Jabiru cooling

[Bruce Tuncks]

My left 2 cylinders run about 20 degrees hotter than the right ones. I can't see why this is. 

 

A perceptive suggestion was that the prop is going up as it passes the LHS duct inlet and this acted in conjunction with the entry of air at high angles of attack to give less air into this side.

 

Today, I flew the plane with a deflector "eyebrow" to make more upgoing air enter the LHS duct. I had tried this before, but decided to try again with a bigger and better-shaped deflector.

 

But this didn't work either... here's the max climbout temps...  T1=136,  T2=157 , T3=144 ,  T4=161.

 

So the LHS 2 and 4 were about 20 degrees hotter than the RHS 1 and 3.

 

Here's my intentions: firstly, to get some clear plastic tube and put one end in a duct and the other end in the lower cowl. Make a "U" tube on the panel and fly to measure the cooling air pressure differential.

 

What comes next depends on the figures . Any suggestions?

 

The ambient was 25 C . I don't think the deflectors made any difference at all. Am I being too fussy with those temperatures?

 

The EGT figures were much closer to each other.

 

 

 

[dlegg]

Those temps are really very good at max climb out on a warm day. Temps can be pretty subjective, measuring device accuracy, location of probe etc. An air cooled engine like a jab should be around 150 at least for optimal operation(in my view) as they are designed for that. I'm happy with my 3300, CHT's at cruise are around the 150, obviously hotter in a climb. The big benefit of individual cyl probes is being able to monitor temps over time and being able to recognise a problem as it develops. IE lower temp, not making power, or running hotter maybe leaning out with valve or intake problems. 

 

 

[Bruce Tuncks]

Thanks dlegg, that's reassuring. I will go ahead though and see what pressure difference there is between the inside of each duct and the lower cowl. My lower cowl has been treated in that the air going into the lower opening is all ducted to go through the oil cooler, with little or no wasted ram air just increasing the lower cowl pressure. But there is no skirt at the lower cowl lip. There is a Jabiru specification on the minimum allowable pressure difference and I would hope to meet this figure at least, even without any skirt.

 

 

[Yenn]

You could also check the pressure difference between left and right side upper cowling.

 

If the left side is hotter and the pressure difference reflects this, you need to get more air in that side.

 

 

[Old Koreelah]

...Here's my intentions: firstly, to get some clear plastic tube and put one end in a duct and the other end in the lower cowl. Make a "U" tube on the panel and fly to measure the cooling air pressure differential...

 

An experiment well worth doing, Bruce. Here is the setup used for testing the CAMit engine.

 

[skippydiesel]

Replace with a Rotax 912/914

 

 

[Bruce Tuncks]

Results from next flight: The eyebrow deflector was removed and the LHS duct and the lower cowl were the ends of a water manometer.

 

EGT's were: 1=687;     2=680;    3=692;  4= 675   ( all degrees C ). Ambient was a warm 29 degrees on the ground.

 

So the hottest CHT cylinders, 2 and 4, were actually the lowest EGT cylinders, but the difference is trivial and the EGT's are quite even.

 

The CHT's were:  1= 129 ( coolest ) and 4= 169 ( hottest) . ( max temps on climb)  Maybe that deflector did a small bit of good?

 

The manometer showed 8 cm of differential at 80 knots. Since the theoretical limit is 10.3 cm, ( at sea-level ) I reckon this shows that there is no big problem with the ducts and the lower cowl pressures. 

 

But bugger, after taxying back, the manometer stayed about 4 cm even after switching off!  This went away when I squeezed the tubes, but now the 8 cm reading is suspect and needs re-doing. And the manometer needs an improvement first.

 

Next, I plan to follow old K's suggestion and plumb the manometer between the 2 ducts.  They should be equal, if not, then that will be interesting.

 

Gosh, you need to find a problem before you can fix it huh.

 

 

[Old Koreelah]

Bruce when doing this test pilot stuff I found it necessary to have a camera recording the in-flight data to avoid being distracted from the important job of keeping the aircraft within safe operational limits.

 

 

[facthunter]

Verily ,thou speakest a mouthful. Set up your go pro. Distraction!!!! A key element of a good proportion of aerial mishaps.  Changing one small bulb put a DC 10 into the Everglades .Nev

 

 

[Bruce Tuncks]

Good advice guys about aviating first and foremost. Today ( overcast, ambient 21 degrees ) the manometer was fixed up and the 80mm water-gauge at 80 knots was confirmed for the LHS duct.

