Bringing TBI and Multi Port Fuel Injection to a New Level.     EFI Conversions and Tuning! Seattle to Portland! E-mail Tuning Consultant!
Page 45 of 47 FirstFirst ... 354041424344454647 LastLast
Results 661 to 675 of 692

Thread: DIY LTCC or similar system for LT1s

  1. #661
    Fuel Injected! spfautsch's Avatar
    Join Date
    Apr 2015
    Posts
    608
    Quote Originally Posted by kur4o View Post
    ... but it also provides power to auto trans solenoids.
    Ooh, that kinda sucks. I doubt everything in the trans draws more than 1 amp peak, until one of them shorts. I'm surprised GM did that.

    In light of that information a relay setup like you describe would be my choice on an auto car.

    Quote Originally Posted by kur4o View Post
    That 3 slots explains the counts inconsistency.
    I think the three slots are to give more light transmission for a cleaner high-speed signal at the phototransistor - as opposed to the low res side which switches very slowly by comparison.

    Quote Originally Posted by kur4o View Post
    I also found why opties fail so much. It is the only japanese part in the car. There is some dot welds that eventually will broke due to high frequency switching and cause signal loss. Not an easy fix since soldering wont last long there.
    I respect everything you take the time to post, but after spending plenty of time researching my personal opinion is that the phototransistor (receiver side) eventually shorts internally from temperature. All things aside, if you put a bunch of semiconductors on a chunk of aluminum that for all intents and purposes has 215F oil sprayed on it it's whole life, failure is inevitable, and I think it's no accident that phototransistor equipped sensors have been almost completely absent from the automotive industry for nearly two decades.

    Quote Originally Posted by kur4o View Post
    Is there a way to get feedback from coil if it is blown or not delivering spark for some reason.
    Unfortunately not. Unless these coils have some internal self-diagnostics that we're not aware of it would require monitoring current to each coil. There aren't enough adc channels to start with, and you would need current sensing shunts for each coil which would increase wire and part count immensely. Maybe if misfire detection works that would help diagnose, but I'm not optimistic it will (work). If GM is doing anything like this currently, keep in mind they have a slightly larger R&D budget than Scott's basement lab.

  2. #662
    Fuel Injected! spfautsch's Avatar
    Join Date
    Apr 2015
    Posts
    608
    Latest release 0.9.38

    Found incorrect sequence detection bug - although I did luck out yesterday and manage to log an instance, it would have been easy to find had I looked. Fallthroughs in sequence detection routines were returning 0 instead of -1 which means #1 is on compression stroke.

    Code:
    int8_t detectSequence(uint8_t seqDegrees) {
    
      // note: returns the index of sequencer array, not cylNo
      if ((seqDegrees > 82) && (seqDegrees < 90)) {
    
        return -1;
    
      } else {
        if ((seqDegrees > 72) && (seqDegrees < 80)) {
          // next cyl #4
          return 2;
        } else if ((seqDegrees > 62) && (seqDegrees < 70)) {
    
          // next cyl #6
          return 4;
        } else if ((seqDegrees > 52) && (seqDegrees < 60)) {
    
          // next cyl #7
          return 6;
        } else if ((seqDegrees > 42) && (seqDegrees < 50)) {
    
          // next cyl #1
          return 0;
        }
    
      }
      return 0; // << plant forehead firmly in palm !!!
    
    }
    Also added a state variable to handle stall detection instead of using the pin 13 state (drives the yellow led). Will allow for using the status led for other purposes in the future (i.e. kur4o's suggestion).

    This hasn't been road tested yet but I will report back if there are issues.

    Update: I've driven about 100 miles on this with no issues. I further simplified the sequence detection routines to eliminate the initial error catch and just let the fallthrough handle all oddball scenarios. I only have about 25 starts on this build so it's too early to make a determination, but so far zero no-start cranks.
    Attached Files Attached Files

  3. #663
    Fuel Injected! spfautsch's Avatar
    Join Date
    Apr 2015
    Posts
    608
    Chiming in with another update.

    Spent the weekend flushing the A/C system, rebuilt the compressor, replaced the receiver / dryer and put an exact oil and refrigerant charge in the mule. Planning to daily her as much as possible over the summer and leather seats and un-tinted windows combined with old fat guys is a recipe for swampass.

    Was all pumped up to run it this morning but lost the "going" commute logs due to laptop freezing up about 1/2 way to work.

