Backfiring is killing me!!!
I can't Kill this backfire and it's driving me BONKERS!!!
AMC 401, warmed over, 454 TBI, Dizzy locked at 0, computer controlled timing, brand new MSD TBI coil, brand new Taylor 8mm spiral core wires, brand new Autolite Copper core plugs (gapped at .0454), New brass terminal Duraspark wide cap, new brass terminal rotor.
My truck runs AMAZING when it is stone cold and going less than 50MPH. If I accelerate to get on the freeway, at about 60 or 65 it starts to backfire out the exhaust. It has been doing this since my rebuild finished (including the installation of my EFI). I have replaced rotor, cap, wires, plugs, and coil. When it is warm, under steady throttle, at 55 or so, it will backfire and run pretty poorly.
1) Could I have a Mechanical engine problem causing this? Cam timing? pushrods too long? obstruction in my intake? leaks out my injectors? Valve issues?
2) I plugged my PCV port for now. I plugged all the other vacuum ports for now, other than trans modulator and brake booster. I watch my MAP sensor and my vacuum is pretty darn steady so I assume I don't have a vacuum leak. Maybe I do?
3) I don't hear any exhaust leaks... My 4 wire O2 Sensor is in my header collector on the passenger side (long tube headers), going to a 3" Y-pipe and out through a single 3" exhaust, no cat, but I do have a Borla Turbo muffler. Could it be an exhaust problem?
I am running out of ideas. I am starting to think I need to pull my EFI and go back to carb... Not sure where to go! This is killing me!!!!
No question/idea is to basic or out of the box. I am not too proud to accept ideas or questions that someone with common sense would have thought of.
I would check your valves first, can be a valve hanging up when warm, but it sounds like an electrical issue when it gets hot also, coil pick up , ignition box etc.
Are you able to get data from your EFI computer?
Am I right to assume that backfiring would only happen if:
a) There are momentary ignition lapses
b) Engine is running super rich and dumping fuel down the exhaust. What's your O2 sensor reading?
I agree with this. Good place to start.
Did you run a carburetor after the rebuild, and did it do this when you were running the carburetor but before the EFI was installed?
If running super rich this would happen hot and cold, miss on the ignition side a possibility since its only happens when hot( ignition module, distributer pick up, coil) or mechanical, valve sticking when it gets hot. You may need to check your temp sensor, could be reading wrong causing it to lean out severely. Did you run this with a carb before, and it ran ok? then you have issues on the injection side. go back and recheck everything.
Has a compression check been done to confirm valves are completely closing?
If valve train checks out...The easiest next check is to pull your distributor and stuff in an HEI dizzy (its only a 1 wire hook up)....if the problem goes away...you know its ignition related......if not I'd pull only as much of the FI as needed and plunk a known good carb on.
a) In addition to ignition lapses, it could be ignition timing errors (igniting very late, or too early).
b) O2 sensor looks "normal" in my data log. I plan to run open loop (disconnect my O2 sensor) and see if that makes a difference. I also plan to spend a lot of time in my data logs while I am on travel this week to see if I can find anything else.
Nope, I was a knuckle head and sold the carb. (more below)
I have not touched my rig in a few days (other home tasks and life keep getting in the way).
Here is where I stand.
1) This week I will order a new/good adjustable pushrod and am going to get the correct pushrods. Whether they are causing this issue or not does not mean I should not fix them. They will get corrected.
2) I need a data log in open loop. Going to run the stock 454 tune in open loop (no O2 sensor) and that could tell me if it is my fuel tables are in the ball park. Also, if my O2 sensor is really hosed, it may run better in open loop.
3) I am going to spend a lot of time looking at my logs this week because I am still new at this. Maybe the issue is obvious but I am not seeing it.
4) I ordered a cheapie distributor cap so i can cut a hole in it and verify my Hall sensor phasing (idea from a guy on theamcforum). I have a little hope my issue may be here. When I locked my hall sensor down, it just kind of laid it where the wires wanted it to be and locked it down. I wonder if it is off and when i got to max timing, it skips to the wrong post?
5) I have a carb spacer between the intake and the TBI adapter. Going to pull that out for now.
6) If, after all that, if I am still backfiring, I am going to put my duraspark back in. I still have the module, wiring, and coil so I just need an actual dizzy. I am pretty sure I can find one and stab it in there. I would also check compression at this point.
