8026 - (-11*) which is -10 - (-11) = 1* which is greater than 0*........so the spark is not limited. So, in order not to limit spark advance then timing > constant. Wanting a max timing of 39* like the stock dizzy would require (MAX_ADV - (TOTAL-INITIAL)) > 0
Using the desired stock 39* of advance yields a MAX_ADV value of (-30 - (39-70)) > 0 or (-30 + (31)) > 0 or 1 > 0

As you can see, the DIS doesn't have the (39+6) = 45*spark advance limitations of the dizzy.

The retard value is done using (MAX_RTRD - (TOTAL-INITIAL)) < 0.
If the result is greater than or equal to zero then clamping is done (spark is limited)
Using the above values with a max. retard of -70 gives
(-70) - (59 - 70) = -70 -11 = -81* which is less than zero, no clamp is done

Using the stock $8D value for retard and say 20* of timing (idle) with the DIS offset gives
(-3.5) - (20-70) = -3.5 - (-50) = 46.5* which is GREATER THAN ZERO and clamping is done
This is why things go wrong when the MAX_RTRD is not changed properly.

This is the max. retard value using the so called proper values.
(-70) - (1 - 70) - = -70 + 69 = -1* so all is good, no clamp is done
The Max retard is 1* and not 3.5* like in a stock $8D bin

Using a MAX_RTRD value of -67.5* would make the max retard like the stock $8D
(-67.5*) - (3.5 - 70) - = -67.5 + 66.5 = -1

If you want the same spark advance value as the $8D you would enter the following:

L8025: 70* decimal
L8026: -30* decimal gives a max advance of 39*
L8028: -67.5* decimal gives a max retard of 3.52*


The values below are for the "standard" DIS module and are the values I would use from what the timing waveforms show:

L8025: 70* decimal
L8026: -10* decimal gives a max advance of 59*, VALUE = (desired_max_adv - initial_adv) + 1
L8028: -70* decimal gives a max retard of 1*, VALUE = (desired_retard - initial_adv)

These values are for the $8D bin.
Note that the values I posted are different than what RBob posted earlier in this thread. These are the values from post #143
L8025: 60* decimal
L8026: -10* decimal gives a max advance of 69*, VALUE = (desired_max_adv - initial_adv) + 1
L8028: -70* decimal gives a max retard of 10*, VALUE = (desired_retard - initial_adv)

The max. retard seems a little high to me. This is better for the non-$8D codes that include both an initial constant and a ref_angle constant. Together
the two constants add up to about 70* like shown earlier.

So there is the code stuff.......all done using a disassembled bin. Now to get some trigger wheels made. Like mentioned earlier in this thread....this would work great on the LT1 intake swaps and the LT1 engines. I think this is an easy change in the $0D code also. I will have to take a look. DIS along with the E-trans would be nice. Gotta do the spark rev limit mod. though because the revs come up fast when the tires break loose from boost. Pulling spark and adding fuel is a crutch for a real spark rev limiter.

On a side note, I think this setup is better than a 411 PCM. The reason being is that the $0D code can easily be adapted to boost, N2O pulsewidth injection (variable N2O), water/alky motor PWM, extra IATs, oil temp sensor, wastegate control. The list goes on and on. This can't easily be done with the new 411 and up PCMs. Not to mention that those PCMs require $500 and up just to tune ONE of them and a lot of the calibration tables are missing. The $0D can be easily modified to send out new variables on the ALDL. Again, not easily done on the 411 and up ECMs. The only thing the newer ones have going for them is SEFI. Until the tuning gets cheaper or I need a new project it is OBD-I for me. I can live without the SEFI and just run a tad richer so the leaner cyls are happy.

junkcltr

iTrader: (1)

The latency table thing. The $8D and other bins using a HEI module dizzy have this table.

For the $8D at 3200 RPM the latency table is 305usec
At 3200 RPM that is 18750us or 52usec/crank_degree
So 305/52 is about 6* of timing advance change due to HEI module delay.

At 6000 RPM the $8D also has 305usec. This is 10000us/360 = 28usec/degree.
305/28 = 11* degree of timing advance.

So if the DIS doesn't have the latency like an HEI then you could be up to 11* advanced if using the stock $8D table.

Corner Worker

no. I just downloaded a cad reader and printed it off of that. I did a layout of the wheel based on the drawing, and transferred my layout to a piece of 3/16 steel.

