Diesel vs miller cycle

[stanislavz]

Dual ECUs with dual sensor suites as Rotax does on their iS engines and we do on Lyconentals. No limp mode in aircraft ECUs since you need relatively high power to maintain flight compared to automotive applications where you only need 20-30 hp to limp home. Most sensors outside of crank sensors are not needed to keep the engine running to get you home safely if you mitigate with defaults.

View attachment 118340

I was in megasquirt bussines for more than a year - connectors in this photo is rubbish to say at least. As other connector too. I have some failures on ms ecu before I moved to some propertiary connector to reuse old wiring in car...

 

[rv6ejguy]

I was in megasquirt bussines for more than a year - connectors in this photo is rubbish to say at least. As other connector too. I have some failures on ms ecu before I moved to some propertiary connector to reuse old wiring in car...

I've been manufacturing ECUs for 27 years. Our ECUs have over 20 million hours on them collectively, 800,000 flight hours, multiple Baja off road wins where they are getting beaten to death, zero failures with these connectors. Rubbish? Based on what? Garmin and most other certified avionics still use D-subs. Rubbish too?

Your older comment here about using Alpha-N on aircraft shows your experience level in aviation. That's rubbish. The aircraft could be at sea level or 20,000 feet. Alpha-N can't tell the difference.

 

[stanislavz]

I've been manufacturing ECUs for 27 years. Our ECUs have over 20 million hours on them collectively, 800,000 flight hours, multiple Baja off road wins where they are getting beaten to death, zero failures with these connectors. Rubbish? Based on what? Garmin and most other certified avionics still use D-subs. Rubbish too?

Yes, D subs was totaly rubish. But - cars have much more humidity + salt in winter.

 

[rv6ejguy]

Yes, D subs was totaly rubish. But - cars have much more humidity + salt in winter.

ECUs are mounted inside cars in most cases until recently. Hopefully no salt or water there and in any case, thousands of customers have driven our ECUs on winter salted roads- zero connector failures. We've also sold hundreds to the marine market. Put in a waterproof enclosure, they also work fine.

 

[PMD]

These last few posts open another subject well worth discussing. I build a lot of stuff that has to tolerate road spray and once a lot of marine wiring and there is simply NO FRIGGING WAY any kind of open connector (i.e. not hermetically sealed) can be tolerated. I haven't built an aircraft harness for decades, but as I recall, NONE ever used a sealed connection arrangement. In both OTR and marine environments it is often the open end of a multi-strand wire that has tremendously effective capillary effect that draws moisture with attendant corrosive ions (chlorides especially) that result in corrosion and stress corrosion cracking that causes the wire to ultimately fail. Why does this not seem to be considered in aviation installations ahead of the firewall????

 

[dwalker]

These last few posts open another subject well worth discussing. I build a lot of stuff that has to tolerate road spray and once a lot of marine wiring and there is simply NO FRIGGING WAY any kind of open connector (i.e. not hermetically sealed) can be tolerated. I haven't built an aircraft harness for decades, but as I recall, NONE ever used a sealed connection arrangement. In both OTR and marine environments it is often the open end of a multi-strand wire that has tremendously effective capillary effect that draws moisture with attendant corrosive ions (chlorides especially) that result in corrosion and stress corrosion cracking that causes the wire to ultimately fail. Why does this not seem to be considered in aviation installations ahead of the firewall????

I built racecars for a wide variety or series from Pro to historic, Baja, Pikes peak, various rally cars, etc. and wiring and connectors are a very simple thing really, that a LOT of people get wrong. I did beta testing for AEM, went to Motecs "school" about 25 years ago, have worked with almost every single aftermarket programable ecu in existence and many that are no longer made, built wiring harnesses from components, rebuilt "professionally built" looms, and thrown away very expensive junk that looked like the right stuff but was completely unsat. I have not built or sold the number of systems Ross has, but I have a very high level of shop and field experience. The reason AEM used me as a beta tester is I used more of thier products capabilities and went so much further than the basemaps that I found problems very few ever would, and I often could solve those problems. I adapted thier systems to fit engines they never intended for them to fit, and I am fairly sure they had a lot of "Did he really" moments.

First is lets talk about the ECU and wiring connectors/housings. The D-Sub connectors are fine for the ECU, as unless there is no other option it should be mounted in a protected location. Forget water or salt spray and think impact, shock, and other violence that will not bother the wiring but can and will give your ECU terminal issues. Now, I *personally* do not like D-subs, but have used them and in general do not feel the need to replace them out of hand.
My personal favorite is the Deutsch DT or DTM connectors and I favor the gold or nickel pins, crimped not soldered. The Tyco/AMP canon plugs using the DT/DTM style pins work very well, and are available in a lighter all-plastic version that locks securely. Not cheap, but then the point is to only need to buy them once.

Denso/Bosch/Metripac etc. connectors as used by most manufacturers are also fine for the most part.
Weatherpac connectors suck and just simply should never be used if it can be avoided. Same goes for the generic Molex connectors. Molex does make some nicer connectors, but for the most part they will cause more issues than they solve over time.

