Page 35 of 55 FirstFirst ... 25303132333435363738394045 ... LastLast
Results 511 to 525 of 825

Thread: DIY LTCC or similar system for LT1s

  1. #511
    Fuel Injected! spfautsch's Avatar
    Join Date
    Apr 2015
    Location
    Montgomery City, MO
    Age
    52
    Posts
    883
    Interesting idea, but I suspect if you lose one injector fuse O2s are going to see a massive lean condition and immediately max out BLMs on both sides. I guess that's better than raw fuel being pumped through the engine. If my memory is correct, on the y-body the injector circuits are split with 1,4,6,7 on one and 2,3,5,8 on the other.

    I spent several more hours over the weekend to no avail looking for my spare connector / pigtail that will work with the D580 coil. Local parts places want $30 for one so I ended up getting one off ebay for $13. Really burns me because I know I got a pair of them with the header kit and only cannibalized one that I still use on my test bench rig.

    I don't anticipate having time to do thorough testing of the coils until into the new year, but should be able to test igniter current tomorrow or the next day on the D580.

    I guess I'm going to need to find or build a good variable DC power supply in order to test at different voltages. I'm also going to be on the lookout for a small toaster oven to bring the coils up to ~100c.

  2. #512
    Fuel Injected!
    Join Date
    Jul 2019
    Location
    Orange, CA
    Posts
    757
    A set of 8 coils with harnesses and brackets for $50!? Sign me up. Heck even if that’s 4 coils with harness and bracket for $50 it’s a good deal. Best I found was 8 bare coils for $140.

    Just gonna grab a pair of coil harnesses and extensions separately for now and figure out a mounting solution once I have it all working.

  3. #513
    Fuel Injected!
    Join Date
    Mar 2013
    Posts
    1,470
    If my memory is correct, on the y-body the injector circuits are split with 1,4,6,7 on one and 2,3,5,8 on the other.
    You are correct the lt1 injectors are not wired per bank. On ls1 are wred per bank, guess the different firing order allows per bank engine operation.
    On lt1 4 cyl limp mode is as 1,4,6,7 or 2,3,5,8.
    I am back on the drawing board. Rewiring the coils to match the injectors will be pain, so I guess I will have to do it or live with the consequences. It will be interesting to do a force open loop in the event a injector or coil to fail.

    Did you thought about the ltcc to listen at the aldl stream and extract some data from it. Anyway all lt1 pcms stream some data already, why not use it. It won`t work during logging so I guess it is a bad idea.

    I guess I'm going to need to find or build a good variable DC power supply in order to test at different voltages.
    It must support at least 5-10 ampers output. Not an easy task to handle. Long and thin cable will also reduce the voltage a little. If you have 14 volt battery source and reduce from there by putting higher load on a thinner, longer cable to get the desired voltage drop.
    d585 are known to self ignite on longer dwell time. Other coils might just melt so be cautious. I am sure GM data is good starting point and there will be very limited tweaking over there.

    I was looking for various connectors and they are really overpriced on ebay. I have to dig some part numbers.

  4. #514
    Fuel Injected!
    Join Date
    Mar 2013
    Posts
    1,470
    Quote Originally Posted by NomakeWan View Post
    A set of 8 coils with harnesses and brackets for $50!? Sign me up. Heck even if that’s 4 coils with harness and bracket for $50 it’s a good deal. Best I found was 8 bare coils for $140.

    Just gonna grab a pair of coil harnesses and extensions separately for now and figure out a mounting solution once I have it all working.

    Good luck.
    https://www.ebay.com/itm/Ignition-Co...temCondition=4

  5. #515
    Fuel Injected!
    Join Date
    Jul 2019
    Location
    Orange, CA
    Posts
    757
    Ah, used. That makes much more sense, I was referring to new. Thanks though, good resource!

  6. #516
    Fuel Injected! spfautsch's Avatar
    Join Date
    Apr 2015
    Location
    Montgomery City, MO
    Age
    52
    Posts
    883
    Quote Originally Posted by kur4o View Post
    I am back on the drawing board. Rewiring the coils to match the injectors will be pain, so I guess I will have to do it or live with the consequences. It will be interesting to do a force open loop in the event a injector or coil to fail.
    No offense but I think you're overthinking things a bit. Granted - it would be better to have the ability to limp home, but the car's gonna run like crap. Seems like a lot of extra wiring for something that should never happen (unless a wire gets pinched).

