Hot rear cylinder example Jabiru 3300A-2575 and plenums

[facthunter]

Corrosion is deadly in these engines.  (or any with a steel bore). I've been barking on about inhibiting if not used  regularly, and always doing a full warm up if the engine is started at all. The gen 4 pistons may be a lot different. They'll certainly run different clearances I don't like the early pistons or the oil rings used. Nev

[RFguy]

Nev, from what I have been able to learn, the HD upgrade on the Gen4 piston thickened up the piston skirts, amongst other changes (dunno). As you know, Gen4 is ally pistons in niksil lined ally bores. there is no thrust skirt on the bottom of the bores in the Gen4, there was for the previous steel bores . There has been talk that that's why they went to the heavier skirt pistons  but that's only talk.

I've read much on the various   effects of temperature and time on 4032 ally cast pistons.  Keep em below 175C....  cumulative high temperature exposure leads to a reduction in the desirable properties....

 

 

 

Edited January 21, 2021 by RFguy

[facthunter]

Cast pistons can be fine except in the most high powered applications. Some don't get the heat out of them very well and may well need oil jets. The Jabiru is not high power for the capacity. I think all such engines need baffles to make the air do the best cooling job.  Nev

[RFguy]

Yes. I am sure the heat related Jabiru engine troubles over decades are 99% related to 'sub optimal' plenum and cooling design.  It can be soo much better so easily. anyway, we'll see what I come up with. There is considerable 'prior art' in this of course, many many ops coming up with improved cooling solutions. For one, its quite obvious  (looking at surface contamination or lack of it) there is almost no airflow on the #6 finning near the exhuast valve and duct - UNDERSIDE.  There is considerable radiated heat from the exhaust header back into the head.. etc etc. 

[facthunter]

Square edged fins don't work well either. Tapered with rounded edges flow air better down deep into the fin. Lots of fins where not needed and little where they are needed Re valve insert seats on your earlier reference, the High nickel and hard ones are fine. Beryllium copper is dangerous and bronze hates lead in fuel. Nev. 

[RFguy]

 

13 minutes ago, facthunter said:

Square edged fins don't work well either. Tapered with rounded edges flow air better down deep into the fin. Lots of fins where not needed and little where they are needed Re valve insert seats on your earlier reference, the High nickel and hard ones are fine. Beryllium copper is dangerous and bronze hates lead in fuel. Nev. 

YEP ! tapered is best but harder to mfr if 100% milling.   yeah wonder why there are not copper slugs pressed into the heads under the seats (via some interface metal) .  that would get the heat out of the small area quickly to the larger area. (instead of being very heavily concentrated) .... that's fairly advanced tech. its what we do in high power electronics. 

 

rather than wrapping the exhaust header in a shield (to stop the high radiation from the exhuast headers to the cly head) - , wrapping the headers  would mean they got hotter and would put even more heat back into the heads via the  flange,  ,so  the under head area needs a shield, which you might as well make as part of the baffle. Problem with too many too tight baffles /wrap around shrouds is there isnt enough air pressure at low airspeeds to make this work as well as you might think.  and then the radiation cooling of the head on the underside doesnt work too well. 

 

I favour extending the plenum over the lower bores, baffles for those gaps, extend the rear of the plenum down tightly at the back,

 *and*  use an air bleed hose from the plenum box  to direct spot  cooling air blast into the head underside side fins near the exhaust seat fins and ex-header. 
* and maybe* piped air blast  or shrouds on the underside of the bores. again, too many tight shrouds needs alot of pressure drop to work (not at low speeds) 

 

optionally, a water vapor mist underside in that blast, hey if it reduces the temps there 20 C then I will  I'll expend the 3 litres per hour of water...

-glen

 

Edited January 21, 2021 by RFguy

[Bruce Tuncks]

Good thinking guys. I recommend measuring differential pressures with a water-filled U tube. This will enable you to see how well the lower cowl is sucking from the ducts.  It also enabled me to see how the 2 ducts were different, something which I thought had to be caused by the prop rotation but further testing this idea proved negative.

The airflows under the cowl are not intuitive. I think it was Yenn who did some tufting and took pics to show just how different the airflow was from our intuition.

But the goal of cool and even temperatures sure is worth a bit of work.

 

[Bruce Tuncks]

There is some good stuff on this site. I was helped a lot. Also there is Limbach tech bulletin 44 well worth a read.

[RFguy]

most people have commented that the negative pressure with good cowling outlet design did more than any other single thing.

[kgwilson]

This requires little debate. The minimum requirement for the cooling exhaust area to be at least 3 times that of the intake is a well proven fact for ensuring adequate airflow through the engine area. Combine this with plenums that direct that air to all of the cylinder heads equally well and enough around the cylinders and other peripherals, then measure it and voila problem solved. The report commissioned in 1981 (40 years ago) by NASA covered it in intimate detail. See attached.

