low EGT values on specific cylinders

[lakis1982]

Hello,

I am trying to finally program my ECU parameters (i am using an Ecumaster EMU device) on my 2100cc oxyboxer turbo injection engine..
I have one EGT sensor installed on every cylinder so that i can monitor their values.. No wideband sensor yet...

My problem is that i see very low temperatures around 70-100C on cylinders 1 and 3, while on cylinders 2 and 4 the values are more normal around 500C.. These are values after letting it idle for a couple of minutes..

I did a test to make sure that my ignition per cylinder is working consistently and is syncronized and i see that all the ingition angles are correct..
So i assumed that the problem  is injection related. I also swapped the injectors between cyl 1 and 2 and the same problem occurs, low EGT values on cyl 1 and 3 and normal values on cyl 2 and 4..

Then I did took off the spark plug from cylinder 2. The weird thing is that in this case, the EGT values on cyl 1 and 3 became normal !!!! around 480C and on cylinder 2 around 80C (as expected) and on cylinder 4 around 180C...
DO you know why does this could happen ?? WHy do I have low EGT values on cyl 1 and 3 , and then when i take off spark plug from cyl 2, why the EGT values on cyl 1 and 3 go normal  

I am not using any EGT based correction yet on the pulse width of every injector
I am using SIemends Deka genuine brand new injectors
The engine is fresh built brand new
i am using batch fire injection

Please give me your thoughts , I feel i am stucked...

REgards

[Basti]

Hi!

Are the EGT sensors in the same Position on all cylinders? I think this is not possible or?
The distance to exhaust valve is very important, also the deepth of the Sensor in the exhaust pipe changes the values a lot.

Did yiu check this?

Cheers,
Basti

[lakis1982]

yes..all EGT sensors are installed in the same way with the same depth and distance from exhaust port of the cylinder( around 10cm)...

I dont think that the values of EGT are misleading.. I can see it in the way that the engine idles that there is not consistent performance on all cylinders... I just try to figure out the problem based on the EGT readings...

[Jesse Wens]

Wasted spark?

[lakis1982]

yes it is wasted spark...

[Martin S.]

Does your engine have individual throttle bodies or a single throttle body? (Pics pls)
How do the plugs look?
Im curious. Why dont you use a wide band?
And why are you not using sequential injection?

[richie]

So the problem moves with the spark plugs? then most likely the 2 low cylinder have fouled plugs, try a new set of plugs and see if problem goes?

cheers Richie

[lakis1982]

I'm not using wideband yet because it is the first steps of setting up the ECU and I'm afraid i will poison the sensor..I will hook one up when I make things a bit better...
I chose wasted spark for simplicity reasons..
[ Attachment: You are not allowed to view attachments ]
[ Attachment: You are not allowed to view attachments ]

I checked that all spark plugs output a strong spark..I will give it a try to change them..

It's frustrating..

[wph]

Cylinder robbing... single throttebody feeding two cylinder. Even worse with carburetors. My first guess, swap EGT sensors from cold to hot cylinders and see what it does

[Erlend / bug66]

Looks like you are feeding a lot more air to 1 and 3...

[Martin S.]

Cylinder robbing... single throttebody feeding two cylinder. Even worse with carburetors. My first guess, swap EGT sensors from cold to hot cylinders and see what it does

Haha, you might be thinking of Kads? The only reason I had the only Kad engine in town idling in all four was that we adapted a larger diameter balance tube compared to the original.
My first thought on this problem is how does the engine drive?
Im usually happy if my engines idle at all, yet alone idle on all four.
(Like, who cares about idle, just crank up the idle speed and be happy)
 

[dannyboy]

i had a bent butterfly that caused big differences in temps on tick over ?

[lakis1982]

Looks like you are feeding a lot more air to 1 and 3...

If  cyls 1 and 3 get a lot more air than required, should't the EGT values for those be elevated due to a lean situation ??



The engine does not drive at all .. the car sits on a lift and we try to make the ECU parameters acceptable so that the engine can run in an acceptable manner... Now it feels like it is working on 2 or 3 cylinders only..
Do you think there could be more air going to one cylinder that the other that sits beside   I am using this throttle system   https://vwspeedshop.com/product.php?productid=16759&cat=395&page=1

Yesterday we also hooked up the wideband sensor just to confirm what we expected.. At around 1400rpm the lambda is around 0.73 and AFR around 10.8 .. SO it appears to be excessively rich...

The ignition is wasted spark as mentioned...so it has the following firing sequence per engine cycle:

igntion event 1: fires on cyls 1 and 3
igntion event 2: fires on cyls 2 and 4
igntion event 3: fires on cyls 1 and 3
igntion event 4: fires on cyls 2 and 4

The injection is batch fire injection and uses the following sequence:

injector 1 injects when ignition event 1 occurs
injector 1 injects when ignition event 1 occurs
injector 1 injects when ignition event 1 occurs
injector 1 injects when ignition event 1 occurs

you can check it below the settings i am using
[ Attachment: You are not allowed to view attachments ]

I am using ECUMASTER EMU as an ecu for the engine ..

