Engine dies with any load

TPI 85 Blazer · Post by **TPI 85 Blazer** » Wed Jul 26, 2006 5:43 pm

I fear that this will be a long post, but I want to give all relevant info. This is my first Megasquirt project.

The engine idles fine, but whenever I put it in gear, it tries to die. I have driven it a block, but I can't move the throttle more than a tiny bit at a time or it will die. When I am easing up past 1800rpm, it starts popping like it was lean or my spark was way, way off.

I started with a crate Chevy 350 running a Quadrajet, probably 85,000 miles on the engine and a recent rebuild on the carb. It pulled about 22-23 psi on my vacuum gauge at idle and ran great, other than occasional cold start problems in the winter, but hey, it's a carb.

I pulled the intake and carb and replaced them with 1985 or 86 TPI manifold, runners, plenum, Throttle body, etc. I created my own emissions system, using the EGR, canister purge and TH-700R4 lockup. I built my own harness from the MS-II 3.0 to all of the sensors and a Ford electrical center that I rebuilt to hold all my fuses and relays for the electric fans, A/C, fuel injection, stereo, etc. I work for Delphi and previously interfaced with Packard on connector and harness design, so the harness and connectors are OEM quality. I have an Innovate LC-1 with the sensor mounted in the passenger side header collector. I locked down my vacuum/mechanical advance HEI so that it no longer has any advance capability. The coil in it is a new MSD. The VR coil in it feeds the VR circuit in the MS and the VB921 drives the coil. The VR signal is through shielded cable and the coil drive is through coaxial cable. For fuel, I put a TPI Camaro pump in the sending unit of the gas tank from a 1990 TBI Blazer. Fuel pressure is about 44psi without vacuum and around 34psi at idle.

The engine starts and idles decent, but I can't get anymore than 20-21psi of vacuum. This is by tweaking the VE and spark advance tables for maximum vacuum. With no load, I can slowly move the throttle and get the engine over 2000 rpm (I haven't tried for more). My timing light is within 1-2 degrees of what MT reports over this range (how I got this much is another story). The engine would die as soon as I put it into gear, even if I set the idle to like 1200rpm. I got it to sort of run in gear by increasing the dwell time. I scoped voltage on the .01 Ohm shunt resistor until the coil current was over 7A. The VB921 is just getting above ambient temp in this condition. If I move the throttle with any real speed, the engine dies.

Here's what I've worked on:

-AE: I tried turning it way up and turning it off, different combinations of MAP and TPS, but honestly, I couldn't tell any difference.

-Vacuum leaks. Yesterday, I was SURE I had a serious vacuum leak, but I couldn't find any with a stethoscope or with propane. When the engine is idling I can put my hand over the intake and it stalls. Then it holds a vacuum for a second or two until I pull my hand away. I capped off all non-critical lines without any change.

-Spark setup. I actually have the VR leads reversed (green to pin 24 and white to ground), but it seems to me this doesn't matter, as you can select "rising edge" or "falling edge" in the menu. I even ran the scope in the vehicle while it was stalling under load. The timing doesn't appear to change (relative to the VR events), nor does the coil current go down.

-Lean condition. I have tried increasing the VE numbers for low rpm and increasing Req_fuel, but I couldn't find any difference.

-Exhaust leak. I tightened the header and exhaust bolts up to make sure no fresh air was getting sucked in, but the problem is there whether EGO correction is active or not.

I just keep thinking that there is a fundamental, simple problem here. I have double and triple-checked the routing of all spark plug wires. I should also mention that the injectors are used and although I did verify ahead of time that they all squirt, they have not been flow tested.

So what should I try? Thanks for the help.

JNygard · Post by **JNygard** » Wed Jul 26, 2006 7:37 pm

Well sounds like a setup similar to mine. Not sure what injectors your running. I still have the 19lb injectors on mine. I say the problem is in your ve table, need to bump up the values above idle so it can catch and recover. I still have the problem with a cold engine and putting it in gear and stalling. But once its warm it goes into gear just fine. Your welcome to look at my msq file its at the bottom of my website. It was created with megatune 225 http://71.37.161.253/ProjectBlazer.htm I have driven a quite a few miles on this tune. Just my cold tune needs work.

Jeff

jakobsladderz · Post by **jakobsladderz** » Wed Jul 26, 2006 11:05 pm

Is there power getting to your fuel pump, these symptoms could relate to that? if you have a fuel pressure gauge and it's showing the numbers your seeing then I guess all's well though.