???

iTrader: (5)

well we spent over 4 hours trying to get the ecm to control the timing right. burnt over 40 bin files and nothing even came close. I tried the max retard and max adv way, and the work around of adding a bunch of timing thru different bais's and spark maps. nothing worked or seemed to make any sense. I would get it within 10ish * at idle of comanded and then any way I went with any value would make it jump 40* in any way it felt like it. at this point I need a week to regroup, go thru my notes I made along the way and hope something makes sense cause right now I'm questioning if this can even be done. other than that, the car runs fine with the bypass wire pulled and running on base timing. needs the tune cleaned up a bit, but drivable. blah

Six_Shooter

iTrader: (2)

I'm still thinking that the cam position signal is needed.

???

iTrader: (5)

I hope not, cause that kinda throws this whole swap out the widow for most people. but we will see. going to change up some stuff on the sensor mounting and see if anything helps. I still have hope.

afgun

The N* DIS module should run with 2 out of 3 of the signals present. So if you have both of the crank signals, it should be OK. http://www.cadillacforums.com/forums...93-99-dis.html You are likely fighting with something in the program...

jwscab

there is one thing that I don't think has been mentioned, and this might be why you have issues getting it to run, ???, in the diy-efi post it mentions that the numbers are
2's-complement, since they are negative numbers. You might not be putting the right values in. Since I have very limited knowledge with respect to the code, I'm not sure what the actual values should be. Rbob or Junk might be able to shed some light.

junkcltr

iTrader: (1)

I will take a look at the $8D again. ???, what values did you enter for the timing in the bin?

The V6 DIS module latency is 153 usec. I would set the $8D latency vs. RPM table to use 153 usec for all RPMs to start.

???

iTrader: (5)

I don't have my laptop in front if me, but the values and locations Rbob posted above. I also found a said to be working v6 .bin and .xdf using dis $8D to check that I had the s_apju xdf defined right. after I made the changes I double checked the hex numbers thru tunerpro and they matched what Rbob posted. so I felt the formulas are right, just with the 2swhatever it allows you to type in any number, even if it doesn't fit within -90 to 90. for retard and adv but as long as you know that's the min and max, it seems to work judging by the hex numbers. I also tried the other way, which you make up the 60* timing of the dis with timing in the bin.(and rasing the max advanced as well) like the bias and coolant tables. a few guys on code59 have got to work. and that didn't work for me either. he made some hardware changes after I left to how the sensor mount in relation to the crank wheel. and we are going to try again.

???

iTrader: (5)

I will try the latency at 153 usec and see what happens.

junkcltr

iTrader: (1)

The values RBob gave neglected the 10* of initial advance. I think you need 60+10 = 70* for L8025. That is a value of 0xC7.

Why is playing with the two crank sensor locations? The diagram shows exactly where to put the sensors. You can't move them. If you do then the base timing will be moved and the bin initial value must be moved.

Are the crank sensors wired right? polarity?

RBob

iTrader: (1)

I'm not sure how I neglected anything. I gave an example and stated it was for 60* BTDC:

As an example for a DIS set up running at 60* BTDC initial timing, here are the values:

???, did you have a chance to put a timing light on the damper with the engine running? First with the bypass open, then with the ECM in charge of the timing? That will tell alot about what is going on.

RBob.

junkcltr

iTrader: (1)

Right, but ??? took it as working for his Nstar DIS which looks like 70* BTDC. Does your value include the base advance, ref angle, or both? I think that confused him.

I think some might get confused with comparing the $A1 to the $8D because the $A1 does an initial of 10* and a ref angle of 60*. The $8D just has the initial which must be 70*.

RBob

iTrader: (1)

The issue is the choice of a word.

And if the timing was off by 10*, it wouldn't cause the engine to have all sorts of problems running.

To ???, besides the timing light, did you get a data log? The ALDL spits out what the ECM is setting for timing. The converted values may not make sense, but we can convert from the raw.

I agree with the week off to re-group. BTDT and it does help.

RBob.

syntax terror

I know this is a different mask but it doesn't use the cam position signal. http://www.fiero.nl/forum/Forum2/HTML/070626.html This thread goes into some detail

???

iTrader: (5)

wow, thinks for the replys guys. i had missed some when i looked thru last night.

as for data logging, i didn't record anything, but i did have it open and watching real time. TP was showing normal looking timing that was commanded in the bin. lt1 spark tables, 25ish idle, 41ish reved to around 1500.

let me read and reread this a few times and i'll try to understand it and also answer the few questions. just wanna make sure i'm answering them with what really happened. because we thru a ton of different things at it. moving min and max spark around quite a bit. also thanks for the part about needing to add the 10* base to the reference. now that i've read it, i remember hearing something about other dis codes having a space to plug that in but i didn't understand it at the time.

what do you guys think of this idea? its from another post on another board. but if it works, it may clear some stuff up.


i also downloaded that $A1 bin and xdf from the link above about running a northstar motor on a 730ecm. its said to be a running bin. so if i load it up, and it works. atleast thats a start. if that doesn't work. something is out of wack.