    Return log was productive. Been working on dialing in the VE tables in SD and fine tuning the A/C IAC adders to smooth out the torque spike when the compressor engages.

    kur4o if you're interested that 77c Motorad thermostat seems to be working great. In 40c ambient air [EDIT: ambient was actually like 29c, IAT was 40c] the hottest coolant temp I saw was 93c even sitting at a stop light with the A/C on max. Disclaimer, summer isn't fully here yet and things may change in a heat island caused by a traffic jam of cars with their A/C's also on max.

    Finally got positive confirmation of bouncing off the 6800rpm rev limiter and what overlap was happening at that point - things look happy there.

    Everything seems to be aces with this build. I've lost my pitch counter but have attempted to keep close track of startups and think it's around 55-60 starts with only one instance that didn't "catch" that I'm sure was operator error - anticipated a quick catch and let off the key prematurely.

    Hoping to see someone win the challenge, at which point I'll call the current build v1.0 and launch a support thread.

  4. #664
    Fuel Injected!
    Join Date
    Oct 2013
    Posts
    968
    I've been following and it's great to see your progress. I'm building a house and I just don't have time to try it this summer. I've only driven my car once this summer. I really need to get it out and clean it up and go for a cruise.

    Also, I've been thinking that just switching to a 24x system may be the way to go if I go after CNP.

  5. #665
    Fuel Injected! spfautsch's Avatar
    Join Date
    Apr 2015
    Posts
    608
    Quote Originally Posted by lionelhutz View Post
    I'm building a house and I just don't have time to try it this summer. I've only driven my car once this summer. I really need to get it out and clean it up and go for a cruise.
    I have a boat that is feeling the same amount of neglect. But I've been there (building a house) and know how that goes. Time is my most prized "commodity" next to my family. It's pretty tough to make more of it.

    Quote Originally Posted by lionelhutz View Post
    Also, I've been thinking that just switching to a 24x system may be the way to go if I go after CNP.
    I respect what you're saying, but if I had it to do over again I don't think I'd change a thing, except to have gone straight to a 394ci stroker kit instead of rebuilding it as a 350, then doing the 383. Call me nostalgic, but I think I'm going to try to keep this car married to the ecu it's maker gave it. If I end up with a 24x or 27x system it will probably be in a bargain-buy C5/6.

    Anyway... wanted to drop in with an update before I forget all the details. Found my pitch counter last week and have been keeping diligent count of starts. At this moment it says 97, and I have only the one aforementioned missed start where the operator probably let off the key before the engine "caught". Every other attempt has been successful on the first try. Today however I've had three kind of long and "slobbery" starts, but I've also been tuning on the closed loop PID tables and have flashed three different bins today (causing BLMs to be reset). In addition to that, the weather has made a wild swing from rainy + hot + humid to 27% humidity + mild, so I'm going to chalk the bad starts up to weather and fueling anomalies.

    I'm still DD'ing this car and driving the beater diesel when the weather's bad. Weather looks great for the next several days so I'm looking forward to collecting plenty of logging data to see how my O2 PID tweaks are working.

  6. #666
    Fuel Injected! spfautsch's Avatar
    Join Date
    Apr 2015
    Posts
    608
    Update - had to flip the sign back to "0 days without an accident" (proverbially). Was sitting in the garage starting it about every 30 minutes last night when it apparently flooded on the 137th start attempt. Unfortunately I wasn't in the driver's seat so I couldn't pedal to clear the apparent flood (don't ask), so no ideas on what might have happened. Naturally I wasn't logging the controller either, but it did start after a long second crank and stumbled initially like it had been flooded.

    With one confirmed problem that wasn't preceded by a tune change in 137 attempts, I'm not sure I'm going to drop what I'm doing to track this one down (if there's anything to track down). Tonight I'll have just over 1000 miles on the most recent firmware build. It's been running phenomenally and averaged 21.7 mpg over the last three fill-ups. Once I'm finished working on the tune I'll probably build another arduino to log every startup so I won't have to fumble around connecting my laptop when I jump in to run to the post office or whatever.

    On another tangent, the past week has given me several days with ambient temps in the low 90s (32C) - the first I've been able to drive it in an average midwest summer. Though I've yet to get stuck in a traffic jam (not that I'm looking forward to that), I've spent a good bit of time idling at stop lights with the A/C on max + recirc and no complaints except being able to feel the cam's 58 degrees of overlap more than I'd like.

    I'm moving on to building a DIY wideband controller so I can verify my WOT fueling. Thankfully someone beat me to that. [link]

  7. #667
    Fuel Injected!
    Join Date
    Jul 2019
    Location
    Orange, CA
    Posts
    299
    Sounds pretty amazing honestly. My project took a sideline as I had an unfortunate conversation with a curb, but the parts to fix that are on the way and if all goes to plan it’ll be fixed by the 26th. Did set me back financially though, so the coils are still delayed. Or rather, finishing the controller.