If we need a truly sealed connection then a DT or DTM connector with adhesive lined heat shrink over the wire side is hard to beat. Fortunately, we do not often need such things.

A side note on EFI in general- I get that for a lot of people EFI is a PITA mystery box that they have only the most rudimentary understanding of, and no desire to learn. That is totally OK, but lets try and keep objections or potential problems to the realm of reality.

 

[rv6ejguy]

These last few posts open another subject well worth discussing. I build a lot of stuff that has to tolerate road spray and once a lot of marine wiring and there is simply NO FRIGGING WAY any kind of open connector (i.e. not hermetically sealed) can be tolerated. I haven't built an aircraft harness for decades, but as I recall, NONE ever used a sealed connection arrangement. In both OTR and marine environments it is often the open end of a multi-strand wire that has tremendously effective capillary effect that draws moisture with attendant corrosive ions (chlorides especially) that result in corrosion and stress corrosion cracking that causes the wire to ultimately fail. Why does this not seem to be considered in aviation installations ahead of the firewall????

It is, all FWF connections are all sealed type.

You folks who crap on what we use and have had ZERO failures on with VASTLY more experience than all of you combined, need to give your head a shake. The proof is in the pudding. We're not the leading supplier of programmable Experimental aviation ECUs in the world because stuff is falling out of the sky as a result of wiring/ connector failures. I've personally built many THOUSANDS of wiring harnesses, many hundreds of those for aviation. I'm building several more today in fact, 2 for a Lancair IV-P. As I said, 800,000 flight hours on nearly 2200 aviation ECUs to date. 8000 more automotive ECUs.

Any of you commenting designed, built and flying your own ECUs and wiring? Any of you doing it commercially 6 days a week, 52 weeks a year? You think Garmin are a bunch of idiots for using D subs on their certified avionics?

 

[stanislavz]

It is, all FWF connections are all sealed type.

Just lets stop here i think. Wiring is always wiring - just too many pieces to get wrong. I am not agains efi any form. Just found mechanical pump in diesel being available.

Back to diesels!

Packard DR-980 - Wikipedia

en.wikipedia.org

240 Hp/ 227 kg dry weight.

 

[dwalker]

Just lets stop here i think. Wiring is always wiring - just too many pieces to get wrong. I am not agains efi any form. Just found mechanical pump in diesel being available.

Back to diesels!

Packard DR-980 - Wikipedia

en.wikipedia.org

240 Hp/ 227 kg dry weight.

So, I have worked with Mechanical Fuel Injection on gas engines in the Porsches (Bosch) and the Alfas (Spica), and a fair number of diesels.
Mechanical injection is simply not going to be efficient enough at altitude and once you get into common rail/electronic injection the weight and complexity goes up. I understand the lure and legend of the "P-Pump" and wanting to make it work, I mean who doesnt want a light, reliable, powerful and fuel efficient motor, it just is very unlikely to happen using a mechanically injected diesel.
Obviously once you get into common rail injection not only do you have the increased weight but you have wiring, ecu, etc. I am sure there is a really viable diesel or CI engine out there in the future for light aircraft, but I have to think the cost is going to be very, very high.

 

[rv6ejguy]

Just lets stop here i think. Wiring is always wiring - just too many pieces to get wrong. I am not agains efi any form. Just found mechanical pump in diesel being available.

Back to diesels!

Packard DR-980 - Wikipedia

en.wikipedia.org

240 Hp/ 227 kg dry weight.

Outside of the Conti CD-200 series engines, just about all certified aero diesels use FADECs today with all that dangerous wiring... Are you speaking about facts or feelings here? The Austro AE series are among the most reliable piston aero engines made today.

 

[imacfii]

The K-cycle engine was inspired but let's not forget that fixed block, rotating swashplate/wobble plate engines have a very long history and have been certified for aviation use (Dynacam). What Kristiansen was developing was really an extension of the original Atkinson Differential engine (1881) but arranged as a swash plate engine. Noteworthy is that it used opposed pistons (as have many very successful engines over the last 140 years) but differentiated displacement for a 4 stroke cycle. This IMHO was one thing Herrmann missed (in 1941) designing the Dynacam.

I have had some exposure to some ongoing R&D using swashplate barrel designs, and from what I can gather from the very real success of the Dynacam but much ongoing difficulty with other R&D projects it is getting the cam and rollers to work reliably and durably is where there is still more work to be done. Yes, my interest in such obscure things has a lot to do with being a ME student at UofM a very long time ago.

The "5 cycle" engines are really just a normal 4 cycle with an integral positive displacement power recovery stage - which limits aviation potential as it would be severely altitude challenged. Much simpler to just add a turbocharger and let it cover the need for more air with altitude.