???

iTrader: (5)

ok, dumb question about location and tunerpro xdf's.

so the location of the values are this.

L8025:
L8026:
L8028:


take max retard, i highlight, hit F2 and edit box pops up. for address it says address(hex) 28.(which doesn't seem like a hex number to me) and when i hit the summarys list, and scroll down to max retard. it shows 28--->29 under location. where's my L80 at? because under summary the spark ret,max for knock lists the same 28--->29(happened to be right under max retard) as its location. i know these both can't be the same. how would i check that i have the location right. as in L8028?

just trying to cover everything i don't fully understand.


i feel like a max retard

junkcltr

iTrader: (1)

Yes, my choice of words could have been better.

Just seemed like ??? was using your example for the $8D code.

junkcltr

iTrader: (1)

What looks like L8025 in the code will look like 0025 or just 25 in the XDF.

For the $8D XDF file you want the following:

address data
0025 70_decimal (0xC7 hex) which is 70*
0026 0xFF
0027 0xE4
0028 0xFF
0029 0x39

???

iTrader: (5)

ahhh thank you. i went to hex editor and put the cursor on 0025-0029 and all the hex numbers matched what you posted above. i'm guessing the 0x part is something you drop.

and i did try adding 10* to Rbob's values he gave me because most people did say 70* is the ref for the n*. but i didn't do the math right lol. looking back i did burn something close to right thou lol. still didn't work.

oh and i think i found the address of the min,max, ref in the $A1 above that's said to run the northstar motor. and came up with

rev at $801B. 65*
max adv at $8023. -6.33*
min retard at $8025. -54.84*

i wanted to look up the spark latency setting, but its not defined and i haven't done a search yet. i'm guessing i have to find a disassemble of A1 and see if someone commented it. which is over my head lol

junkcltr

iTrader: (1)

from the $A1 code
address 8021 shows the 153 usec latency correction. It appears it is always constant for the DIS. The $8D dizzy HEI module has a table that varies.

The "0x" means that the number is being displayed as a hex number.

???

iTrader: (5)

thank you.


that 11* would have deff confused me, or atleast not helped.

junkcltr

iTrader: (1)

Good luck. I hope you get it going.

afgun

For those of you considering using the northstar DIS system, I just dug up some interesting information.

Indmar, a marine engine company, "marine-ized" the LT-1 engine back in the 1990s. They did so by mounting a reluctor to the harmonic balancer, and bolting a bracket to the front of the engine that holds a pair of sensors that then feed the DIS module. The DIS module is of course then wired up to the MEFI controller.

Unfortunately I don't have an indmar catalog available to me to look the part numbers up, but if you're interested in adapting the N* DIS to an SBC, you might want to call up your local indmar dealer to see if the parts are available...

89 Iroc Z

afgun,
The LT1 N* setup was used on mastercraft LT1 boats up to 1995 if I remember correctly, I have some engine and shop manuals pictures of the setup I can post next week (they are on my other computer). I was never able to get a part number or price on the mounting bracket.

That would be cool to get a hold of but they probably want a arm and a leg for it.

afgun

I bought a copy of the Indmar GM engine book and can scan the page with the drawing if it might help...

???

iTrader: (5)

couldn't hurt to have a look if you have a scanner and wouldn't mind. the more info online the better for guys trying this. I have a buddy working on lazer cutting 3 wheels out for me now. so I'm not giving up on this for a while. I want my 3rd gen to run on this as well as his 4th gen.

Corner Worker

i used the cad file from luke skaff. I printed it out, and did a layout on some sheet metal. then transferred it to .187 steel plate. I then used a hand held plasma cutter with a custom protractor attachment I made. I then cut it out, and drilled a .201 hole for every notch. I used a band saw to cut the grooves down to the drilled holes. after it was mounted to the back side of my balancer, I chucked both of them up in my lathe and trued the edge. it has as much runout as my balancer, which is basically nothing. btw, I have a complete metal shop at my disposal, and 10+ yrs experience with metal. ??? is one of my closet friends and he is really the inspiration for this project. I knew he could get the software side of the project handled, I just had to make a wheel, a perch for the sensors and do a repin. I thank you all for the massive amount of input....