NASA_CR3405.pdf

[RFguy]

KG said - "Combine this with plenums that direct that air to all of the cylinder heads equally well and enough around the cylinders and other peripherals,"
 

and that is what is difficult to acheive  in the  stock jab setup. 

 

and full wraps and close fitting shrouds of heads and cylinders requires more pressure drop  than you have available at low speeds... and the shrouds prevent radiation  cooling  unless they are matt black at long IR on each side... 

 

and dont forget the plenum will deform around the edges  at pressure and little brooks will become big rivers (or leak flow) 

Edited January 21, 2021 by RFguy

[Bruce Tuncks]

Thanks kg, that was an interesting paper. More detailed than the Limbach one for sure. I was especially interested in the inlet losses through to the upper cowl being up to 50 % of dynamic pressure. Limbach say to streamline this entrance but do not have a figure.

But the NASA article ( I reckon it was NACA when written ) still refers to plenum chambers which are different from the ram-air ducts that Jabirus use.

A particular difference is that there can be cross-flow in a plenum chamber, so there can be no real difference left or right. This is not what I actually measured on my Jabiru.

 

[RFguy]

The plenum concept- constant pressure box filled uniformly and pumped with a high head of air, and ram air jab  (which is a mix of everything because the interaction between the inlet and the baffles and the deflectors etc all changes with airspeed) ) are quite different. the plenum approaches a set of bagpipes. 

[facthunter]

Don't OVERestimate radiated heat. It's not that much in this instance. Low speed and nose high go together. Air flow will go the path of least  resistance so it won't go  in the fins if it can go some place else. The lip on the cowl creates pressure drop at the expense of drag.  It's pretty clear some form of "in cockpit" adjustment is the way to go.. It's a proven fix.   Less drag when cruising too and quicker warm up if you want to use it that way.. Nev

Edited January 22, 2021 by facthunter

[onetrack]

The best air-cooled engines have tight ducting and fan-forced air going over their fins, at a positive rate. Is there a way to install some kind of simple, fan-propagated cooling air forcing, on the Jab engine?

[RFguy]

this topic is well discussed in depth elsewhere  in other topics, and also JabCamit group.io  ...

[facthunter]

I don't know about the BEST but some two strokes do, and some "pushers". Talking of aeroplanes and not Industrial engines or cars. Aeroplanes do generally have sufficient speed to not need fans. so IF you can do without, why not? There is a practical limit for HP/Capacity where aircooling is getting marginal, particularly with four valve heads. I'd put it at around 60 HP/Litre of engine capacity. Dust where it's a problem, is an aircooled engine issue. Nev

[Bruce Tuncks]

Onetrack, I remember the old Volkswagen. I loved the sound of that fan. But I think it took quite a lot of power to run it. Mind you, applied to a Jab, it would only need to run at low speed.

[facthunter]

Once you completely cowl an engine you are committed to that FAN keeping on running to operate.  It would only run a couple of minutes at load without it. Nev

[Bruce Tuncks]

I like Nev's point about the square cylinder fin-ends.  It takes a fair bit of work with a fine dremel bit to improve them, so it is not something you could expect to be done at the factory.  I only did a few fins on the hottest cylinder and it helped. Also, it does not increase complexity at all like a fan does.

 

[APenNameAndThatA]

I'm *not* an expert on this. Can a cracked piston or ring problems cause the oil to be burnt, or vice versa? If not, they why test that particular cylinder. I would have thought that ring problems would have actually decreased the risk of oil being burnt because of inefficient combustion. 

[facthunter]

You can make fins too deep and finely spaced,  where you are actually worse off unless air is directly forced through them. Best fins actually strengthen the head,  Radially orientated ones. Harder to cast and machine. Cast heads can have porosity issues, which is more of a problem with liquid cooling than air cooled.  Nev

[facthunter]

Piston cracks are usually at the skirt and will cause excess rock or disintegration, it you are unlucky.. A broken compression ring will increase blow by and that doesn't increase oil consumption. It can actually reduce it. If a valve head drops off you  will get the crown damaged and maybe holed and it may blow a lot of smoke and not fire at all. There's plenty of variables. IF your engine develops a vibration  it's a warning not to be ignored. Get safely on the ground at your earliest opportunity if you only have one motor. Nev

[Yenn]

I never did tufting tests but they have been done in the USA by Ron Brierley I think his name is. He runs a business selling top quality injectors among other things. He found on a C182 that standing tail to wind on run up was no worse for airflow than head to wind.

If you increase the cooling area covered by the Jab plenums to include more of the cylinder lower areas, you will be reducing the cooling effect on the cylinder head. You have a finite amount of cooling air and spreading it over a bigger area decreases its effectiveness.

You could look for any places where air flows out of the plenum and isn't in direct contact with the fins. That could include a gap between adjacent cylinders, which may need a V shaped duct to get the air between the fins.

[RFguy]

There is a Gen 1,2,3 piston upgrade available now, also....