Backfiring, intake manifold during cranking

[beev]

Hello,

I am currently trying to install a microsquirt v3. I have made it to section 2.20 of the Megasquirt-2 Setup QuickStart Guide - expanded written by Jame Murray in 2019, (http://www.msextra.com/doc/pdf/Megasqui ... Up-3.4.pdf).

I have verified the timing, and confirmed that TDC is where I expected it. Attached is the MSQ file during cranking for section 2.19, RPMs are steady at 200, the MAP does drop but isn't super steady (I may need to use Alpha-N) but this is a known issue, the remaining sensors are clear. RPM has also been verified with a digital strobe lights set to 2 stroke.

The engine itself is a Datsun L20b 4 cylinder that has been modified to have higher compression. I have converted it to a 36-1 wheel with wasted gate. Injectors are low impedance, there are 2 of them on a modified dual SU sidedraft intake manifold. Firing order is 1,3,4,2, I have 1 and 4 on ignition 1, 3 and 2 are on ignition 2. The coils are D585.

For section 2.20 I started cranking and it takes a while but it starts to backfire out of the intake, I have seen fire a few times. Sometime it jams the starter and makes a terrible terrible grinding sound, I assume it is my poor ring gear taking a beating. Below are links to 2 videos of me apparently trying to destroy my engine.

https://photos.app.goo.gl/NpTh2rp6tTE28sXcA
https://photos.app.goo.gl/qFwA3wers79QjQiaA


While searching the forums I have tried adjusting the cranking advance from 10 down to 5 and zero. I have tried giving a little throttle while cranking. I have delayed the priming pulse up to 3 seconds, and I have totally disabled the priming pulse. All to no avail. I couldn't see any other things to try on the forum so here I am. The only remaining items I can think of is it is running too lean (loosely suggested on the forum), or I am missing something about the timing, perhaps the speed is changing inconsistently.

I am sure there are settings, details, etc that I am missing and can provide.

Thanks in advance for your help. I am a noob at this so please explain is as detailed and simple as possible.

[beev]

Sorry some technical difficulties. Attached is the MSQ.

EDIT: Deleted old MSQ file that was wrong. See lower post. My apologies for wasting time.

[SQLGUY]

Sounds to me like timing might be off... maybe way off.

Are you sure your tooth #1 angle is set correctly? Can you get enough of a signal while cranking to verify about where it's firing with a timing light on the 36-1 wheel?

[beev]

That is part of the earlier steps. With the injector fuses out I can confirm the timing is correctly aligned. I even pulled plug #1 to make sure the timing mark is actually TDC.

[SQLGUY]

I didn't load the msq... just looked through it. It looks like crank dwell is 6ms, but the nominal dwell is much lower... (0.6ms). Is that correct?

[beev]

Ah that is weird. I grabbed a really old file (early development from last year when I was just trying to understand), please see the newly attached one. I am going to delete the old post. I have it at 3.5ms same as the nominal. I am not sure if it is supposed to be different from nominal or not. 6 would for sure be a problem as it would discharge early (and likely give the problems I am seeing). I have re-tested just to make sure I haven't messed this up. The problem still exists.

[rickb794]

What kind of coil? Ah, LS coils? Is the main coil ground connection short and to the motor?
With the connector unplugged from the MS there should be no continuity to ground on pin 20. see 3.2 of the manual.

I'm guessing spark is out of phase.

I am concerned as you have made multiple compromises in your selection of hardware.
I am not sure you will get good fuel distribution with only 2 injectors. It may need more squirts?
Low impedance injectors with a microsquirt needs resistors. No resistors will overload the MS.
2 x 326cc injectors are only good for 95hp, a high compression 2.0L ought to be good for 110 - 130 hp maybe more.
Is there a balance tube or common plenum between the TB's?

Is compression even between the cylinders?
Is it getting sync and an RPM indication in TS?

I'd suggest adding some supplemental fuel to see if it will fire, but that could have exciting results if the firing order is off or a valve is not closing.
Do not flambe' the L20...

[beev]

Hi Rick,

Could you explain out of phase? I have checked with a timing light the spark on all cylinders. The timing is within +/-2.5 degrees. It is harder to tell for cylinders 2-3 as there is no notch in the pulley, but I verified when cylinder 2 is at at the top of its stroke and marked the pully.

The coils are D585 LS2 coils.

What do you mean by compromise on hardware?

The system was previously running with the same injectors on a GM ECU from a 1995 chevy pickup (one of the old eprom type), I know the injectors work for this engine. I upgraded to the microsquirt for improved control and to go the the 36-1 wheel. I know that they work with this setup.

I have wired 3.3 ohm resistors inline with each injector.

Yes, there is a balance tube.