[Erlend / bug66]

The divider in the intake is not in the middle of the air path. You could help the engine with moving the centre tube to above the fan shroud. The distribution would be much better.

[Basti]

I am not an expert in your ecu but for me it Looks like the Injection and ignition are at the same time or? No injection angle??

Looks like you are feeding a lot more air to 1 and 3...

If  cyls 1 and 3 get a lot more air than required, should't the EGT values for those be elevated due to a lean situation ??



The engine does not drive at all .. the car sits on a lift and we try to make the ECU parameters acceptable so that the engine can run in an acceptable manner... Now it feels like it is working on 2 or 3 cylinders only..
Do you think there could be more air going to one cylinder that the other that sits beside   I am using this throttle system   https://vwspeedshop.com/product.php?productid=16759&cat=395&page=1

Yesterday we also hooked up the wideband sensor just to confirm what we expected.. At around 1400rpm the lambda is around 0.73 and AFR around 10.8 .. SO it appears to be excessively rich...

The ignition is wasted spark as mentioned...so it has the following firing sequence per engine cycle:

igntion event 1: fires on cyls 1 and 3
igntion event 2: fires on cyls 2 and 4
igntion event 3: fires on cyls 1 and 3
igntion event 4: fires on cyls 2 and 4

The injection is batch fire injection and uses the following sequence:

injector 1 injects when ignition event 1 occurs
injector 1 injects when ignition event 1 occurs
injector 1 injects when ignition event 1 occurs
injector 1 injects when ignition event 1 occurs

you can check it below the settings i am using
[ Attachment: You are not allowed to view attachments ]

I am using ECUMASTER EMU as an ecu for the engine ..

[lakis1982]

here is the explanation for the injection phase
[ Attachment: You are not allowed to view attachments ]

im non sure if i should use an injection offset angle..


Erlend,
if the problem is the location of the center tube that appears to sit in a non symmetric location, shouldn't the differences for the EGT values appear between the left and right sided cylinders In my case, the problem is that the differences on EGT values appear between the front and the rear cylinders.

[Erlend / bug66]

If you follow the air path on the outside of the bends, you see that 1 / 3 will get most of the air. If you move the centre tube to above the shroud, the air will still travel at the outside wall of each bend, but will meet the divider in the intakes straight on.

[spoolin70]

Why not try removing the silicone hoses and just let the throttle bodies take air from directly above (at least to set the idle).

That will prove if the bends are causing the imbalance/unequal air supply

[lakis1982]

Spoolin79,

thats exactly what i was about to ask right now.. I will take off the silicone hoses and let the throttle bodies take air directly from the air.. Besides i am now testing around the 1300rpm region only.. If  the EGT values get much closer for all cylinders then we could say that the assumption of the air distribution through the tubes issue is correct.

If indeed is correct, could you tell me what i could do about it ?? I mean, if i cant move the center tube because of  limited spacing above, is there something else i could do to improve the situation Or you can just live with that and acccept that the idle will rely mostly on 2 cylinders..As far as i understood, this situation with uneven EGT values and air distribution should not exist on higher rpm regions e,g at 4500rpm ,

I will let you know tonight the results after i take off the silicon hoses...
All comments are appreciated..

thanks

[Martin S.]

If  cyls 1 and 3 get a lot more air than required, should't the EGT values for those be elevated due to a lean situation ??

Lean doesnt necessarily mean higher temps. As mixtures go leaner they cool down until the flame goes out.

10.8 is super rich! Thats like what you might see with boost enrichment, not at idle.
The air intake setup should not make a difference at idle. There is no boost and the turbo is just freewheeling.
Dont worry about the spark as you have lots of sparks going on, its the 10.8 mixture thats killing the combustion.
Can you reset the ecu back to default and start over with some kind of known baseline that works?

[lakis1982]

Martin,

thats the main problem.. That i dont have a baseline to start from.. Apart from a megasquirt file that someone gave me, that actually has the same engine specs with me...2100cc, turbocharged, same heads, same cam, etc.. SO i used his file as a reference to convert it to an Ecumaster ecu file...

Now as far as i get super rich values, should i start playing with the VE tables values so that the actual lambda starts getting close to the Lambda target table

Please eee below  the target lambda table that i am using and give me your comments .. I guess the lambda target table is one of the most important parameters of all

[ Attachment: You are not allowed to view attachments ]



Regards,

[cassa]

Cylinder robbing... single throttebody feeding two cylinder. Even worse with carburetors. My first guess, swap EGT sensors from cold to hot cylinders and see what it does

X2

Try leaning out cyl 1 and 3

[lakis1982]

So here we are..

