Airflow Turbulence at carburettor flange

[Neil Spooner]

Hi Guys/Girls,

Please may I seek your collective minds for help?

I have an issue with poor mixture distribution from a carburettor that uses a spray bar and throttle slide (known as a Throttle Body Injector TBI) The holes in the spray bar are at the side of the spray bar in full rich, the spray bar is rotated so the holes progressively face the airflow to reduce fuel flow and lean the mixture.
The device matches fuel flow to air flow as the slide is opened by having a greater population of holes in the spray bar toward the full throttle end. This is fine except that if there is insufficient turbulence downstream of the TBI there is a bias of fuel toward one side of the intake duct. In my case this is leading to some cyl's running rich whilst others are lean only when at Wide Open Throttle (WOT).
The TBI is bolted to the bottom of the engine sump, the airflow goes through the TBI, into the ducting in the sump and then branches out in 4 directions to the individual cyl's. The two cyl's closest to the full throttle end of the TBI run rich, the other two lean, but only during WOT operation.

My question is this: How do I create an effective turbulent mixing without reducing airflow? (A contradiction I am sure!)

The possibilities as I see them are:
1-Fit a lip between the TBI and the sump flange that projects into the airflow creating turbulence.
2-Offset the TBI to one side of the inlet duct which creates a duct mismatch.
3-The sump inlet has a conical adaptor fitted to allow the use of larger carb' throat dia' of 50mm, I could remove that adaptor.
4-Modify the adaptors shape by cutting some away only on the WOT side, or the idle side.
5-Adding to the adaptor on either side.
6-Fit an aerofoil shaped fin in the sump entrance attached to the conical adaptor. If I go down this route what dimensions should the aerofoil be? And what should the "angle of attack" be?

Here is a link to a Flikr photo album which has pictures of the installation (the pertinent photos are at the end):

https://www.flickr.com/photos/62722241@N00/sets/72157647983156255/

Any suggestions or guidance would be very gratefully received thank you. I have roughed out some options that I think may be viable and are shown in the Flikr album.

TBI throat dia': 40mm

Turbulence before and after the TBI can affect its performance, that is why I have fitted a honeycomb air flow straightener before the TBI.

Best wishes,

Neil

 

[Greg Bernhardt]

Thanks for the post! Sorry you aren't generating responses at the moment. Do you have any further information, come to any new conclusions or is it possible to reword the post?

 

[jack action]

Maybe you could install a curved elbow between the carburetor and the plenum, positioning the rich side of the TBI towards the outside of the elbow. See Advanced engine technology by Heinz Heisler, p. 236-237 in attached file for more info (Note that figures 5.17 and 5.19 are mistakenly inverse).

 

[theOrange]

Airfoil idea intrigues me, probably because I'm an aeronautical engineer and love aerodynamics. I wouldn't be able to tell you the dimensions to use, however using symmetric airfoil is pretty obvious. For the angle of attack; you would have to run some tests and see what the effect is with increasing angle.

Have you looked through the PDF from jack action? Definitely worth looking at also, even if it doesn't help you it is still interesting.

Aside from that I wouldn't know how to solve your problem.

 

[Danger]

I'm very confused. The only throttle-body injectors that I'm familiar with simply replace the carb on top of the intake manifold, which takes care of distribution. What on Earth is yours doing in the oil pan, and how does anything get from there to the intake? Is this a 2-stroke that intakes through the crankcase?

 

[jack action]

What on Earth is yours doing in the oil pan, and how does anything get from there to the intake? Is this a 2-stroke that intakes through the crankcase?

I thought it was pretty obvious from the pictures. It's a boxer engine and since it is for an airplane, so access for maintenance is easier from the bottom than from the top. Also, maybe it is meant to be used in colder situation (altitude), so pre-heating the air through the oil pan could be a smart idea.

 

[Danger]

I thought it was pretty obvious from the pictures.

Thanks for prompting me, Jack.
Neil...sorry. My bad; those images didn't open for me the first time, so I was going by your words and thought that you were referring to a car. In particular, when you said that the injector was bolted to the bottom of the sump, I thought that you meant inside. That's doubly embarrassing because I used to fly. I really have no significant knowledge of pancake motors in general and Rotex in particular. All of my wrench-twisting was on Wedges and Rats; I never took a Lycoming apart. My apologies for the misunderstanding.
Now that I have a better idea of what you mean, I'll further demonstrate my ignorance by asking a couple of questions. What about just "roughing up" the interior of the manifold runners (the opposite of "polishing" which one would do in an automotive engine)? Maybe the more turbulent flow would cause enough mixing? It would probably just result in decreased performance, but perhaps might lead to something useful. Secondly, by the same reasoning, is there any chance that your "honeycomb" straightener is counter-productive? Would more turbulence before the injector be of any value?
That's all that I've got, and it's probably useless. If I somehow think of anything worth while, I'll get back to you.

 

[theOrange]

It would be nice to see if a solution was found by Neil, and what the solution was...

 

[Danger]

It would be nice to see if a solution was found by Neil, and what the solution was...

Yeah, it's been a while.

 

[Neil Spooner]

Thanks for your input.
Sorry I have not been back to you sooner.
The TBI (Throttle Body Injector) is a bit of a misnomer...it is really a pressure carb', similar to that found on a chainsaw/strimmer/leafblower, but uses a slide rather than a butterfly. Here is a link if you would like more: http://www.rotectbi.com/Installation.html
The units performance is rather complicated, in that at some throttle settings the mixture distribution is good, at other very poor. Some of this will be due to the density of the spray holes in the spray bar being greater at the full throttle end. It is also more complicated by the fact that the spray bar rotates to adjust the mixture.
Overall I would say I am happy now with its performance, as it is a compromise. If I could afford the expense fuel injection would be the perfect option!
Thanks for your support. I will close this request.
Best wishes,
Neil