Compression on cylinder 2 is a little low, 170 vs 185 for the remaining cylinders. Still within the 10% suggested value but I agree close. I will be checking the lash.

What method would you suggest for adding supplemental fuel?

[beev]

I realized the specific question might have been what exact coils do I have.

I purchased these DELPHI GN10119

https://www.rockauto.com/en/moreinfo.ph ... =16&jsn=16

[rickb794]

I asked what kind of coils and then re-read and see you have LS coils.
Is the main ground for the coils short and connected to the block?

My supplemental fuel is a can of Berryman's spray carb cleaner.
Adding supplemental fuel may cause it to spray flaming fuel if an intake valve is not closing or ignition is not right.

Out of phase is TDC on the other half of the 720° cycle of a ICE.

Throttle body injection is not a good choice, there are reasons it is no longer used.
Port injection with a long runner and a large plenum is more efficient and will make more torques.

A single injector running 2 cylinders may have fuel robbing issues.
It seems like you would want to double the number of squirts to lessen possible fuel robbing.

Your injector size is small according to a recognized sizing chart. Found here http://users.erols.com/srweiss/tableifc.htm#ITABLE
30#/326cc = 48HP times 2 = 96HP a high compression 2.0L should be good for 110-130HP maybe more. I'd probably go with a 42#/442cc injector.
I ran (4) 24#/252cc injectors = 148hp capacity on my 2.0L 4 valve motor.
Low impedance injectors are old tech, modern high impedance injectors may have favorable spray patterns and are less electrically noisy.
The mini's with shared ports had troubles and it was found that they responded to a larger injector because they could get better fueling by messing with injector timing and using a shorter (but larger) squirt.
You might be able to overdrive your small injectors or raise the fuel pressure to cheat it, but for a high compression motor you don't want to come up short on fueling at WOT, and a larger injector may improve throttle response.

[vr-vagman]

I realized the specific question might have been what exact coils do I have.

I purchased these DELPHI GN10119

https://www.rockauto.com/en/moreinfo.ph ... =16&jsn=16

May be not related to the current problem but be aware these coils should not be charged over 5ms. They don't need this either.
Also, I'd remove the WBO2 sensor from the exhaust during the initial starting attempts. Also I'd turn the EGO control OFF/Disabled . Also could you verify your cranking spark timing also on 2nd/3rd cylinder?

[SQLGUY]

I asked what kind of coils and then re-read and see you have LS coils.
Is the main ground for the coils short and connected to the block?

My supplemental fuel is a can of Berryman's spray carb cleaner.
Adding supplemental fuel may cause it to spray flaming fuel if an intake valve is not closing or ignition is not right.

Out of phase is TDC on the other half of the 720° cycle of a ICE.

Sounds like you're talking about TDC at exhaust vs. TDC at compression. He's running a wasted spark system.

Throttle body injection is not a good choice, there are reasons it is no longer used.
Port injection with a long runner and a large plenum is more efficient and will make more torques.

A single injector running 2 cylinders may have fuel robbing issues.
It seems like you would want to double the number of squirts to lessen possible fuel robbing.

Your injector size is small according to a recognized sizing chart. Found here http://users.erols.com/srweiss/tableifc.htm#ITABLE
30#/326cc = 48HP times 2 = 96HP a high compression 2.0L should be good for 110-130HP maybe more. I'd probably go with a 42#/442cc injector.
I ran (4) 24#/252cc injectors = 148hp capacity on my 2.0L 4 valve motor.

That chart sounds pessimistic... sort of like the carburetor charts with ridiculous CFM requirements. For instance, Porsche ran 19lb injectors on the 5.4L 32-valve 928 engine, and that's for 345HP stock. By that chart, Porsche should have only been able to get 243HP out of that setup. In fact, these stock injectors are even good with changes like more aggressive cams and X-pipe exhaust setups.

Low impedance injectors are old tech, modern high impedance injectors may have favorable spray patterns and are less electrically noisy.
The mini's with shared ports had troubles and it was found that they responded to a larger injector because they could get better fueling by messing with injector timing and using a shorter (but larger) squirt.
You might be able to overdrive your small injectors or raise the fuel pressure to cheat it, but for a high compression motor you don't want to come up short on fueling at WOT, and a larger injector may improve throttle response.

[beev]

Hi thanks for the input on the overall system . Although I may lose some performance on the old injectors and throttle bodies, I have no performance right now and would like to focus on having it actually start. I think it is unlikely to be the injectors or the settings with the injectors as the car has run with those injectors before and prior to installing the 36-1 wheel it did start on the microsquirt on a fuel only setting. I will however disable EGO control and pull the sensor (I am guessing to protect it, hopefully I haven't caused too much damage already)

In regards to 720 degrees out of phase, I am running wasted as pointed out so it can't be that.