7) If that still does not work, I will buy a carb, swap fuel pumps, and try that.
I am tossing the idea around of a new hall sensor. The one in the dizzy is the one that was in my truck when I bought it (it looks old and a little crusty. But when I look at my timing and RPM in my data logs, it looks OK. I would assume if it was going, I would see either spikes or something, right?
Very good. I was going to suggest looking at your tach input very closely and seeing if you are getting the correct counts just before, and during, the backfire. That's always step #1. Sometimes a lean/rich backfire or misfire can actually cause the tach signal to drop out, which then causes the problem to amplify.
Right now, I'm of the opinion that this is a tuning issue, and your fuel table in the MAPxRPM ranges at these higher steady loads is messed up. You are not seeing the issue below 55mph at the same load because the engine isn't spending enough time in those RPM areas as you romp it. I'd verify the fuel and spark tables are correct for those ranges before moving on.
All of this occurs when it's cold? So when it's cold and you get up to 60/65 it still backfires?
So whether it's hot or cold, it still runs crappy at 55+? Any video of this happening?
Can you post up some graphs of your timing, RPM, and injector pulsewidth while it's acting up?
I can't imagine a stock 454 tune would run anywhere near decently at any RPM on any 401.
Probably not related to your issue, but how tight is the fit between the actual PCV valve and the rubber grommet on the intake manifold? Sometimes they loosen up and induce a vacuum leak.
Temp sensor/switch working OK? Maybe the computer thinks the engine is cold at all times and makes it run very rich when it gets warm.
Couple questions about your setup:
1) Are you saying that you are running a throttle body EFI taken off a GM big block 454?
2) Has the fuel mapping been adjusted for use in an AMC small block 401 or are you running stock 454 mapping?
2) Are you sure the backfires are occurring out the exhaust and not out of the EFI air horn?
3) Do you have the ability to adjust the acceleration pump squirt?
4) Are you certain that locking the timing at zero degrees is correct for this application? Are you removing vacuum advance when you lock the base timing? I'm not sure that the timing required by a big block 454 will work for a small block 401. Perhaps you need to adjust the base timing so the computer will produce timing closer to what the 401 requires. I would be concerned about the timing curve as well. I am certain that a 454 has a much different timing curve than a 401. The 454 is a much different animal than the 401 in many ways.
2) Both. As a wise man once said (babywag) you can't tune away problems; I started to adjust the fuel and timing tables but stopped because I have problems that need fixing. So I have a set of tables that are rough draft for a 401 and I have stock 454. I used them both and recorded data on both.
4) Yep! I have the ability to set my timing wherever I want in the BIN. I have the BIN set up so the Dizzy is at 0 and the computer does the rest based on inputs from the MAP and engine RPM signal from my hall sensor. Yep (vacuum advance removed, also). Good point (454 vs 401 timing). In my first set of logs below I have a set of fuel tables and timing tables I rough drafted and drove on for the 401. The second set are basically stock 454 fuel and timing tables. I can change from tune to tune very quickly and easily. I do not have my knock sensor mounted in the block yet but that is coming. Once it is mounted, I will really dial in the timing. Very true! (454 different animal than 401). My plan is to get it "working", then I can dial in the fuel and timing curves to work for a 401. 454 is the cloesest engine to work with because I am .030 over so it's really 407. 454 is only 47 cubic inches from me vs. 350 being 57 cubic inches. In the MAP I am able to set cylinder volume so my BIN knows its working with 407 cubic inches and not 454.
I was on a plane for several hours so I dug into my data logs. Here is my driving route (this is basically cut and pasted from my build thread):
Red lines are 45 or 50 MPH speed limit, black is 65. Run 1 is about 2 miles, run 2 is about 3/4 of a mile, run 3 is about 2.25 miles.
I made one lap with my tables, then swapped to the stock 454 tables. Power enrichment is turned off. I rarely get a flash of Acceleration enrichment here and there but it does not coincide with a backfire.
Dark blue= Vehicle speed sensor
Light blue=Engine RPM
Red=Throttle position Sensor
Sky blue=Degrees of timing
Here is how it looks as I leave my house on run 1. It runs AMAZING here. You can see the temp is pretty cool and rising. It is open loop for most of this.
