Temperature Balancing by Turning/Twisting the Carb

[vwglenn]

Mike,

All VWs are notorious for the temp issue you describe. #3 is consistently the hottest cylinder in the cars for a multitude of reasons. It might be extremely difficult to beat that out of it.

[lutorm]

Our AeroVee also has a very uneven mixture between front and back cylinders. I tested a redesigned (3d-printed!) intake elbow to see if it would alleviate the problem:


The short story is that avoiding the 90-degree turn helped even the mixture out some, but did not fix the problem entirely. I've written the experiment up here. It has EGT plots comparing the stock and custom intake.

Ultimately I think the solution is to run a separate intake runner for each cylinder, none of this tying together of cylinders (especially the ones on the same side that are firing unevenly.)

[Fastcapy]

I have similar intake elbows and it bumped up my rpms a bit (150ish static) and evened out the temps fairly well. With the rounded elbows and the Rotec TBI I have pretty consistent temps across the cylinders now and haven't had problems with high CHTs like I did before the changes.

[MichaelFarley56]

That's a cool experiment! I can't help but agree with you guys; I tried to accomplish the same thing by pulling the red 90 degree intake elbows and installing a set of Great Plains (had to modify them a little to fit) intake elbows. In fact, they look a lot like the 3D printed ones above.

They helped slightly, but just slightly overall. On my engine, the front two cylinders nearest the prop are certainly richer while the back two are leaner overall.

I also think there may be some limitations with the airflow through the cowling as well, at least on my engine. My back two cylinders nearest the flywheel have nearly identical EGT readings, but the pilot side cylinder (Sonex calls it #1; I call it #3) runs around 30 degrees cooler than the rear copilot side (Sonex calls it #3; I call it #4). If there's a balance of mixture but a cylinder head difference, that almost has to be baffling/cooling air flow. I do have the top mounted oil cooler so maybe that's blocking some air? I have played with some air dams in front of the cylinders a little to get more air to the back cylinders but haven't had much luck there either. Oh well, it's something to mess with!

Ultimately I also agree that a redesigned intake system with 4 equal length intake pipes, one to each cylinder, would help balance things. If someone wants to make me one, I'd be happy to try it out!! :smile:

[kevinh]

Wow lutorm,

That is a great experiment (also a cool 3d print). I'm really curious about your future experiments!

Kevin

[Bryan Cotton]

Nicely done! Fun stuff!

[lutorm]

Yeah, it was fun!

Yeah, there's clearly also something to do with cooling, apart from the mixture isuue. My assumption was that the difference in temperature between the left and right was due to the propeller rotation which would make the airflow go down towards the cylinders on the left side and up towards the top of the cowling on the right side. I don't know how much difference that would make in flight, though, all my tests have been ground runs.

(It would be an interesting experiment to take pressure measurements inside the cowling, just above the left and right rear cylinders, and see if they differ. I'm also in the process of getting set up to measure cooling airflow in the spirit of http://www.n91cz.net/Pressure/PlenumPressure.pdf, but I was only planning on measuring the average upper plenum pressure.)

[kevinh]

FWIW - Even with a lycoming in my previous plane I spent plenty of quality time filing flashing off of heads and tweaking baffles to get the EGTs closer between the front and rear cylinders.

[peter anson]

I think there is only so much you can do to even up temperatures. The photo below shows the internal passages of the Jabiru intake manifold. It's really quite beautiful. They obviously went to lots of trouble trying to get even flow distribution, but my EGTs vary considerably, with a very noticeable difference between #5 and #6, the two outermost branches of the tree (top in the photo).
Peter

[DCASonex]

For what its worth, my old Jabiru 3300 which had intake manifold with the single airfoil flow divider, tilting the carb very slightly did have the desired effect, also with that manifold, replacing the airfoil shaped flow divider with a plain 1/2" (12 mm) dia. round rod cut the EGT spread at WOT about in half but at a cost of slight increase in spread at cruise.

My new CAMit engine appears to have an intake manifold identical to later Jabiru ones, with two round rod flow dividers. That does seem to result in reduced EGT spread, but tilting the carb has little effect with this manifold.

Briefly tried the new manifold on old engine after some work on it, and seemed to behave same on both.

David A.