Which brings us to why CI vs. SI. No aspirated charge means no such detonation limits that drive current fossil fuel/fossilized tech engines to burn stunning amounts of avagas to try to overcome fixed antique mag timing issues and lack of cooling air at altitude. Much simpler to just throw the mags in the dust bin and use compression to ignite. Again: one of the reasons that pretty much ALL newly certified piston engines for some time are diesels, not gassers.

I looked at Dynacam a few years ago, I think I actually lost money on them, but it seemed to me that as piston #1 was at TDC, so was piston #7 which resulted in a intermediate rocking couple, which destroyed the mid section. Why not a 5 cyl or 7 cyl which would resolve that issue. Any thoughts? I seem to recollect Ford Australia did some development but shelved it when they had ises balancing the swash plate.

 

[lelievre12]

In defense of CI vs SI: EVERY diesel now certified has a dismal power/weight ratio. Once someone comes up with Rotax power density (that is IMHO the benchmark) things will change.

80HP Rotax 912 power to weight = 1.7lb/hp
100HP Rotax 912IS power to weight = 1.5lb/hp
150HP O-320 Power to weight = 1.8hp/lb
155HP Technify TDi power to weight = 1.9lb/hp

TDi is heavier but not "dismal". It burns less fuel and is turbocharged so offers better power to weight than all the above at >5000ft. Of course SI turbo is better again if one is just looking at power to weight. However I expect the market is concerned with many matters aside from power to weight such as cost, reliability, complexity and nationwide service support.

 

[stanislavz]

80HP Rotax 912 power to weight = 1.7lb/hp
100HP Rotax 912IS power to weight = 1.5lb/hp
150HP O-320 Power to weight = 1.8hp/lb
155HP Technify TDi power to weight = 1.9lb/hp

TDi is heavier but not "dismal". It burns less fuel and is turbocharged so offers better power to weight than all the above at >5000ft. Of course SI turbo is better again if one is just looking at power to weight. However I expect the market is concerned with many matters aside from power to weight such as cost, reliability, complexity and nationwide service support.

This is dry weight or firewall forward ?

 

[PMD]

it is, all FWF connections are all sealed type

Have not seen one of your installations, but not at all surprised you do this. However: I was referring to genav overall. I do not recall anything actually hermetically sealed from days long gone by (except of course mag leads). To me, one of THE most important components of any connector is some kind of strain relief to move the load off of stress riser point at the crimp. Modern heat shrink terminals with adhesive lining do that extremely well, but are there any such things today that are AN standard and PMA supplied? My last aviation harness was around 40 years ago, so no such thing around then.

 

[lelievre12]

That Dynacam is interesting.

Another is the Brueckmueller variable stroke swash plate engine.



US4174684A - Variable stroke internal combustion engine - Google Patents

Like the Dynacam it has a tiny 'turbine' like profile but more importantly offers variable compression so cruise BSFC could be optimised.



I dont know much about how 'successful' it was however variable compression in aero engines is tantalizing.

 

[PMD]

I don't know much about how 'successful' it was however variable compression in aero engines is tantalizing.

MUCH simpler to just use a turbocharger to vary the cylinder pressure as needed instead of a bunch of monkey motion hardware to vary the stroke. Then of course using the Diesel cycle instead of Otto allows you to use very high cylinder pressures.

 

[PMD]

The engine that embraces almost all of the "good things" that one can do, and still very simple is the Fair Diesel camplate engine: The Engine | covaxe It has more in common with Dynacam in that it uses cam plate and roller followers. Being a 2 stroke diesel cycle, the load side of the cam is always in compression with the tiny offside cam and roller just there to get things started. I know their biggest problem has been camfollower bearings, but Dynacam in its original certified format seemed to be able to get that right a long time ago.

 

[stanislavz]

Being a 2 stroke diesel cycle, the load side of the cam is always in compression with the tiny offside cam and roller just there to get things started.

Yes. But you could make all this with jumo like opposite piston engine in lighter way.. blower is cheaper and lighter than piston to compress air for 2 stroke one.

 

[rv6ejguy]

The engine that embraces almost all of the "good things" that one can do, and still very simple is the Fair Diesel camplate engine: The Engine | covaxe It has more in common with Dynacam in that it uses cam plate and roller followers. Being a 2 stroke diesel cycle, the load side of the cam is always in compression with the tiny offside cam and roller just there to get things started. I know their biggest problem has been camfollower bearings, but Dynacam in its original certified format seemed to be able to get that right a long time ago.

Interesting design however, I wonder what they are trying to say here: "Thus full combustion pressure is not generated at Top Dead Centre" referring to traditional piston engines as if this is some sort of bad thing. PCP occuring at TDC would do ZERO work and simply try to compress the connecting rod. PCP has to occur AFTER TDC to impart useful work.

It's statements like these on new engine design websites that make me question if they even understand fundamental engine operation as they make a case for their advantages over existing designs.

 

[stanislavz]

Thus full combustion pressure is not generated at Top Dead Centre

Continuous combustion or combustion at constant volume... Same in M4+2 or other six stroke engines or split combustion engine.