As a test, I removed the silicon hoses from the throttle bodies so that they could breath physically ...the same problem.. the cyls 1 and 3 still see egt values below 100c around 1500rpm, while cylinders 2 and 4 still have egt values around 500c..I also cross checked this condition of extremely uneven temperatures using a laser gun pointing to the the exhaust ports of the cylinder.. cyls 1 and 3 have temps of 50c and 60c while cyls 2 and 4 have temps of 150c and 180c..this happened also after I have managed to bring the AFR at 1400rpm around 13.2...so, why still that extreme uneven condition between cylinders? It's not only that I am obsessed with the egt numbers but I hear that the engine is not running smooth at those low Rpms..sounds like some cylinders are not functioning properly


Obviously i think the problem is not the center tube that feeds the throttle bodies with air..let me know if I am missing something...should this problem get minimized when  rpm are getting higher?

I also started playing and modifying the VE table in the region between 1000rpm and 2300rpm for Map 100-110kpa , so that the actual lambda is close to the target lambda...

Could someone share their lambda or afr target tables that they use?  I need to know that my current lambda target table is consistent..

Thanks and regards

[JLaw]

Have you double checked your valve clearances are correct on all four cylinders, just a thought..

[modnrod]

Cylinder robbing... single throttebody feeding two cylinder. Even worse with carburetors. My first guess, swap EGT sensors from cold to hot cylinders and see what it does

X2

Try leaning out cyl 1 and 3

Yep. You could tune it up beautifully by adjusting fuel to reach exhaust temps so all cylinders match, you don't need a wide-band.

[Erlend / bug66]

My list would be:

Check spark plug leads: Meassure ohm on all four.
Check for vacuum leaks
Check injector wiring
Change sparkplugs


As a starter

[andy198712]

If you follow the air path on the outside of the bends, you see that 1 / 3 will get most of the air. If you move the centre tube to above the shroud, the air will still travel at the outside wall of each bend, but will meet the divider in the intakes straight on.

my fluid dynamics is a little rusty but i don't believe thats how it works.

lets say when a cylinder draws in its air, the air in the intake moves, say 12 inches down the tube, depending on volume, so does the other side as it draws in its air, so long as the volume of one cylinder's worth of air doesn't exceed the volume of the air in the intake tract to the TB opening, it shouldn't effect it.

do that seem right?

i case that off alot of OEM engine, look at the 6 cylinder BMW diesel engines, they have an inlet manifold shaped like a J, so number 6 is closest to the TB and number 1, a bout 2 feet further from it.... works well for them.

your thoughts?


good ideas on the general service items too, if its not following the injectors and its spraying and sparking as it should, could well be valves.

[Martin S.]

My screens are different but heres one for you.
I did not tune the engine so I cant help you there unfortunately.
I do have all my current tune screens available if you want to see one. 

[lakis1982]

i have just checked the valve clearances and all good...no problem there...

no vacuum leaks..

tomorrow i will try new spark plugs..

but anyway, today i was trying to play with the fuel trims per cylinder, so that i could lean or rich the cyls 1 and 3 that have low egt tamperatures...

In the following two videos you can see all the live data and the EGT values that i had... one video is when i was running the engine around 1200rpm and the other when i was running around 2000rpm..
the egt values in cyls 1 and 3 are in both cases quite different compared to cyls 2 and 4...no matter if i was trying to rich or lean cyls 1 and 3 , no big changes would occur...

https://www.youtube.com/watch?v=PMKB0FXEgfo&feature=youtu.be

https://www.youtube.com/watch?v=KfoF01tGzh4&feature=youtu.be

The guy where i took the injection system told me that the fact that i have one common throtthle plate per 2 cylinders is mainly the fact that i have uneven conditions between cyls and unstable idle... He told me that i cannot expect much with that setup.. WHen i bought the turbo injection system from him, my initial plans was to install it on a 1776 engine.. he insists that this setup could work with small engines like 1776 ...

do you agree ?

should i go to individual throttle bodies per cylinder as he proposes (which means more money to spend  ) or things will not change much and the problem is hidden somewhere else. ??

[Martin S.]

He does have a point. Most people on here have one throat per cylinder with Webers. (Although my turbo engine only has one in total!) Although I havent measured EGT on Kadron engines, I do know they idle badly and not on all 4 and need a stupid balance pipe like dual carbed stock T4 VW Bus motors to idle. They are one throat per 2 cylinders.
One weakness of adapting EFI to this setup is the need for manifold pressure to give the signal to the ecu that there is a change. Carbs seem to be less sensitive to this and will meter fuel nicely regardless. Where is the MAP sensor mounted on your setup?
To make things even worse in this regard can be the cam. Carbs wont care. But if you like the thumpa lumpy idle sound that cam with big overlap will have even less vacuum signal when the lobes keep both valves open.
How does the car drive? If the problem is near idle related, does it at least rip when you floor it?