TPI 85 Blazer · Post by **TPI 85 Blazer** » Thu Jul 27, 2006 3:32 am

jakobz, I can crack the throttle under the hood and the fuel pressure goes up slightly while it dies. I was trying to think of a way to run the hose between the hood and the cowl and tape the gauge to the windshield.

JNygard, Your VE table is huge compared to mine, but my required fuel isn't much different. I am actually not sure of my injector flow rating. The only thing I ever found on-line says they are 24lb/hr, but I can't be sure of this. It's always possible that they are 19lb/hr, since there are so many of those floating around. I'm curious why your VE goes so high above 100. Doesn't this just mean that you need to re-scale it and adjust req_fuel? I'm also wondering if it works OK at high rpm / high load or if you had to change out the injectors for bigger ones.

I will try upping all of the VE values above idle MAP values and see what happens.

JNygard · Post by **JNygard** » Thu Jul 27, 2006 7:43 am

When I originally built mine I still had the 305 in it and the 19lb injectors were enough. I just scaled everything up when I put the 350 cu/in engine in. It runs and drives down the freeway with no problems. I have not run it at high rpms very much. It's on my to do list to find some 24 or 30 lb injectors and swap out.

TPI 85 Blazer · Post by **TPI 85 Blazer** » Thu Jul 27, 2006 8:21 am

The more I think about this, the more I am thinking that the VE table is just super lean everywhere but at idle MAP readings. I will try turning everything else up. Maybe the WBO2 is lying to me. It jumps around a lot when things are acting up. Maybe it misfires and the raw fuel screws it up. I would think that misfires would read leaner though since there woudl be so much oxygen. I will listen more to the engine and less to the O2 sensor for a while.

JNygard · Post by **JNygard** » Thu Jul 27, 2006 10:09 am

Yeap once you get the map closer then the WB02 should really work for you. It is a bit easier working from richer to lean. So fatten the rest of the table up a bit.

ChevelleFan · Post by **ChevelleFan** » Thu Jul 27, 2006 2:22 pm

Without seeing a datalog, I would have to guess the VE table is wrong too. If you can crack the throttle *slowly* to get it up to 2000 RPM, then your values at 20-30kpa and 1000-2000 RPM are close. If you crack the throttle quickly and it dies, then the values at, say 70-90kpa and 1000-2000 RPM are bad -- bad enough to not give the engine the right fuel and it dies. When you drop it in gear and it dies, look around the 500-1000 RPM values and 40-50 kpa. My kpa numbers might be wrong for your engine, but the idea is that you'll have less vacuum (higher kpa) in gear than in park. Check these area of your table.

Since the engine seems to idle well (or, at least consistently), that tells me the fuel pump & injectors are working. It doesn't matter what the real flow number is for the injectors, yet. You compensate this by adjusting REQ_FUEL.

When you're making changes to the VE table or REQ_FUEL, how much are you changing the values? I think you need to make large (20-25%) changes in both directions to see what happens. If that proves to make no noticable change, try bigger (35-50%) changes.

Post a datalog of it running with 1) idle 2) slow throttle movement to get to 2000 RPM 3) quick throttle movement (or dropping it in gear) to get it to die.

-Dave

TPI 85 Blazer · Post by **TPI 85 Blazer** » Thu Jul 27, 2006 5:53 pm

OK, you guys were absolutely right. I wasn't getting enough fuel. I generated a new VE table that is fatter all over and now I can idle in gear. It's not great yet, but I'm able to drive it a little.

BUT, I found a bigger problem. The popping and surging at higher rpms (in this case I am just trying to break 2k) is the spark dropping out. The coil current is consistent as the engine speed changes, as measured with a scope. However, the timing light cuts out altogether. This is why it feels like a rev limiter. My guess is that the mechanical assembly of my dizzy is causing the rotor to not be aligned with the brass stud when the coil is firing. I will go test this out by adjusting the advance in MT. Less advance should let me reach a higher rpm. I suppose I can turn the dizzy CCW and then adjust the timing back with the offset in MT. Otherwise I will have to pull the dizzy and move it back one tooth or something like that.

TPI 85 Blazer · Post by **TPI 85 Blazer** » Thu Jul 27, 2006 8:31 pm

Sweet. I drove it and got to 35mph.

I turned the dizzy CCW and then used the trigger wizard in MT to offset the timing by 18 degrees so the timing light matched again. I was able to get Autotune to work and it started making some adjustments to the VE table. I am actually planning to drive it to work in the morning, which is only 2.5 miles away, with 30mph speed limits.