 

Strangely, the hottest CHT today was number 2! on climb, the readings were T1=130, ( coolest) and T2=155 ( hottest) Hotter than T4 by 4 degrees.

 

So the LHS 2 and 4 were still hotter than the RHS 1 and 3 by about 20 degrees, but the front ( cylinder 2)  was hotter than 4. I have no idea why. Of course 4 degrees is not worth bothering about.

 

On the weekend, the next measurement comparing RHS and LHS ducts will go ahead.  I will use a camera next flight. I agree there is too much to do  with taking measurements and flying the plane.

 

On the cooling-air pressure difference between the ram-air duct and the lower cowl:

 

Jabiru state they need a minimum of 60mm at 1.3 Vs. If Vs=45 knots then 1.3Vs= 58.5 knots=30m/s then the water pitot head on my calculations is 55mm What have I got wrong here?     If a perfectly sealed pitot can not make the minimum specified pressure difference on my calculations,  then there is some problem with my theory or calculations. All I did was to equate 1/2times  rho( air) times  v squared with rho ( water) times g times Ht where Ht is the water-gauge reading. 

 

 

[pmccarthy]

I don’t know but air density is 1.225 kg/ cu m and water is 1000 kg/ cu m.

 

 

[facthunter]

 In that formula I think air density is in SLUGS (whatever that is in other measurements). Nev

 

 

[Bruce Tuncks]

Well old K's suggestion of putting the manometer between the 2 ducts sure paid off.

 

The RHS duct has 4.5cm more pressure than the LHS duct. This is far more than I expected, and it means that the RHS duct would be 12.5 cm of water head at 80 knots.

 

This is more than the pitot pressure and so it answers another question, that is the one about the Jabiru specification of minimum water-gauge of 6cm at 1.3 Vs.  The answer is that the extra pressure must come from the prop. Even though it is going slow near the inside and the airfoil shape is poor there, it must be making a bigger contribution to the pressure  than I thought.  

 

Looking at the ducts, they are the same except for the LHS ( the hot side) being further back and that the prop is going upwards as it passes the LHS entry hole.

 

What can be done? A few things come to mind, none of them real easy to implement.

 

PS.  Nev, pmc's units for air density at sea level are correct. Aviators like yourself had the metric changeover harder than anybody else, what with the USA sticking to imperial stuff and aviation staying with the US for lots of things.  And slugs are one of the worst things of the old imperial system, because they are not part of ordinary life. Who goes and buys a slug of spuds?   

 

 

[skippydiesel]

Has anyone mentioned the air exiting the cowling as a possible contributor to your problem(s) ??

 

 

[facthunter]

Angling of the intakes to account for AoA on climb and the prop shape near the hub would have to be beneficial.. I've flown Jabs in 42+ degrees and just increasing the climb speed a bit stops the temps going too high. (Higher than normal). You have no idea what the piston crowns temps are and could easily make a huge difference.  In pre war Chevrolet cars the original pistons were cast iron and they pinged very easily. When alloy pistons are fitted it's nearly impossible to get any pinking even a very low rpm and a lot of load. That's because of the reduced piston crown temps.  Slotted pistons as are in the Jab are not good at getting the heat to the skirt .Nev

 

 

[turboplanner]

 I've flown Jabs in 42+ degrees and just increasing the climb speed a bit stops the temps going too high. (Higher than normal). 

 

J160 and J170?    (A lot of people in those have been told not to fly above 35 OAT)

 

 

[Kenlsa]

Good advice guys about aviating first and foremost. Today ( overcast, ambient 21 degrees ) the manometer was fixed up and the 80mm water-gauge at 80 knots was confirmed for the LHS duct.

 

Strangely, the hottest CHT today was number 2! on climb, the readings were T1=130, ( coolest) and T2=155 ( hottest) Hotter than T4 by 4 degrees.

 

So the LHS 2 and 4 were still hotter than the RHS 1 and 3 by about 20 degrees, but the front ( cylinder 2)  was hotter than 4. I have no idea why. Of course 4 degrees is not worth bothering about.

 

On the weekend, the next measurement comparing RHS and LHS ducts will go ahead.  I will use a camera next flight. I agree there is too much to do  with taking measurements and flying the plane.

