Jabiru Gen 4

[RFguy]

10 thou. wow. . My pistons rocked up and down I think about 8 thou cold clearance at the crown on the thrust axis.  apparently rather normal, and inside the jab spec. Here is a rotax 912 ULS piston . solid skirt, note the drilled ports *I think* behind the oil ring. 

Edited April 17, 2021 by RFguy

[onetrack]

No need to get too picky about absolutely precise clearances, apart from the specified factory clearances.

After all, engines still run very satisfactorily with a few thou wear in every component, they just become a bit noisier, that's all. Good oil is important.

Around 70% of the parasitic frictional losses in IC engines is caused by the shearing of the oil molecules as the metal bits slide past each other.

Heat is definitely the killer of all engines, and of engine oil. Excessive heat alters clearances substantially, alters metal properties, and substantially degrades oil and its additives.

So keeping the engine heat within the specified range is the most important thing to concentrate on. It's better to have clearances a fraction loose, than to try and obtain perfect tight clearances.

New engines are "tight" and can be seized by excessively hard work, and high temperatures. New engines produce more heat from friction than "run-in" engines.

[facthunter]

They are just oil drain holes. A lot of the engines friction is from the oil rings, and some MODERN bearing materials. (alutin and copper lead) which must have extra clearances to allow more oil flow. Short con rod to crank throw  figures give high side loads and more piston slap noise. There's always one side of the piston firmly against one side of the cylinder. Heat is not particularly a problem in a LOW HP/Capacity motor. 85 HP for 2.2 litres. About 38/litre PROVIDED you get most things right. A new aero engine should not be tight, or it  would risk failure.   You can't run them in at low speeds like one used to with a car The running clearances have to be right from day one. .

 It does generate more friction than when it's got a few hours up and I would avoid high ambients at this time.  (and/or climb at a faster speed as well. Don't overfill with oil. Nev

[turboplanner]

On 16/04/2021 at 8:04 PM, RFguy said:

OK on that.

In terms of reducing piston temps, for this engine, any other suggestions ?

 

You can cool the combustion chamber with richer mixture.

[RFguy]

My feeling is, at least in cruise, its probably rich enough already, judging by the deposits.

However, Is it rich enough when it matters- in WOT  max TO RPM,  or max RPM is another question.  Need to look at what the fuel flow and EGT does. The engine is 36 hp/litre at 3300 RPM..

The pre Gen4 pistons which were adapted from an engine that ran 54 hp/litre @5200(natural asp)  and 61 hp/litre @ 5200 (superchaged) in wet cooled bores.
The so $64 question  is, do the pistons run any hotter at roughly half the output than they would in the water jacket bores ?  

My guess is yes, as the water jackets are 80 deg C, the bore jackets >180C, that's more than the difference shown in temp versus RPM variations.  However, the bores is not the only way the pistons get their heat out (oil, cool gases).

 

Edited April 19, 2021 by RFguy

[facthunter]

It's one way and probably can disperse quite a bit IF the piston is designed properly Heat can't get to the bore unless there's contact there. Ovality and the slipper design aren't helping that.. nor is a steel cylinder not cooled well..( Blue colour near exhaust ports. Distortion and lubing problems

  Heat is much more  at full throttle and that's where it should/must be enriched. Lycs and Conti's are operated at full throttle to get the richness. coming in.  Doing reduced power take offs is definitely  NOT recommended for them.. 

 Note a lot of Jab's engine problems happened when they experimented with Lean Jetting.  Nev

 

[turboplanner]

6 hours ago, RFguy said:

My feeling is, at least in cruise, its probably rich enough already, judging by the deposits.

However, Is it rich enough when it matters- in WOT  max TO RPM,  or max RPM is another question.  Need to look at what the fuel flow and EGT does. The engine is 36 hp/litre at 3300 RPM..

The pre Gen4 pistons which were adapted from an engine that ran 54 hp/litre @5200(natural asp)  and 61 hp/litre @ 5200 (superchaged) in wet cooled bores.
The so $64 question  is, do the pistons run any hotter at roughly half the output than they would in the water jacket bores ?  

My guess is yes, as the water jackets are 80 deg C, the bore jackets >180C, that's more than the difference shown in temp versus RPM variations.  However, the bores is not the only way the pistons get their heat out (oil, cool gases).

 

I don't think it's liner calculation to compare hp per litre; everything's cooler in a water jacket, but I found it you get combustion wrong in aircrooled barrels, even high speed fans bloing them cool won't save the pistons, which in my case were the canaries that signalled disaster.