    Glad to see it appears to all be working correctly. Without any logs to see why that one start happened I’ll just chalk it up to random chance and stick to this code being good to go. I can’t wait to have it running on my end too, but yeah, need a car that tracks straight again first.
    1994 Corvette (Automatic)
    1995 Corvette (Manual)

  8. #668
    Fuel Injected!
    Join Date
    Mar 2013
    Posts
    967
    I am getting some ebay coils in the mail box pretty soon. Also waiting on some china crap connectors.
    Now trying to arrange some stock wiring look.

    Stay tuned.

    Any hints on controller`s case or some good hard gel coating similar to the pcm`s.

  9. #669
    Fuel Injected! spfautsch's Avatar
    Join Date
    Apr 2015
    Posts
    608
    Quote Originally Posted by NomakeWan View Post
    ... as I had an unfortunate conversation with a curb
    Ah that stinks. Wheel come out ok? I have my original turbines that only need a mild refinish. Tho I'd like to keep the rears to put drag radials (or burnout rags) on someday.

    Quote Originally Posted by NomakeWan View Post
    Without any logs to see why that one start happened I’ll just chalk it up to random chance and stick to this code being good to go.
    It's definitely "good enough". My ocd will prevent me from letting good enough get by when it comes to starting, so I will eventually build a passive, autonomous logger and try to get to 5 nines in the starting reliability. But for now I'm really interested in perfecting the tune.

    Quote Originally Posted by kur4o View Post
    Stay tuned.
    If you knew how cheap I am, you'd understand how significant this statement is. I'll never be happier to give $100 away as the day someone wins the challenge. Trust me, I'm tuned.

    Quote Originally Posted by kur4o View Post
    Any hints on controller`s case or some good hard gel coating similar to the pcm`s.
    I wish I had something on the case. I'm using an extruded aluminum box tube that fits really nicely and came from another DIY type project that one of my co-workers gave me for parts. Try as I might, I've been unable to find a source for the extrusion, and I highly doubt these were made to order but more likely cut to length. I think there are some pictures about 20 pages back with the prototype board in it.

    Somewhere around those pictures of my case will be a pic of the silicone sealant I've been using to seal the wires going into the acrylic case ends I made. I got this off Amazon and I believe it's intended for use as a conformal coating. I just haven't tried it for that purpose (yet).

    Edit: the case pictures mentioned were here (post # 356).

  10. #670
    Fuel Injected!
    Join Date
    Jul 2019
    Location
    Orange, CA
    Posts
    299
    Quote Originally Posted by spfautsch View Post
    Ah that stinks. Wheel come out ok? I have my original turbines that only need a mild refinish. Tho I'd like to keep the rears to put drag radials (or burnout rags) on someday.
    The wheel was actually what absorbed about 1/3 of the impact, disintegrating the lip. Thankfully the tire survived unscathed despite going flat pretty much instantly. The toe rod took the rest of the kinetic energy, bending at the threads. I've got the old tire mounted onto a new wheel already, just waiting for UPS to deliver my shiny new heim joint toe arm assembly so I can get back on the road.
    1994 Corvette (Automatic)
    1995 Corvette (Manual)

  11. #671
    Fuel Injected! vilefly's Avatar
    Join Date
    Sep 2017
    Age
    49
    Posts
    217
    Drat. I'm still waiting for a chance to get both sides of my frame on the wagon welded up and aligned. Was a little rough on it inside the roundabout exits in my area. Weak factory welds gave way right where the rear lower control arms attach, just ahead of the rear wheels. The body shops are now open.....and flooded with business, because everyone forgot how to drive again.

    Congratulations on the successful debugging, spfautsch.

    It looks like everyone will be battling the shipping demons that I have been dealing with. Nothing but delays these days.
    Gawd. I am so late to the game again. Ack.

  12. #672
    Fuel Injected!
    Join Date
    Mar 2013
    Posts
    967
    I almost assembled the kit but I am having some questions.

    There are 7 0.1uf caps and 8 positions on the board. Should I get one more and populate all 8 positions or 1 can be omitted.

    The sts resistor have no value printed. What is the suggested value there.

    I was hoping to stuck in there a bluetooth module. Is it safe to source a 5v to it from the board regulator.

    What is the default baud rate used for communication.

    Do we need some heat sink for cpu or voltage regulator. I found the perfect box from a 2.5 external hdd but it is plastic. I do have some aluminum too but it is too low to clear the 100uf 16v capacitor. Can I substitute it with something shorter.

    And last question are the leds loads the system with more heat and power consumption. What are the gains if not installed.


