Intake length experiment.

[sonex1566]

Hi Paul and everyone else who took an interest in this experiment,
I stumbled down a rabbit hole which has turned into the Bing rabbit hole not the Aerocarb rabbit hole! As to whether I made the right choice or not is hard to tell, I'm not sure myself. Originially I fitted an Aerocarb to my Jab 3300. As you know, the Aerocarb is easy to adjust, however getting it even across the rev range can be a bugger. My problem was that I wasn't getting even fueling between the front 4 and rear 2 cylinders. I tried a 2" intake extension in which I copied the shape of the original Jabiru fitting where I had to do a bit of creative shaping on my el-cheapo milling machine to fabricate it. After that work, plus fiddling around with different needles in the Aerocarb, it was better but I still wasn't happy......

Now this is where you get to choose which rabbit hole to go down.....In the Jabiru manual they state that to never exceed 1300 degrees F EGT. With the Aerocarb I was dilligently trying not to exceed this figure, however, after one flight someone on the ground commented that I was leaving a trail of black smoke. It also sounded rich, plus the soot on the bottom of the fuselage. I was getting almost 1300 on the front 4 whilst the rear two were in the 1100's. This was the point at where I spat my dummy out (I apologise for my Australianism). I put the original Bing carby as supplied and ran from Jabiru. As you know, it takes about 5 mins to swap the carbies over and about a month of stuffing around changine throttle directions, choke cables, fuel pumps, electric backup pumps and wiring. Phew!

Now Jabiru states to not exceed 1300 degrees, well straight out of the box my temps went to over 1500 degrees at Full throttle and not much less and any throttle setting more than 1/4. I believe that my EGT probes are generally telling the truth, but I can't accurately judge it. I swapped one over from my old AeroVee and got similar numbers. Not identical, but close enough. If it wasn't such a pain in the bum to change from the Bing to the Aerocarb I think that I would have stuck with the Aerocarb......well maybe!

Now in the 'Olden days', you could obtain a nifty plastic tray that contained a whole range of jets and jeedles for a particular brand of carby. It was easy to swap needles and jets until you got your tuning right. However, in Australia they just simply don't exist anymore. It's really expensive and shipping costs and shipping times are just broken. So instead, I spent $20 on some bar stock and proceeded to make 4 main jets, 4 needle jets and in the end 7 needles. Just changing the main jet from the STD 2.85mm to a 2.90 mm made the engine stutter and run too rich at WOT. It doesn't matter what size the main jet is, it has zero effect until you go from about 7/8 to full throttle. I then experimented with enlarging the needle jet. The STD is 2.90mm, I went all the way to 3.20mm. This instantly transformed my half and 3/4 throttle mixtures, however completely stuffed up my idle and especially the off idle and 1/4 you need when taxing. Fowled plugs, and pushing the bloody thing back to the hangar! Now for all you smarties out there who say, why don't you just raise the height of the needle? Well I can tell you......DON'T EVEN THINK ABOUT IT! If you raise the needle above the std Jabiru setting, at full throttle the needle can come out of the needle jet and get stuck on the edge and hold the slide wide open. Amazingly the engine still runs OK even down to about half throttle, but oh boy are you in trouble when you need to land. I have a successful, dead stick landing to my credit after the engine cut whilst turning base. Luckily the only pain suffered was the surgical removal of the seat cushion from my arse. Lesson 1 learned.

The end result which is now working really well over all the rev ranges is a new larger diameter needle, which has the high speed end lengthened which has the same effect of raising the needle, and then a constant 5 degree taper up to the 'idle' diameter of 2.60mm. I arrived at this measurement, by giving up trying to work out the resultant area and just make the needle the full 2.80mm diameter and machine a couple of thou off in two goes until the engine would idle and throttle up smoothly.

So after all of my fluffing around, my OCD is finally satisfied! I think that with the Bing the air/fuel seems to flow better within the plenum chamber into each of the intake manifolds. If I had a second Sonex, I would experiment with fitting maybe a butterfly or some sort of disruption within the intake trumpet to mix up the airflow a bit. One of the blokes I know has installed a single fuel injector body which replaces the carby and he had similar issues with unevern EGT's. Another friend installed the SDS multipoint EFI system which after a lot of messing around works beautifully. The big issue he has is that he can't start the engine on the EFI as the battery voltage drops too low at cranking and turns off the EFI ECU. Which means starting it with the Rotec throttle body and then switching over to the EFI once the engine is running. It has taken many, many hours of experimenting but I've finally got there. I'd be very interested to see how others have faired. So I'm not sure whether I've been a smart guy who has wasted a lot of time doing something a bit dumb or a dumb bloke who has stumbled onto a smart fix for the problem. Whatever the answer is, I have provided plenty of entertainment for the Airport experts where I spend far too much time!