 

For example, big block V8 (c. 400 cu in) engines in sprintcars now produce 1,000 hp and you can flip  car over its rear wheels onto its back if you were so inclined, but they still run reliably at full power in races.

 

Top Fuel drag racing cars which are still based on V8 blocks, albeit with a lot of changes are producing 11,000 hp

 

Both those engines are powered by methanol, the top fuelers with the addition of nitromethane and take advantage of the fact that you can run an engine so rich on methanol that a fair amount pours out the exhaust cooling all the way through.

 

Plugs are read after a long burst of WOT acceleration with an instant engine cut, and with petrol, there are charts showing the colours from extreme rich to extreme lean, with the aim usually being a honey brown colour.

 

Coming back to petrol and 80 hp behind a prop, trying to read plug that way is a safety problem, but if you have a long strip with no other aircraft using it, you may get some good readings on a warm engine with WOT as in take off and mixture cut before it leaves the ground, then look at a plug.

 

In about 40 cases I looked at, 16 had burnt exhaust valves, which was telling me the chambers were very hot, and in those the pistons had stood up to it, so I was looking at mixture being too lean at WOT.  I didn't go any further and experiment with mixture.

 

I don't think combustion chamber temperature relates to EGT, so I don't think you'll pick up the interior changes on EGT gauges even though they quite accurately reflect changes outside the combustion chamber.

 

 

 

[facthunter]

IF a nitro engine doesn't fire reliably it's likely to hydraulic the  cylinder, there's so much liquid gets there. They have incredibly powerful Igniters. and only do about 900 revs total for the whole run. They don't even bother with a  cooling system or any jacketing..

  Fuel flow (with required oxidant , air) is proportional to power roughly and there's a relationship with heat that has to be dissipated if you aren't using the motor as a heat sink. Water meth injection on take off would help an engine with marginal heat dissipating ability. It's not a new  idea and it works. Nev

[WaterWings]

Is there a reason we don’t simply use a $300 wideband oxygen sensor in the exhaust and EFR gauge to see the actual mixture in all stages of flight, at all speed and throttle settings, hot and cold day, hot and cold motor etc?

I know that unequal mixture distribution won’t show up.

[Old Koreelah]

A simple fuel injection system would solve many Jab engine problems, but EFIs are dependent on a totally reliable battery. 

Old-fashioned mechanical fuel injection might be a better alternative, if only one could be found.

[RFguy]

I dont really think its that a big deal. easy enough to get the mixture close.

 

the lambda (mixture)  sensors dont like the lead much....

 

 

 

 

 

[facthunter]

Individual mechanical injectors will get the mixture quite even if you do the Gami type thing.  Float equipped carburettors  are a risky thing to have on an aero motor. Better off with something else really but it needs to not require electricity   Some of the mechanical injector systems were easy to light up a grass fire on start up if you weren't careful. .Nev

Edited April 21, 2021 by facthunter

[turboplanner]

The standard system controls the mixture exiting the carb by the jet hole size, which of course is fixed. Does’ t cost anything to look at a plug after the proper test procedure.You just need patience to fit a slightly bigger main jet until you get the right colour. We are talking tiny increments because the mfr usually leans the engine for maximum power so you don’t want to solve one problem by creating another. If oxygen sensors could be fitted easily fitted To the standard engine that would be good. The ones I’ve seen normally inform the ECU which makes an adjustment. If you could get leads to a recording chip that’s all you would need and you could adjust main jest from the data. Some people have mentioned uneven manifold distribution and this would allow you to run varying jet sizes, again in tiny increments so you don’t transfer the work to other chambers. No one ha reported that they’ve got under the aircraft and measured the space to see if an intake could be fabricated for even feed. You have to solve the even feed problem before spendin money on fuel injection and if you do you won’t need it.

[Bruce Tuncks]

Very interesting RF that not all older jabs do the tight turnover thing when hot. Around here they all seem to do it, and it was very noticeable that the gen 4 didn't. I really liked how the gen 4 felt bouncy even when hot.

The thought that it could be just one piston has never occurred to me, thanks for the idea. How could I test that out?

[facthunter]

With most engine assemblies if one piston is tightening the other's won't be far behind. It's ridiculous to have an engine doing this especially in an aeroplane. Some new/rebuilt Con Lyc engines in the tropics with high ambients do burnish a piston slightly in the first few hours. That means your solid piston running clearances are as small as one should ever go. Nev

[RFguy]

Hi Bruce. I dunno how you can test that without some fancy sonic gear (tap on the case, the tap noise will be damped by piston contact)

 

Understanding pistons and bores a little more now, I think the Gen4 is a significant improvement in that department.