Then I sit at the stop light and head up onto the freeway for run 2. Where I have the marker, is right before the first backfire, of 2'ish.
Then I get off the freeway and head down run 3. This is usually a steady throttle, 50MPH run. Where you see me let go of the throttle, that is because it is back firing.
Then I swap over to the 454 BIN.
Run 1, it was stumbling a little but the idle was amazing.
Merging onto the freeway again, run 2 with the 454 bin. It backfired sooner and more often, like 4-6 times. Where the marker is, is about where it was the worst.
Then the last bit as I headed down run 3. It was backfiring pretty bad here.
So I do not see a common indicator of when it backfires. I do not see any of these sensors showing me anything that looks abnormal. The data looks consistent warm and cold.
Anyone else see anything odd in these?
Looking at my RPM signal, it looks normal to me. No spikes, no flat spots. See graphs above.
Copy (tuning). I am starting to think I may have a distributor phasing issue. I "think" it seems to backfire when timing is max. If my hall sensor is locked down in the wrong spot, I wonder if my spark is jumping to the wrong post. I have a cheapie cap in route so I can cut a hole in it and will verify my phasing (hopefully) this weekend with my timing light. Then I will start looking into tuning more.
Its hard for me to test this cold. There are too many stop lights and cops for me to get up to 65 without getting on the freeway. By the time I am getting on the freeway I am not up to operating temp but I am getting pretty darn close. Does not seem to be any different between below operating temp and operating temp.
I will take some video next drive (hopefully) this weekend. Good idea.
Graphs posted. I did not put injector pulse width but will look at it now. I looked at duty cycle and it never went above 40% that I saw.
Rang, can you post up screenshots of both your 401 fuel table and the 454 fuel table for us to look at?
How exactly is your ignition set up? Is the Hall Effect sensor output feeding directly into your ECU, or is the sensor output going to some sort of ignition module first, which is then feeding your ECU a tach signal? If it's the former, does your tuning software allow you to see an actual tooth count rather than just RPM ticks? You should be getting one tooth count each time a tooth passes by the sensor. That's what I would double-check while it's backfiring, though it seems ok.
Also, is your Hall Effect setting in the ECU/tuning software set to "normally high/on" or "normally low/off"? I think GM sensors are "Normally On," but you're running a Motorcraft distributor right? Ford sensors are "Normally Off". I don't even know if the engine would run if this setting was wrong, but it's something that can be easily checked.
I'm definitely no expert, and I'm not familiar with the details of your system because I run MegasSquirt, but everything looks ok to me in your graphs, including the tach signal. Just by eyeballing it, it SEEMS that the MAP drops first (from the backfire), before the throttle is closed, which then cascades to some of the other variables. Is that correct? Let's see what the injector pulsewidths are, and what the timestamps are for everything.
Are all of these logs in open loop (O2 correction disabled)?
Great thinking. Triple check all the basics, always. We just want to rule out the EFI system as a cause, so don't get so focused on it being an EFI problem that it stops you from doing the other simple tests that were suggested in this thread. If it were my rig, I'd be tempted to go back to fuel-only EFI with mechanical/module ignition and see if that stops it, like Levelhead said.
And here is stock 454. When you look at the graph, notice the scale is different. So on initial look, it looks like I increased the values but really, they are less.
Ignition set up: Factory pickup is screwed down to the plate, wires come out the side of the dizzy, through a weatherpack connector, and are hooked to the 7 pin ignition module pins P and N. Ignition module pins + and C go to my coil and pins E, R, B, and G all go to my computer.
I don't see tooth counts, just engine RPM.
I searched all my parameters for "Hall", "Pickup", and "normally" and cannot find anywhere to adjust it or verify it.
Only run 1 of the 401 bin is open loop. My engine went closed loop toward the end of run 1.
here is Run 2 with 401. New color codes because I removed some items and added pulse width:
Light/Sky blue=throttle position
First backfire is around 3:31
Here is the 454 bin with pulse width added.
Will do. I have found another duraspark distributor so if I cannot get thsi to work, I may go to fuel only EFI. These buisiness trips are killing me! I just want to go home and fix this thing!!!
Couple of quick points...