Yes, I have verified the timing during cranking of coils 1-4. I have however not done that for all coils with the injectors also firing.

For the charge time, it is set to 3.5ms nominal as per the microsquirt manual. Should it be the same for cranking?

Yes the injectors are grounded to the block. The longest path is about 2 feet, shortest is about 6 inches.

So under the assumption I have TDC correct and verified with a timing light without the injectors, what else could it be?

What about my settings for during cranking is the battery correction too high, maybe it occasionally causes the coils to fire early to protect themselves?

What about the change in speed for the trigger wheel should I not use predictive?

What other settings can effect when the coils fire?

[vr-vagman]

Hi beev,
from the attached videos I could see you kinda have occasional ignitions. IF your timing is close enough to the wanted advance BTDC, you should fire the engine even by sparying carburetor cleaner inside the manifold. Choose some "Fixed" advance (i.e. 10 deg) in the settings to avoid any unwanted timing corrections at that moment. If you manage to fire it and get some relatively stable idle on all 4 cylinders while you're still spraying, then probably the problem is in the fuel supply.
How may gaps do you have between tooth N1 and TDC cyl1 tooth?

What do you mean with that:

"Yes the injectors are grounded to the block. The longest path is about 2 feet, shortest is about 6 inches".
The injectors should not be grounded.

[beev]

Sorry jumping between injectors and coils before coffee is a bad idea.

COILS are grounded to the block, 6 inches and 24 inches. They also have the ground back to sensor but an earlier question was specifically about power ground.

Injectors are not to ground.

There are 8 gaps from the N1 to the TDC cly 1 tooth.

For spraying the carb cleaner do you do that continuously while cranking or fire some in before. Also throttle bodies closed while cranking?

[rickb794]

That chart sounds pessimistic... sort of like the carburetor charts with ridiculous CFM requirements. For instance, Porsche ran 19lb injectors on the 5.4L 32-valve 928 engine, and that's for 345HP stock. By that chart, Porsche should have only been able to get 243HP out of that setup. In fact, these stock injectors are even good with changes like more aggressive cams and X-pipe exhaust setups.

That chart was created using math science and sound engineering principals
Like I said you can overdrive the injectors. (at your own risk)
There is no normal reason to drive an injector over 80%DC.
For street use and longevity it's probably best to limit the coil heating caused by increased Duty Cycle.
Injectors were not made to run wide open. They were made to pulse, not imitate a fire hose.

[beev]

Any chance the injector conversation can be taken offline from this thread. I except the risk of the injectors, they should not prevent me from starting the car.

[rickb794]

Sorry jumping between injectors and coils before coffee is a bad idea.

COILS are grounded to the block, 6 inches and 24 inches. They also have the ground back to sensor but an earlier question was specifically about power ground.

Injectors are not to ground.

There are 8 gaps from the N1 to the TDC cly 1 tooth.

For spraying the carb cleaner do you do that continuously while cranking or fire some in before. Also throttle bodies closed while cranking?

No NOT CONTINUOUS SPRAY!!! Open the throttle and give each TB a 1 second blast. Close throttle and crank.
The motor should start and then die in a second if it not getting any other fuel.
If it is getting fuel it may continue to run.

If it doesn't start and continues to belch fire you have a valve that is not closing or incorrect ignition events.

Increase VE or ReqFuel to get this better so you can get the motor warmed up.
Don't raise ReqFuel If you have small numbers in the lower left corner (less than ~20?)
Don't bother changing warm up fuel or cranking fuel as these trims are modifiers to the hot tune numbers.
You need good hot tune numbers to base those modifiers on.

I have a "getting started" tuning write up if you are interested.

[Matt Cramer]

Backfires out the intake are usually timing related. And I'm seeing a massive sync loss issue in your data logs, which could either account for the timing problem or be caused by a cylinder firing out of time. Can you get a tooth log with the injectors disconnected, and one tooth log with the injectors working?

Also, I see your tune is set to wasted-COP, four spark outputs. How exactly are these wired up? Can you make sure the following outputs work in output test mode:

Spark A - cylinder 1
Spark B- cylinder 3
Spark C - cylinder 4
Spark D - cylinder 2

[vr-vagman]

Sorry jumping between injectors and coils before coffee is a bad idea.

COILS are grounded to the block, 6 inches and 24 inches. They also have the ground back to sensor but an earlier question was specifically about power ground.

Injectors are not to ground.

There are 8 gaps from the N1 to the TDC cly 1 tooth.

For spraying the carb cleaner do you do that continuously while cranking or fire some in before. Also throttle bodies closed while cranking?

I see 90 deg tooth N1 angle in the attached MSQ. You should change it to 80 i.e. 8gaps X 10 degrees