Results 1 to 9 of 9

Thread: Maxed out Knock Counts??

  1. #1
    Fuel Injected! beasleyrb's Avatar
    Join Date
    Apr 2020
    Location
    Arkansas
    Posts
    24

    Maxed out Knock Counts??

    I've had this engine on a stand for 2 years and never done anything with it while on other projects. I was going to rebuild it but thought I would see if it would run. After buying some parts, cleaning injectors, modifying the harness to standalone, buying an ecu and loading it with a bin it does, in fact run. I bet it hasn't run in years. maybe decades but it starts immediately and is responsive. I used every old part on it that looked good but had to buy a few things. All sensors are on it and hooked up. I bought it through a Facebook Marketplace listing. Anyway,

    It is a 94 LT1 that came out of a Fleetwood Cadillac on an engine run stand. No transmission. Firebird ECU with an EEHack " robot tune" bin code on it. I modified the harness to a standalone setup following things I found online. I'm just starting with EEhack and looking at the engine running. I've noticed that the knock counts are at 5217 from the time the engine is started. There is no knock retard. One of the knock sensors is new the other one isn't.

    What does this mean? Is the knock sensor circuit open or shorted? I suppose that's my question.

    I've attached a zipped Excel log file. I'm sure it's pretty crazy but the engine runs good. Well, to a point. I think the open headers, old injectors, and who knows what else affect it's performance. I do wonder about the knock counts. The engine is not noisy at all at idle or revved up. Thanks for your thoughts.
    Attached Images Attached Images
    Attached Files Attached Files
    Last edited by beasleyrb; 05-29-2020 at 10:34 PM.

  2. #2
    Fuel Injected! spfautsch's Avatar
    Join Date
    Apr 2015
    Location
    Montgomery City, MO
    Age
    52
    Posts
    883
    You might get better help faster if you posted the .eedata file. It's hard to tell from what's in the spreadsheet how long the engine was running, but your knock count didn't change until it died at frame / row 5609 near the end of the log.

    If your knock counts aren't increasing, it's not seeing "more" knock. And the starter bendix engaging the flywheel / flexplate could be the cause of the initial knock count. It's what I look for in a start log to know exactly when I hit the key, then I step through the log frame by frame in eehack until I see the spark advance jump to what the run table is commanding. At that point I know roughly how long it took to "get running" from the moment I hit the key.

    By the way, the knock count is stored in a 16 bit register, so it rolls over at 65575. 5700 is a typical number I might see after four to six really harsh shifts (manual trans). Having run numerous variations of roller rockers with severe spring load, I've got some logs where it's rolled over more than three times from false knock.

    Edit: sorry, I incorrectly typed that a 16 bit register overflows at 65575. Meant to type 65535.

  3. #3
    Fuel Injected! beasleyrb's Avatar
    Join Date
    Apr 2020
    Location
    Arkansas
    Posts
    24
    Thank you. Here is the EE datafile zipped up. I think I don't understand what the Knock count number means. I thought it counted how many times the knock sensor found a knock.
    Attached Files Attached Files

  4. #4
    Fuel Injected! spfautsch's Avatar
    Join Date
    Apr 2015
    Location
    Montgomery City, MO
    Age
    52
    Posts
    883
    I'll look at your log in the morning - currently into my third adult beverage and it wouldn't do much good for me to look now.

    The knock count, as I understand it, is a cumulative register. To simplify it as much as I can (with the limited understanding I have) is that the ecu starts at key on engine off with the knock count at zero. When a knock event is detected the routines assign a value to the event based on the magnitude of the signal received from the piezo sensors. This value is added to the knock count register that's displayed in eehack. Larger magnitude events generate appropriately larger knock retard. This is what you really need to pay attention to because knock retard is what will be felt in loss of torque. But the knock count is another way to identify exactly when a knock event was detected. Further along these lines, as mentioned previously the register holds 16 bits, so when the knock counts accumulate to 65535 (sixteen binary 1s) the register rolls over to zero.

    Generally there's always some initial knock count by the time the engine is running, caused primarily by the starter bendix drive engaging the driven gear ring. If you're running your stand setup without a torque converter or a flywheel mass of some significance, that could account for the relatively high initial knock count you're seeing due to the compression fighting the acceleration without any additional ballast to help smooth things out.

    Incidentally, where are you at? Your setup is begging for me to come install a diy-ltcc ignition.

    Edit: I couldn't help but look at the log even in a semi-impaired state. Seems like a really healthy engine at idle. Excellent map numbers and nothing that I would consider concerning (at first glance). The .eedata file starts at 25 seconds into running, so it's impossible to tell where your knock counts occurred, but I think it's safe to assume it was the aforementioned starter bendix.

  5. #5
    Fuel Injected! spfautsch's Avatar
    Join Date
    Apr 2015
    Location
    Montgomery City, MO
    Age
    52
    Posts
    883
    After looking a bit more carefully without the aid of ethanol, let me retract the generally healthy statement.