[lpaaruule]

Thanks for the update. Keeping EGTs below 1300F at full throttle has eluded me, so I set the limit to the max allowed for cruise 1364F. Otherwise I’d probably be leaving black con trails too.

I’ll see if the length change helps me at all. I guess going from 1 too 2 inches didn’t have much affect. I also plan to roughen the inside wall.

[peter anson]

On my Bing equipped 3300, #5 always had higher EGTs than the other cylinders, about 720°C (1330°F) on climb and about 680°C (1260°F) on cruise. Coincidentally, or maybe not so coincidentally, at 470 hours the head came off the #5 exhaust valve totally destroying the engine, so maybe it's worth worrying about those high EGTs.

I also at one time had an odd problem with the engine running rough at full throttle, but cleanly at lower throttle settings. I messed around trying different main jets but everything I tried made no improvement or even made things worse. To complicate matters, at about that time the mechanical fuel pump on the engine partially failed and replacing that improved matters but didn't entirely solve the problem. To further complicate matters I was trying to "fine tune" a new adjustable pitch prop. The original problem turned out to be self inflicted. I had made a new NACA intake to the Jabiru air cleaner box. The air box has a small rubber flap to relieve pressure if inlet flow is too great. In an attempt to see if the NACA duct was feeding too much air in I had put a couple of small dabs of Torque Seal on the edge of the rubber flap so that I would be able to tell if the flap was opening in flight, and it appeared that it was not opening so all good. But it seems that Torque Seal was strong enough to glue the flap shut. As soon as I broke the seal the engine ran beautifully, well, until it committed suicide.

Peter

[Murray Parr]

I installed a honeycombe air flow straightener From http://www.saxonpc.com/120mm-airflow-ho ... us120.html on my Rotec TBI. I haven't ran it yet but I believe it could help with these issues. Might be worth a try

[sonex1566]

Hi Murray,
I have thought about putting, vanes or even just a round rod into the intake pipe to experiment mixing up the airflow. I just didn't try it. One of my mates went to a lot of trouble making an insert that created a spiral airflow. It looked really clever, it's just that it created too much restriction to the airflow and just reduced power. It does look pretty sitting up on his shelf now after all that work! The guy with the single fuel injector throttle body on the Jabiru 3300, he ended up pulling the plenum chamber apart and machining grooves and 'roughing up' the inside surface to disrupt the airflow, which is the exact opposite of the straightening vanes. I must admit that I've spent far more time trying out ideas that didn't work than ones that do. As always, good luck it's more interesting to talk about than the weather.

[lpaaruule]

I welded up one of my own yesterday, and roughened the inside. Also added some dimples.










I also have a honeycomb air straightener. It did help, but not as much as I had hoped.

If the valves are getting too hot, hopefully they will show some signs before just breaking, as one did for Peter. I recently borescoped valves, and they looked good; but my engine only has 175hrs.

[lpaaruule]

I was able to get in a short flight today. The results aren't quite as good as I was hoping for, but better. Maybe quite a bit better, as the following isn't exactly an apples to apples comparison because the first image if from over a year ago, and I likely had the AeroInjector tuned richer for winter. I didn't retune for winter this year, I decided to just back off the throttle a little once the EGTs got to the upper range. I figure the less I mess with the tuning, the less chance of be not getting the needle set screw tight enough.

Note: EGT #6 has a new probe but it's noisy for some reason, and I'm still looking into why. The old probe was spiking to over 2500F (not actual temp), after flying and at idle. (it's always something).

A year ago, and likely tuned a little richer


Today, with extended length intake :

[BRS]

Paul,
Are you sure using EGT temps is what you want to normalize? I've always learned to consider each cylinder as it's own little separate engine and any tuning for consistency between them should to try and get the EGT's to all peak as close as possible. That's not peak at the same temp, but to peak at the same fuel flow. This will make the engine run as smooth as possible.

Though on a carb'd engine there is really not much to adjust except what you are doing now. Let your peak fuel flow for each cylinder be your guide as opposed to the actual temperature. Then use final carb mixture to make sure they are not running too hot.

Take the above with a grain of salt as I'm used to flying injected Lycomings.

-brs

[lpaaruule]

I think they do peak at around the same time, and you're right, there isn't much, if anything, I can do with the AeroInjector to change that.

The Jabiru seems to favor the left side of the engine in both cooling, and intake flow. I suppose that rotating the carb would have some affect, but it would take a lot of work to rotate it more than at few degrees either way due to the fuel line, and the distributor caps being in the way.

I think that Richard is probably right about needing some other object to direct the flow. I'm reluctant to do that for fear of icing though. If I'm feeling energetic I may remove the carb mount/intake and weld in a vane, and try a ground run and some taxiing.