1) The 454 tune is, of course, dumping in WAY more fuel in a lot of areas, so it makes sense that the backfire is worse with that tune.
2) Looks like the commanded timing is fine and steady? Of course we have no idea what is actually happening at the output of the module.
3) What would you say your redline is? 4,800rpm? 5,200rpm max? I would get rid of every row after 5,000rpm and then smooth out the RPM axis for better resolution.
4) This is where a wideband O2 is essential.
Try changing the following on the 401 map just for fun:
40x1600 from 53.9 to 58
40x2000 from 56.6 to 60
40x2400 from 60.9 to 63
40x2800 from 65.6 to 67
50x2000 from 59.4 to 65 ****Important
50x2400 from 61.1 to 68
50x2800 from 67.6 to 70
60x1600 from 61.7 to 65
60x2000 from 68.0 to 66
Then go hang out in that same area of the map with as steady a throttle as possible. If that makes things worse, then restore the map and then try going the other way:
50x1600 from 63.3 to 55 ****Important
40x1200 from 60.2 to 50
40x1600 from 53.9 to 51
40x2000 from 56.6 to 53
Be careful when going lean, you are flying blind with no wideband O2 readout so you want to do so in small steps. And don't change any bins you know for a fact work better than the numbers I'm giving you.
You can always restore to your current map if it doesn't work out.
I'm not saying the fuel map is the issue, even though it's very rough indeed, but I did have some misfires/cuts and near backfires that drove me crazy - all due to an improperly aligned VE/fuel map. There was just too much distance between bins, in certain areas of the map, for the ECU to interpolate correctly.
Where did you get your base 401 map? Is there anyone else around running the same system/cam/compression who can give you their map that might be more dialed in? Does your tuning software have a base VE table generator?
How high have the RPMs got on this EFI system? I'd be very leery of pushing them too far with a questionable map and no AFR data to watch as you're leaning on that throttle in real time.
What about EGR? How do you guys deal with it on this EFI system? Is it opening up or somehow leaking on cruise and messing up the mixture? You said it's disconnected for the time being?
2)The only way I can think to verify actual timing is I stick my laptop on my inner fender, while it's running, with my commanded timing showing on TunerPro, and my timing gun pointed at my timing marks. I have verified it at idle with my timing light and it seems to be spot on.
3) I built the motor for 5500RPM. I had it balanced, ported the heads, and the cam is supposed to be good to 6k. I have not gone north of 4k yet though, some because I cannot tune there without a wide band but, mostly because I backfire before I get there. I will try your recommendation, though.
4) concur! (wideband). Will do (update 401 map to those values). Copy (careful when going lean)
Base 401 map started with the 454 map. I smoothed it, drove on it, data logged on it, plugged my numbers into an excel spreadsheet I got from gearhead-efi, re-did the map with the excel numbers, and repeated. I don't know anyone running the same system/cam/compression. I know several with 2 of those 3 (babywag has same system, Full Size Jeeper has similiar cam and compression). Honestly, I feel like if I can get this thing to stop backfiring, I am pretty confident I can get the maps close enough to run really good. Nope, no base VE table generator in TunerPro. I have many options on the the interwebz to download though.
Highest I have gone is about 4k becuase backfiring. Concur (leery of pushing too far).
I have an EGR mounted, I have an EGR solenoid mounted and wired, but it is disabled in the bin. I forgot to disconnect the vacuum line (a guy on theamcforum noticed that in a picture) during troubleshooting so there is a possibility that it is leaking but I am not too worried about that.
I really appreciate you guys all helping me out! This is all new to me and I would be lost if it weren't for folks like all you guys on here! I have been fighting with this and other things for so long, it's easy to get distracted and miss things.
What is your FP set @
Did you change the injector flow rate in .bin to match your injectors @ that FP?
If the value isn't correct the pcm calculations for fuel will be wrong.
It appears to be rich, but without seeing the logs or the fuel trims just a guess.
You might want to look @ INT & BLM in monitor to see what's occurring ~backfiring?
Before significant changes to fuel map as I suggested a couple times already force it OL and see how it runs.
|All times are GMT -6. The time now is 02:32 AM.|
Powered by vBulletin Version 3.5.4
Copyright ©2000 - 2019, Jelsoft Enterprises Ltd.