    I'm sure there are more knowledgeable folks here that could give better detail, but here are a couple general observations:

    Looks like your IAC is hitting zero, so the ecu is unable to command the target idle speed. Also your MAF numbers seem low (although you didn't specify the engine's displacement). All these things point to a vacuum leak. If this is a 4.3 those MAF #s are probably fine but if it's a 5.7 it would indicate air is getting into the intake that's bypassing the MAF.

    Also, the left bank BLMs indicate a lean condition, and max out when you open the throttle. Likely related to the vacuum leak.

    I'd look for a missing or cut injector o-ring or a manifold gasket broken / sucked in on the left side.

  6. #6
    Fuel Injected! beasleyrb's Avatar
    Join Date
    Apr 2020
    Location
    Arkansas
    Posts
    24
    Thank you for looking at the data. It is a 5.7. I thought I plugged all the vacuum ports that aren't used but will check for leaks. I didn't want to spend $300 on a MAF sensor because I didn't know if the engine would even run so I bought one of those Mostplus chinese parts off Ebay for $50. The low MAF reading might be the cheap part. I've seen some reviews where people said the parts worked but who knows who actually did the review. The sensor looks well made though. I am using the old injectors but did put new O-Rings on them. They were really varnished up and it took me some effort to get them clean but they did clean up. They didn't leak and did pulse on the little test and cleaning rig I made. The long open 3 inch headers probably affect things. Long way for the gases to travel to reach the 02 sensors and no back pressure at all. 02's are new of course. I read Steveo's words on tuning for headers (on his website) and did allow for some delay, etc. The IAC is the original one. I cleaned it when I cleaned the throttle body. It's probably working just not getting the correct command. I suppose that's how it works.

    I really appreciate your thoughts. I'm playing and learning about LT1's with this thing. I suppose the real world is in a car or truck going down the road. I may rebuild the engine but I have 58 PSI oil pressure and it doesn't burn oil that I can tell. You asked and I live in Western Arkansas on the border with Oklahoma.

    I'll look for leaks and put pipe plugs in the un-needed vacuum connections on the intake plenum.

  7. #7
    LT1 specialist steveo's Avatar
    Join Date
    Aug 2013
    Posts
    4,007
    just because a hose is plugged into something... doesn't mean the thing on the other end of the hose isn't leaking

    you should consider replacing those rough old injectors, tuning with home cleaned injectors is a bit crazy

    if you suspect your MAF just unplug it. if it runs better, it's suspect. the chinese one is definitely a bad idea. why not just dump the MAF and run speed density? no shame in that

  8. #8
    Fuel Injected! spfautsch's Avatar
    Join Date
    Apr 2015
    Location
    Montgomery City, MO
    Age
    52
    Posts
    883
    Speaking from experience, the chinese MAF is definitely something better suited for use as a paperweight. For the purpose of proving that the engine is healthy I'd take steveo's suggestion and run speed density to try and suss out any demons.

    Thinking further into the MAF readings, unless you have the entire intake ducting from <whatever MAF calibration you're running> connected on your stand setup, it's pretty much useless anyway. Almost every piece of the intake system can skew MAF readings. I see there's a metal elbow piped into your throttlebody, so your MAF is going to need to be recalibrated to account for that (and any other differences) assuming you can find an OE Delco MAF element. That's a job best performed on a load cell dyno.

    Whatever the case, your fully closed IAC counts still seems to point to a vacuum leak. However, you might need to reset the IAC system so the ecu knows where it's actually at versus where it thinks it at. And I don't know if that's possible without a vss signal and simulating driving. Any ideas on that <everyone>?

  9. #9
    Fuel Injected! beasleyrb's Avatar
    Join Date
    Apr 2020
    Location
    Arkansas
    Posts
    24
    I'm sure your right. Like in so many things, garbage in garbage out. The combination of old engine, cheap parts and old parts probably make tuning and trouble shooting difficult to impossible. Well, I've actually accomplished what I set out to do and that is hear it run. I'll need to decide now what to do with it. Putting it on the road would mean new "real" parts and going through it probably. Youngest son wants to buy me an old pickup to put it in but I told him I would rather have a 71 Chevelle Malibu. A car I had in college I wish I had never sold. Last year I restored a worn out Jeep CJ for him and put a 2014 L83 in it with a 5 speed. That was fun. Thanks again.
    Attached Images Attached Images

Similar Threads

  1. 24* Knock retard, no knock counts
    By hellomynameis in forum GM EFI Systems
    Replies: 14
    Last Post: 01-15-2018, 06:17 AM
  2. odd knock counts
    By mdj in forum GM EFI Systems
    Replies: 1
    Last Post: 01-23-2013, 03:18 AM
  3. Knock Counts and Knock Retard Degrees – AUJP $8d
    By 84Elky in forum GM EFI Systems
    Replies: 1
    Last Post: 01-19-2013, 07:10 PM
  4. Knock Counts and Knock Retard Degrees – AUJP $8d
    By 84Elky in forum Fuel Injection Writeups Articles and How to New and Old
    Replies: 0
    Last Post: 01-19-2013, 12:33 AM
  5. Knock counts, Knock retard...
    By RobertISaar in forum GM EFI Systems
    Replies: 6
    Last Post: 01-16-2013, 10:19 PM

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
  •