    Quote Originally Posted by kur4o View Post
    Did you thought about the ltcc to listen at the aldl stream and extract some data from it.
    The only parameter I can think of that would be marginally useful that we couldn't easily measure with the controller is coolant temp. And I'm afraid the overhead of parsing the ALDL data could potentially cause starvation in the main loop which could cause missed or delayed dwell. It would be much easier to just read ECT with one of the unused ADC inputs. But I'm not convinced temperature compensation is necessary. Remember - the Bailey LTCC kit only connected to the opti and EST lines. It's unknown whether it compensated for operating voltage.

    Quote Originally Posted by kur4o View Post
    d585 are known to self ignite on longer dwell time.
    I fully intend to test and document dwell limiting behavior. In my initial tests I actually found that the D514a coil seemed to discharge a lot when it wasn't expected to. I didn't notice this with the D585 or the 8183. But I didn't test that function methodically - this is just something I observed casually.

    Quote Originally Posted by kur4o View Post
    I am sure GM data is good starting point and there will be very limited tweaking over there.
    I'm not as convinced as you are.

    I spent a good bit of time one night last week looking over the bin dumps you posted back here [link]. It raised more questions than it answered. For instance looking at the main dwell table for the 98 f-body ls1 and the 02 ls6 they seem to match exactly.

    For a frame of reference, here's the 2000RPM dwell row.

    Code:
            0      1      2      3      4      5      6      7      8      9     10     11     12     13     14     15     16     17     18     19     20     21     22     23     24     25     26
    2000  26.00  26.00  26.00  26.00  26.00  26.00  26.00  17.29   8.60   7.29   6.00   5.24   4.50   4.04   3.60   3.30   2.99   2.99   2.99   2.99   2.99   2.99   2.99   2.99   2.99   2.99   2.99
    Then there's a dwell multiplier on coolant temp that's also voltage compensated.

    Code:
            -40      -20       0        20       40       60       80      100      120      140  
    13    1.00000  1.00000  1.00000  1.00000  1.04004  1.11499  1.14502  1.17993  1.20996  0.99512
    Just looking at the results of the 13v range, this gives a minimum dwell below 40c of 4.04ms, increasing with coolant temp to a maximum of 4.89 at 120c and then tapering back at 140c. Based on the megasquirt info here they're calling for a target of 5.6-5.8ms, yet in the GM tunes that never happens except at voltages below 10. Baffling.

    This same range in the 2014 vette tune (13v@2000rpm) has a minimum dwell of 2.88 and max of 3.2, also increasing with coolant temp up to 140c where it seems to be reducing as a protection mechanism for an overheating condition. I presume this application would have been equipped with the 8183 coils.

    Moving on to the 2004 Yukon tune that would have had the D585 coils the same range has maximum dwell at 20c of 4.59ms and then drops to a floor of 3.52ms at 60c and stays there. This is completely opposite of what the other temperature compensation tables do.

    Then there's a dwell vs MAP multiplier table in the LS6 and the Yukon tune, but every cell is effectively set at 1 (0.9999999). Actually the Yukon bin dump has suspect data in this table - 8.1567373 from 20 to 70 kpa then 0.9999999. And there's no reference to dwell vs MAP in either the f-body LS tune or the 2014 vette.

    Looking at the differences and commonalities, it's not terribly clear what the function of the temperature compensation is.

    The only universally common data here is:
    1) dwell is increased as voltage decreases
    2) dwell is reduced with RPM

    This last item is patently baffling to me. In a coil per cylinder setup you have the ability to increase dwell at higher RPMs, but these tables seem to mimic how a distributor setup would be mechanically dwell limited.

    I'd like to think "surely there's a reason GM did this" but I'm not necessarily convinced.