    I found some good coatings on amazon but the price is ridiculous and it will add 50% to the budget build. I will search for something more affordable locally.


    A thought on future improvement. The 26 wires soldered to the board really spoil the good feel of the controller and are some weak point on bulletproof design. The wishlist is to get the outputs lined up so they can take some 2x16 connector like the PCM style connectors.

  13. #673
    Fuel Injected! spfautsch's Avatar
    Join Date
    Apr 2015
    Posts
    608
    Quote Originally Posted by kur4o View Post
    There are 7 0.1uf caps and 8 positions on the board. Should I get one more and populate all 8 positions or 1 can be omitted.
    Yours shipped before the addition of the coolant sensing on the spare analog IN1. If you can find a 0.1uf do it. Otherwise, if you omitted one I would leave out C1 which is a decoupling cap for the 12v input.

    Quote Originally Posted by kur4o View Post
    The sts resistor have no value printed.
    If there's no value printed on the board the component is optional. C20 for IN1 should be the only exception to this "rule". The STS solder pad is for optionally relocating / duplicating the status led so it depends on what component you're driving externally. O1 and the resistor R29 next to it serves the same purpose for the OUT1 led. So if you want those leds remotely located just move the 1k current limiting resistors from R9 and R26 to R29 and R30.

    O2 and O3 and their associated r-c networks should remain unused / upopulated.

    Quote Originally Posted by kur4o View Post
    I was hoping to stuck in there a bluetooth module. Is it safe to source a 5v to it from the board regulator.
    It depends on how much current the bluetooth module draws.

    Quote Originally Posted by kur4o View Post
    What is the default baud rate used for communication.
    115200,8,N,1

    Quote Originally Posted by kur4o View Post
    Do we need some heat sink for cpu or voltage regulator.
    Shouldn't need anything on the AVR, but if you're going to source power to your bluetooth module from the 5v regulator it wouldn't hurt to add something for a safety margin.

    I would highly recommend lightly soldering your bluetooth module in and testing that you're able to upload firmware and log without resetting the AVR before you get too far into packaging things up.

    Quote Originally Posted by kur4o View Post
    ... it is too low to clear the 100uf 16v capacitor. Can I substitute it with something shorter.
    You could go to a lower voltage but not capacitance. But I think you'll be hard pressed to find a 100uf 10v radial can cap in a shorter package.

    Quote Originally Posted by kur4o View Post
    are the leds loads the system with more heat and power consumption. What are the gains if not installed.
    The power, status and out1 leds could be handy for troubleshooting if things don't work out of the gate. The coil indicators may also help in this regard, but the amount of power they consume is fairly marginal. Unfortunately I think you got a version 1.0 pcb so there are no jumpers to easily take these out of the circuit. My recommendation is to install the power, status and out1 leds minimally and omit the coil indicator leds. If you have issues getting things to work we may need to put some of those in.

    Quote Originally Posted by kur4o View Post
    A thought on future improvement. The 26 wires soldered to the board really spoil the good feel of the controller and are some weak point on bulletproof design. The wishlist is to get the outputs lined up so they can take some 2x16 connector like the PCM style connectors.
    The reason for the solder pad wiring is well, try finding a source for such a weatherproof connector and a mate at a price that doesn't bust the budget. Trust me, I wanted all this and more such as a case, but the kits would have cost $250 instead of $25.

  14. #674
    Fuel Injected!
    Join Date
    Mar 2013
    Posts
    967
    Thanks for the in depth explanations. I will look for some caps locally.

    The bluetooth module draws around 40ma peak. I look at the voltage regulator max output is 1.5a normal, 2.5a peak. I am planning to put an external box[It will get to tacked up inside the 2.5hdd enclosure] for the bluetooth with on/off switch and rip some 5v from r1. If it doesn`t work I will add external 5v source to the module.

    Any idea how much those leds draw. If I put higher value resistors will they lower consumption.




    Now the main issue is uploading firmware and fixing the crank tables. I will need some links and brief step by step instructions.

    Is there any test mode or the only way to test is to put in an opti signal and give it a spin. My main concern is bad batch of components or bad soldering skills..

    Edit:
    Just to make sure the shorter red coded 25v can cap goes to c3 and the longer 16v purple coded goes to c4


    Edit2:

    Most of the issues are cleared I had to pay more attention to the readme files in the latest upload. Great write up, all clear now, unless how reliable is firmware upload. Should I be worried of bricking or it is pretty straight forward and safe.
    Last edited by kur4o; 1 Week Ago at 01:55 PM.