 

On the cooling-air pressure difference between the ram-air duct and the lower cowl:

 

Jabiru state they need a minimum of 60mm at 1.3 Vs. If Vs=45 knots then 1.3Vs= 58.5 knots=30m/s then the water pitot head on my calculations is 55mm What have I got wrong here?     If a perfectly sealed pitot can not make the minimum specified pressure difference on my calculations,  then there is some problem with my theory or calculations. All I did was to equate 1/2times  rho( air) times  v squared with rho ( water) times g times Ht where Ht is the water-gauge reading. 

 

Bruce,

 

I look at your CHT numbers and know you run digital but in reality if you look at the jab issued vdo temp gauge they are barely the thickness of the difference of the needle.

 

I look at my Colt CHT and is says 1-2-3 so i guess when it goes furter to the right on the dial that would be called very hot as it has hit the needle stop pin. I tend to be happy if the needle isnt bending on the pin and that the slip ball is still in the glass, and hasnt blown out the end.

 

Remember all the traffic on the EGT a couple of years ago, and i am reminded in the "old days" you leaned the mixture until it was in the green, and this was on only one cylinder and had no numbers.

 

I wonder if all we get is a headache :)

 

Ken

 

 

[Markdun]

Bruce,  I've read a bit about the extremes some of the RV guys go to to get an extra knot or 2 by reducing cowl inlet and outlet. Anyway a common thread is that the outlet, or more precisely just before the outlet,  is very important.  Stuff like gascolators, airboxes battery and even engine mount tubing can reduce the air flow and in particular affect port and starboard sides differently.  It would seem to me that as the inlet pressures are the same it has to be the outlets.

 

I note your EGTs are very close for a Jab. I'm still getting a spread from 640C to 700C.  I don't see CHTs up around 150 except on 30+C days with two people in the plane and after a long slow climb out, mostly they are between 120 and 130.

 

 

[Hargraves]

Jasus H Darwin how come your doing all the R&D work for this engine, get a jabarax and be done with it sir.

 

 

[Yenn]

As I suggested comparing the pressures on each side and you find they are different, you could try putting in a pipe between each side to even up the pressure. That could drop the high side temps, but it would lift the low side temps. I would think a 1/2" pipe would do the trick.

 

 

[skippydiesel]

Has anyone mentioned the air exiting the cowling as a possible contributor to your problem(s) ??

 

Still waiting for one of the "experts" to comment on this - As I understand (little to be sure) for air to enter, it must also exit. Effective air flow design must address both in & out let.

 

 

[coinz]

Still waiting for one of the "experts" to comment on this - As I understand (little to be sure) for air to enter, it must also exit. Effective air flow design must address both in & out let.

 

Yes,and i would be thinking of air exiting the ram duct,at the rear.I would be thinking lowering the pressure of the LHS even more,are you able to lift the rear of the duct 5 or so mm,to help more air flow into the front of the duct,thru less resistance at the rear.

 

colin

 

 

[skippydiesel]

Yes,and i would be thinking of air exiting the ram duct,at the rear.I would be thinking lowering the pressure of the LHS even more,are you able to lift the rear of the duct 5 or so mm,to help more air flow into the front of the duct,thru less resistance at the rear.

 

colin

 

Sometimes the fitting of a small lip/spoil (mini cowl flap if you will) on the trailing edge of the exit "duct" creates sufficient additional  negative pressure in the vicinity to improve air flow through the cowling.

 

 

[Bruce Tuncks]

There is plenty of room around the engine in the lower cowl, it is hard to see how the pressure can differ much from one side to another.  Still, the option of adding a skirt is there, and it would be easy to impede the airflow in the cool duct to make cooling  more even. That skirt is a drag-raising thing though. But worth it if it is needed.  Jabiru say that a 20mm skirt at 60 degrees (?)  gives  2cm of water-head gain.

 

Running a connection between the 2 ducts should help to equalize the pressures and therefore the temperatures as old K says.   My feeling is that this pipe  would need to be as big as possible. Maybe a bit of scat hose.

 

Most planes have a plenum chamber, and I reckon the Jabiru ducts make for a nice clean installation, but this uneven pressure business is a downside.

 

Yes the EGT's are now very even. They used to be very uneven, and I tried lots of things. The breakthrough was when I was told ( on this forum ) that nothing BEFORE the carby made much of a difference. So now the engine has a cross-piece vane just downstream of the carby with the vanes bent to deflect fuel droplets into the rear inlet pipes.

 

What I think was happening was that unvaporized fuel droplets were flying past the no 3 and 4 pipes to finish up going into the front (pipes 1 and 2 ) and making their cylinders  run richer.    

 

 

[Yenn]

Where is old K involved in this conversation?