    Quote Originally Posted by kur4o View Post
    I was looking for various connectors and they are really overpriced on ebay. I have to dig some part numbers.
    If you wanted to make your own harness, I recently bought a set (8) of the newer LS2 style connectors / pigtails off ebay for $25 shipped here.

    I have a 120w 20v power supply (about 6.2 amps) for an all-in-one PC that I think will probably suffice. But I'll initially test primary current with a battery so I know what to expect. My guess is the D585 coils will have the highest primary current, but we'll see.

  7. #517
    Fuel Injected!
    Join Date
    Mar 2013
    Posts
    1,470
    Found some good source for all kind of connectors. The hard part will be to identify the needed connector , terminal pins and insulators.

    How many different connectors the coils use.

    Ah, used.
    A coil is a coil no matter how much abuse it had. As long as there are no physical damage[melted, cracked or shorted] and corroded contact points it must act as new.
    It is only copper wire, no wearable parts in it.

    I will take used oem over new aftermarket all day long. So it will be safe to get used coils if they are oem and are working.

  8. #518
    Fuel Injected!
    Join Date
    Mar 2013
    Posts
    1,470
    That 2004 file looks a little weird. Some of the conversion might be wrong.
    I see in all calibration the max and min multipliers are capped at 1.3 to 0.9. So the multipliers might be added together before beeing applied to main table. On 98 ls1 there are some crank dwell settings that might be not defined on the later bins.

    The coolant must be converted to some coil heat soaked scalar matrix, so the pcm predict the temperature of the coils based on coolant, engine runtime and iat. I am sure alot of dwell tables are missing from the files. I will have to start that dwell disassembly to figure all of the unknowns.
    The estimated temp of the coils might be used to compensate for the copper resistance change related to temperature. The higher the resistance the more current is transferred to heat , the less efficient the coil gets. The more likely a melt down condition to occur real fast.

    The dwell might decrease with higher rpm for more that enough reasons.
    I can only speculate. It could be pcm software/hardware limitation of firing more than x coils at a time.
    Not enough time to cool the coil between cycles.
    Keeping coils colder to extend service intervals.

    On a side note truck coils are 2 different designs and are interchangable within the same tune. First design have the mitsubishi star logo and the later are the d585 delco units. They use different lenght spark plug cables due to different height and are only identified by visual inspection. Can`t be figured just by vin number. So the tune might be for one of the coils and not optimized for the other or somewhere in between.

  9. #519
    Fuel Injected! spfautsch's Avatar
    Join Date
    Apr 2015
    Location
    Montgomery City, MO
    Age
    52
    Posts
    883
    Quote Originally Posted by kur4o View Post
    That 2004 file looks a little weird.
    Agreed.

    Quote Originally Posted by kur4o View Post
    I see in all calibration the max and min multipliers are capped at 1.3 to 0.9. So the multipliers might be added together before beeing applied to main table.
    I had the same thoughts, but asked myself "self, do you even care". No, I don't think I do.

    Quote Originally Posted by kur4o View Post
    The coolant must be converted to some coil heat soaked scalar matrix,
    I understand physics as well as the next guy and have lots of first-hand carnal knowledge of what happens to transistors when they're driven into saturation after reaching cascade temperature. But it seems like the calibrators were making the majority of their dwell modifications using the temperature tables when it made absolutely zero sense to do so.

    Quote Originally Posted by kur4o View Post
    I can only speculate. It could be pcm software/hardware limitation of firing more than x coils at a time.
    It sounds like you might be reaching. If the TPU in the $ee era PCMs could handle smoothly sequencing 8 injectors independently to 100% duty cycle, the newer ones can surely handle something a dipstick of my caliber cooked up in a few months on an 8 bit mcu.

    You're a real soldier kur4o. I love that you'd rather stare at machine code for hours to solve a problem than to strike at it with a large hammer. I appreciate the value of reverse-engineering, but know when to use the hammer.

    I've read enough of the Banish material to have a lot of faith in the GM calibrators and the reasoning behind their choices. But everything you stated in your previous post could be dismissed with thorough, empirical testing. Why don't we do that?