89 Iroc Z

Corner Worker or anyone who is getting trigger wheels cut using my CAD file do you mind making an extra and shipping it to me for a cheap price or at cost. For the work I did on the CAD design, etc.

I have the original trigger wheel cut out of a N* crank which I used to make the CAD file but it would have to be milled flat on both sides to mount and I no longer have access to a machine shop.

- Luke

???

iTrader: (5)

if it works I'll see what I can do to hook you up. my buddy in another state is doing if out of scrap metal at his work between jobs and shipping them down to me. only enough for my two cars and corner workers. he didn't seem to into the idea of making them for other people but I'll work on it. what kinda setup do you need? like how do you need the wheel done to mount it? or just a blank wheel and you drill holes?

89 Iroc Z

???, That would be awesome. I was thinking to mount the trigger wheel in-between the harmonic damper (balancer) and the serpentine belt pulley. Three holes or circular slots would have to be machined into the trigger wheel to match the holes on the pulley. Then use the corvette pulley (which has a smaller offset then the f-body pulley) and have a spacer machined to push the corvette pulley out to match the F-body offset.

Another option, which would not require changing the trigger wheel used in the above example would be to mount it behind the damper with a LT1 crank damper hub and a LT1 damper. Anyone know if a LT1 damper setup can be used on a gen1 SBC? Are the engines balanced differently?

???, how and where are you mounting the trigger wheel?


To better explain what I said above about slots. The black slots in the image below would be cut out slots where the bolt would go through the pulley and squeeze the trigger wheel between the pulley and the damper. This is just a rough drawing, it would make more sense to use three slots instead of four since the pulley only has three bolts.

The other option is to just do a ring of bolt holes, which would be easier but not allow you to dial in an exact angle for the trigger wheel.

afgun

Here it is...

jwscab

the lt1 damper is the lower pulley, which is no big deal, but the hub is not machined for a keyway, so you'd need to just use the bolt. I think the balance it close, but closer to a 305 bobweight due to the lighter components from the factory. The hub and balancer also don't have a symmetrical bolt pattern, one of the bolts is slightly off center, took me a bit to figure out what was going on.....

if you use a normal balancer and use the right thickness plate for the reluctor, I think you could bolt it onto the back of the balancer using the three tapped holes for the puller, and make the bolts long enough that the pulley would get mounted with nuts and washers from the front, so you'd have like studs protruding from the balancer.

I would get the reluctor on the balancer as exactly as possible, then make fine adjustments on the sensor bracket.

???

iTrader: (5)

I like the slots idea but I am going to try one step simpler. just had him add a 4.xx inch hole in the center, slide it over the back on the pulley and tack weld it in place. my pulley seems to run very true and I think it will work.

???

iTrader: (5)

???

iTrader: (5)

new wheels are in..






edit fixed the uploaded pic

???

iTrader: (5)

you have a pm

???

iTrader: (5)

i'd kill for the mounting bracket lol

jwscab

I dug up some old northstar stuff from when i prototyped this (never got any further), and I see I made up a spreadsheet with some reference angles listed. I can't remember how I did this, but I have the northstar reference angle to be measured on the bench at around 25 degrees, not 70 as has been referred to here. I might have looked at the delta between the first pulse listed as the first notch, compared to the first 4x ref pulse, and then verified it to be close to that with a dial back timing light. I went further and calculated what the min and max retard should be, so I came up with this:

ref angle: 25
max advance: 18
max retard: -28

I dunno if they will work or not, but it's what I had calculated here. Oh, the overall advance range with these values is -3 to 43 degrees.

???

iTrader: (5)

i will try those values for sure. he's working on mounting the new wheel and making sure the sensor's are in the right place. about a week or so i'd guess and we will know for sure if its working, or if we are scrapping the idea on his car since its been down for a few months now over this and go another way.

maybe something lsx based or off the wall.

???

iTrader: (5)

i just tried emailing them about this and attached your scanned pic to it. hope i hear something



http://www.indmar.com/About/ContactIndmar/index.html

???

iTrader: (5)

afgun you are the man...

89 Iroc Z

???,
I love the idea, makes things a lot easier

- Luke

89 Iroc Z

Whoa, I can't find afgun's post with these part numbers. I am tempted to get ahold of that bracket and put it into cad.

- Luke