Exhaust Gas Temperature Variance at Idle

Hi,

I have a general question about the exhaust gas temperature at idle. I have the following setup:

- Naturally aspirated 615cui Chevy BBC. Single intake manifold, compression ~11:1, runs on race gas

- Cam has a large valve overlap. At 1000rpm I get a MAP reading of ~0.8bar (with 1bar being atmospheric pressure, so _very_ poor vacuum)

- Engine is controlled by Haltech ECU and runs sequential injection

I just fired the engine up and noticed that the exhaust gas temperature at idle varies a lot and gets colder and colder with higher cylinder numbers (cylinders further away from the intake of the manifold). While the first two cylinders see ~300-400C, number 7 and 8 barely get up to 150C. This was while running at ~1600rpm, no load (idle). I can further see that the reading is correct since the exhaust pipes started changing their color on the front cylinders, but not yet on the rear ones.

The ECU has no corrections set up yet, the engine has not yet run under load. All I'm trying to find out is whether I should be worried about the temperature difference, or if this can be explained by sth like "the rear cylinders see less air because the front cylinders already take all the air you got in your manifold".

Any advice apreciated

I think that's an interesting observation, and I would keep an eye on it as you tune things. It would be interesting to see if different injector timing results in any change. Or if closing the throttle more and using more ignition advance (which should result in lower MAP values), helps to even out the EGT variance.

One thought, is it possible you have a manifold leak at the front of the engine, resulting in leaner mixture and possibly hotter EGTs. (or the inverse - leaner is not always hotter, only if the target is a richer mixture to start with) If you have per-cylinder fuel trims, it might be an interesting experiment to attempt to trim all cylinders for the same EGTs at idle.

Janek,

As David mentioned, I'd check for leaks first. I'd also confirm all spark plugs are alike if you aren't sure.

Beyond that, this is a good example of wideband data per cylinder being what you really want, rather than EGT data, if the goal is determining how much richer or leaner some cylinders are compared to others.

When measuring EGT, there are factors besides air/fuel ratio which impact temperature. Some examples are probe distance from exhaust valve, exhaust manifold design i.e. bends, EGT probe depth.

I understand buying 8 widebands (or 10 if you also have collector data), is a significant expense. While that's ideal, if you have one extra handy, I suggest placing it in a front cylinder, then a rear cylinder on the same bank, referencing each reading against your exhaust collector wideband reading in that same bank. I suggest testing at idle and 2500 RPM in neutral so you have a few data points. If you end up having large variance between cylinders during that test, and you've exhausted reasonable mechanical and electrical diagnosis, your particular engine seems it would significantly benefit from per cylinder wideband sensors and tuning.

In terms of diagnosis, if you haven't performed cylinder compression and leakdown testing yet, I'd pull plugs and check both. Cylinders in varied condition can have varied VE, resulting in varied result AFR.

Hard to say, but pulling the spark plugs and having a look may help identify if it's fuelling variation issue?

How is the fuel rail plumbing run, is it sized to suit, and is it possible the injectors are protruding into the rails and restricting their flow to the rear or front? With a large engine like that, I would assume you're using a dominator style TB - is it staged or all four opening at the same time - or there are some TBs that just use one huge butterfly - I was wondering if the TB may be biasing air to the front or back of the engine and amount of work the front cylinders and/or mixture variations were affecting the burn temperature?

Thanks for the responses, appreciate it. Let me start by answering a few questions:

- I do have one lambda sensor per cylinder, the readings are very noisy at idle though, especially given that the rear cylinders don't seem to fire at every revolution. On average the rear ones show a richer mixture, which lines up with the EGT reading

- The spark plugs show that the first two cylinders are happy, 3 and 4 are okay and after that everything is black in black with the rear cylinders barely firing

- The engine is a fresh built, compression is good on all cylinders

- The intake manifold is a plenum style manifold from Plazmaman. I don't think there are any leaks, the plenum is new.

I ran a few more tests tonight, played around with the ignition angle, but all that didn't change a lot. I then disconnected the injectors for the cylinders 1-4. Once I did that, the temperature of the rear cylinders did start going up into the normal range.

At this point, all the signs point to "the rear cylinders suffocate", so I guess I will have to put the engine under load with higher rpm to get any meaningful reading out of the EGTs. For now I won't be too concerned about the lumpy idle performance.

Thx!