    Keep in mind that we're making something completely from scratch that isn't designed to reduce costs, emissions, minimize warranty claims, or otherwise serve the financial interests of General Motors Corporation. My only concern is that the thing runs and performs as well as it possibly can, period. If that means I need to replace all the coils after 75,000 miles vs 150,000 I'm fine with that.

  10. #520
    Fuel Injected!
    Join Date
    Mar 2013
    Posts
    1,470
    That 04 file is not correct. It is set as the other files 1 for map multiplier and 0.9 to 1.3 for the other tables. The start address of the 2 tables is switched.

    SIgn me in when the experiments starts. I have no idea how the output will be measured but running the max of the coils is always good to do. I suggest 90% exploit to meling point.
    The coolant table makes sense when you look at the usefull data, between 7 and 16volts. I have seen different GM strategies running the same output. Higher base values with big multiplier offset and lower base values with opposite multiplier produce the same result at the end. I have no idea why they do it that way.

    There are definitely some scalars for dwell used at stratup. You can consider adding something similar to the ltcc logic.

    I am too far from complete disassembly so can`t give much useful info how it is used.

  11. #521
    Fuel Injected! spfautsch's Avatar
    Join Date
    Apr 2015
    Location
    Montgomery City, MO
    Age
    52
    Posts
    883
    Quote Originally Posted by kur4o View Post
    The coolant table makes sense when you look at the usefull data, between 7 and 16volts.
    I was ignoring everything outside of 7-16 volts. They would make more sense if the same patterns were demonstrated throughout all the tunes. It just seems very haphazard.

    Quote Originally Posted by kur4o View Post
    I have seen different GM strategies running the same output. Higher base values with big multiplier offset and lower base values with opposite multiplier produce the same result at the end. I have no idea why they do it that way.
    Neither do I, and that's exactly why I'm not placing a lot of faith in this data. In addition, for all we know the definition files for the bins may have incorrect conversion formulae. Without a good understanding of how all these tables interact and exactly what results they produce, I'd rather rely on measured data and come up with sane dwell targets based on that.

    Quote Originally Posted by kur4o View Post
    There are definitely some scalars for dwell used at stratup. You can consider adding something similar to the ltcc logic.
    Startup is kind of a dangerous area and I'm not sure I want to try emulating what the factory stuff does there because I want to avoid any possible built-in dwell limiting. At low voltages and very low cranking speeds the target dwell can reached in just a few degrees of engine rotation, and without interpolation a big jump in degrees of dwell could easily double or triple dwell time. We don't want a coil firing well before the EST circuit commands it with the starter bendix engaged.

    Edit:

    Quote Originally Posted by kur4o View Post
    running the max of the coils is always good to do. I suggest 90% exploit to meling point.
    I guess you were zinging me here - sorry I overlooked this initially.

    Whatever the case, it wasn't my intention to target saturation or peak spark energy as the target dwell. This is sort of why I'm confused by the lack of MAP compensation in the factory tunes. My thought process is that I find the dwell for each coil that produces maximum spark energy for a given input voltage. Then use MAP to ramp up to that dwell at 100kpa - definitely a parabolic type curve where the adders are near zero until 65-70kpa.

  12. #522
    Fuel Injected!
    Join Date
    Mar 2013
    Posts
    1,470
    Here are some more files from 07-08 trucks.

    Intersting the pcm is a newer design but the tables are the same style and format.

    If you add a start up spark table to ltcc and ignore the pcm signal at starting, might give much better starting time.
    Some kind of semi independant mode found in some icm modules.

    I suspect a 10ms dwell for square coils and 7ms dwell for round coils. Readily compensating the voltage drop at cranking.
    Attached Files Attached Files

  13. #523
    Fuel Injected! spfautsch's Avatar
    Join Date
    Apr 2015
    Location
    Montgomery City, MO
    Age
    52
    Posts
    883
    Quote Originally Posted by kur4o View Post
    Here are some more files from 07-08 trucks.
    Can't extract the archive. Will try some other tools tomorrow, but your initial analysis makes me think I'm time ahead not even looking at them. Full disclosure - I have a bowtie tattoo, so you're talking to someone with tons of GM brand loyalty. But I'm not entirely convinced the calibrators had any earthly clue WTF they were doing with dwell on these coil per cylinder setups, and just went with "what worked", or "what the last guy did".