  15. #675
    Fuel Injected! spfautsch's Avatar
    Join Date
    Apr 2015
    Posts
    608
    Quote Originally Posted by kur4o View Post
    The bluetooth module draws around 40ma peak.
    I don't anticipate that causing any kind of problem for the regulator. If you're worried try finding a sheet metal type heatsink for it and you should have plenty of safety margin.

    Quote Originally Posted by kur4o View Post
    Any idea how much those leds draw.
    Discussed here [link] (fourth paragraph of first post).

    With all leds on, and all 8 coil igniters being driven high total load on the AVR would be around 36 milliamps. This doesn't concern me whatsoever because in operation there will only be 4 to 5 coils being dwelled at max rpm, and the status led should be off.

    Quote Originally Posted by kur4o View Post
    Now the main issue is uploading firmware and fixing the crank tables.
    Go here [link] and install the latest Arduino ide package for your OS platform.

    When you launch the Arduino IDE for the first time I think it asks you where you'd like to store your sketches, and that will probably default to a folder under Documents named 'sketchbook'. I haven't used it on windows in at least 10 years so this could have changed, but I think you should be able to figure that out.

    Extract the latest firmware to your sketchbook folder. The file / folder structure should look something like:

    (Documents)\sketchbook\diy_ltcc_0_9_38\diy_ltcc_0_ 9_38.ino

    Find that .ino file in windows explorer and double-click it. You should see a number of header (.h) files and several .ino files (no extension will be displayed in the IDE for these) represented as tabs at the top of the editor window. Find config.h and click it.

    The default build is set to use the D580_dwell.h table - if you have different coils edit the include line appropriately.

    You can fine-tune your voltage divider resistor values here as well - plug in the values measured for R1 and R2.

    Scroll down to the definition of cranking spark table and put the values from your tune in the second column.

    Code:
                                                         v 0x12075 table goes in this column
    const uint8_t PROGMEM crankSpkTable[17][2] = {{  0, 14},  //  -40c
                                                  { 16, 14},  //  -28c
                                                  { 32, 14},  //  -16c
                                                  { 48, 14},  //   -4c
                                                  { 64, 13},  //    8c
                                                  { 80, 13},  //   20c
                                                  { 96, 12},  //   32c
                                                  {112, 11},  //   44c
                                                  {128, 10},  //   56c
                                                  {144,  9},  //   68c
                                                  {160,  8},  //   80c
                                                  {176,  7},  //   92c
                                                  {192,  6},  //  104c
                                                  {208,  5},  //  116c
                                                  {224,  4},  //  128c
                                                  {240,  4},  //  140c
                                                  {255,  4}}; //  152c
    Click the save button or press [ctrl]+[s].

    In Arduino IDE go to Tools > Port and select the com port your bluetooth adapter appears as.

    Then go to Tools > Board and select 'Arduino Uno'

    Click the check button or press [ctrl]+[r] to test compile the sketch. If no errors you can try uploading ([ctrl]+[u]). If you get any errors uploading follow the troubleshooting links that are presented. If you get a timeout from avrdude check that your bluetooth dongle is properly controlling the DTR pin. The drivers for the common bluetooth boards don't support this in the linux kernel, so I can't upload firmware using them. But I believe they work in windows.

    Note: always disconnect the DTR pin from serial device you're using when operating the board. DTR must only be connected for uploading firmware to the board.

    Quote Originally Posted by kur4o View Post
    Is there any test mode or the only way to test is to put in an opti signal and give it a spin.
    Different test modes are documented at the top of config.h. Simplest test if you have all the coil indicator leds installed and an opti signal is to uncomment

    Code:
    #define STANDALONE 1
    and upload. Do not use this mode with coils attached unless you're ready to fire the coils. Spin the opti or crank the engine, and you should see the controller find sequence and begin a light show on the 8 indicator leds.

    Quote Originally Posted by kur4o View Post
    Just to make sure the shorter red coded 25v can cap goes to c3 and the longer 16v purple coded goes to c4
    Correct

    Quote Originally Posted by kur4o View Post
    Should I be worried of bricking or it is pretty straight forward and safe.
    The bootloader on the AVR is pretty robust, and it wont let you write any of the fuse bytes when programming over the UART. So you'd really have to make an effort if you wanted to brick it. The only time I've killed a 328P based system is by applying reverse polarity power. I'm not even sure it killed the AVR, but the 7805 regulator shorts immediately in this situation. I did this to my original prototype board when I started testing coils, and the AVR is soldered in so it's not worth the time to try removing the 328P.

    P.S. - the reverse polarity "trick" mentioned above is also a really fast way to destroy a coil pack. Ask me how I know.

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
  •