Finally Aluminium bores and solid skirt pistons (like Rotax) . Much less likely to have a piston throw IMO, considering the pistons should be stronger as they should be running cooler..... (solid skirts) , get the heat out better (solid skirt, finned ally bores) . I know of one piston throw in a Gen4, but that was an very early Gen 4, and put down to a assembly error by the factory (and before the change to the solid skirt pistons) .

[facthunter]

The alloy bores expand more than steel so you can get closer running clearances and less Piston rock giving better ring sealing. . The water cooled heads of the Rotax would help the top of the bores as well. Compression is usually held well with Rotax 912. Nev

[Thruster88]

The use of slotted pistons in the gen 1,2,3 engines makes no sense to me. Slotted pistons are used to make engines quieter,  not really important in an aero engine. The elevated crown temperature compared to a non slotted piston must surely make ring sticking from carbon as experienced by RFguy more likely.  

[facthunter]

Not only that but they are extremely weakened. Totally unsuited for the job. High crown temps affect detonation also.. nev

[RFguy]

I'm fairly satisfied now that possible reasons for piston throwing in < gen4, which I  beleive that I now have a good handle on, have been substantially reduced in Gen 4.  (piston temps, piston type, bore casing expansions, bore casing heat removal) .

 

The 4032 pistons (high silicon)  have a lower expansion rate v temp than the alloy used for the bores and head.  

 

The Gen2,3- cooling the bores is a bit of a double edged sword. You need them to get hot to expand enough as not to have an interference fit when the pistons get red hot.

 

I hypothesize  that the pistons running with initially high clearances, get hot enough , expand and  make good contact with the bores, and then pistons then get their heat out to the bores.

This establishes some sort of equilibrium. 

But it is a hairy edge of not becoming a tight fit if either :

 

a) pistons are suddenly hot, without time for heat to propagate to the bores . The heat capacity of the pistons is little compared to the heat pumped in, so I would expand piston temperature changes to be almost instantaneous.  It takes time for the thermal flux to propagate to the bores , so the bores may be caught off guard. That is the bores cannot expand fast enough to keep up with the piston expansion. This probably suggests avoiding sudden increases in power, go from idle to WOT over a few seconds, dont punch it.

 

b) engine shutdown - The bores cool fast,  and the pistons heat soak- that is the usual designed temperature distribution changes  ,and the piston expands much more on its major ovality axis, meanwhile the bores are contracting fast, and the engine becomes hard to turn when recently shutdown, thus interferring with the bores.

 

 

 

Edited April 22, 2021 by RFguy

[turboplanner]

In my experience, pistons under full accelerating load went from normal running to start of seizure in about 2 or three seconds.

[onetrack]

Quote

 This probably suggests avoiding sudden increases in power, go from idle to WOT over a few seconds, dont punch it.

That's a handy restriction to have, when you're 15 feet off the runway on landing, and the aircraft is getting seriously crossed up, and you need full power instantly to go around.

[RFguy]

45 minutes ago, turboplanner said:

In my experience, pistons under full accelerating load went from normal running to start of seizure in about 2 or three seconds.

Turbs, was this application of WOT from low power setting 'applied  instantaneously', and the time  till the seizure ?

[turboplanner]

8 minutes ago, RFguy said:

Turbs, was this application of WOT from low power setting 'applied  instantaneously', and the time  till the seizure ?

Usually towards the end of a straight (in racing) with WOT and the car accelerating. Sometimes you could instantly back off and the seizure wouldn't occur, most times there wasn't enough reaction time to stop the seizure.

[RFguy]

Turbs my GUESS is that there was some heat capacity available beyond the continuous heat duty and this got used up. 

 

In electronics, we have usually high thermal mass 'slugs' under power silicon chips: 

 

IE a sandwich of  chip >> copper slug >> aluminium heatsink.

 

For sustained heat removal , the heat just travels through the lot to the heatsink. But the amount of heat that can be removed per second is limited .

 

For transients and momentary overloads , the high heat capacity of the copper slug can absorb the heat pulse  out of the silicon chip, and stores it for (slower) forwarding onto the heatsink over time.  

 

I gather there is a bit of that going on in said drag strip service, a 2nd order effect. IE you have transient loads that have some heat capacity , and sustained dissipation limits. the piston will heat up fast, and there will be some heat capacity in the conrod, bore walls etc , and once those heat wells are filled up (come up to temperature), the piston temp rise will accelerate.