    Thinking about it during my commute, the only reason I can concoct for reducing dwell linearly with RPM is spark discharge time / coil "recovery" time. So I guess I have another parameter to test for when I get around to the task at hand.

    Quote Originally Posted by kur4o View Post
    If you add a start up spark table to ltcc and ignore the pcm signal at starting, might give much better starting time.
    I already have a default spark advance # built-in that will help dwell happen before the controller has been able to pick up advance from the EST line. The problem here is that depending on where the engine came to rest, the controller needs to see a certain amount of data from the opti (i.e. a certain number of crankshaft degrees) before sequence can be determined and dwell can safely be initiated. Whereas the $ee PCM can start firing the ignition immediately after the first TDC / low res pulse because it's 100% batch mode on the spark side. This plays into my cautious attitude with excessive cranking dwell. Accidentally fire a coil on a cylinder that just started the compression stroke on a cold (i.e. pig rich) engine, and there could be carnage. Like the outboard ear of the block being jettisoned when the outer starter bolt reaches it's maximum elastic deformation. I've seen it happen on a hydrolocked 4.3 Mercruiser engine, and it's not pretty.

  14. #524
    Fuel Injected!
    Join Date
    Jul 2019
    Location
    Orange, CA
    Posts
    757
    Quote Originally Posted by spfautsch View Post
    Can't extract the archive.
    It's a WinRAR archive--remove the .zip that he added just to be able to upload it as an attachment. Worked fine on my PC using WinRAR. Not sure what your compression tool of choice is, but just putting it out there. Anyway I've attached a ZIP version here just for kicks.
    Attached Files Attached Files

  15. #525
    Fuel Injected!
    Join Date
    Mar 2013
    Posts
    1,470
    just use an amps clamp on the 12v in and measure charge time to were it current limits > easy < charging past this point ie more dwell just overheats the coil
    just google coil & current probe for a detailed explanation.
    We do it all the time to set dwell with aftermarket ECu's and different coil combo's.
    Cheers,
    Mike
    I was searching on a dwell meter and stumbled upon this. It looks as the rigth way to measure the max dwell a coil needs.
    The voltage and coolant corrections are not something that can be easily measured.

    Someone knows a good, DYI method of measuring dwell time and the current that the coil draws from the pcm.
    I think that the resistors are kind of current limiting device protecting the chip. So how much is needed to ignite the coil and can that be measured reliably without a scope or some fancy equipment.

    I managed to confirm the resistors value on the 1mb pcm with blue color. It is again 430 ohms.

    The baseline of measuring is 12 volts. At 12volts the voltage correction must be zero. Below 12 volts the correction is positive and above 12volts the correction is negative.
    For the coolant baseline of measuring we have different settings for different calibration. It might be that the base dwell table is done at that temperature and coolant correction is offseted from there.
    We have coolant baseline at 100*C -98 ls1, at 125*C -14 vette and at 20*C -04 d585. That might explain the weird calibration data accross the years and platforms.

    You can be suprised but the end of injection target calculations in ls1 pcms is borrowed from $ee code, so it is continuous code upgrade in the years. If you missed the beginning, it will be very hard to understand the later code.

Similar Threads

  1. Which TBI system is better?
    By KeyAir in forum GM EFI Systems
    Replies: 41
    Last Post: 05-13-2019, 09:39 PM
  2. Hard start 93 LT1 with LTCC Ignition Mod
    By beestoys in forum GM EFI Systems
    Replies: 0
    Last Post: 05-18-2015, 08:58 AM
  3. ABS system?
    By K1500ss4x4 in forum Gear Heads
    Replies: 3
    Last Post: 02-06-2014, 06:21 AM
  4. Vortec EGR System?
    By EagleMark in forum OBDII Tuning
    Replies: 40
    Last Post: 06-02-2013, 10:07 PM
  5. Quicker way to do Spark Hook test on the street for LT1s and others?
    By sherlock9c1 in forum Fuel Injection Writeups Articles and How to New and Old
    Replies: 15
    Last Post: 03-03-2013, 01:52 AM

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •