A question about aero diesel engine in WWII

[Trilisser]

It is an absolute fact that the primary reason for electronically controlled injection for diesels has been political, i.e. emission regulations. One has to be blind not to see the link between certain technical features and emission regs. Do you really believe that engine makers would intentionally adopt features that increase sfc of their engines if they could avoid it? E.g. diesels of current logging machinery have poorer efficiency than those of say 7-8 yrs ago due to stricter emission regulations (=politics). Yet these newer engines have many more "modern" features.

And yes, the problem of adequately accurate mechanical injection was solved prior to WW2. One good example is the DB 605 whose injection pump follows 99.999 % diesel practice except for provision of additional lubrication to compensate for the different fuel.

 

[Dogwalker]

Calling a fact "absolute" doesn't make it more true.

To have some answer, often it's sufficient to mind how stuff works.
What's a prechamber diesel? Why was it invented?
It's obvious that diesel engines didn't born that way. They were single chamber engines at first.
The problem was that this kind of diesels worked well in big static application, or marine engines. When someone tried to do automotive diesels, the stuff proven to be too fragile.
Why?
Cause in a direct injection diesel, injection pressure and good timing is all. The more is the pressure and the more precise is the injection time, the more efficient, less mechanically stressed, and lighter is the engine.
But, back in the '20s, high pression pumps and injectors are heavy and costly things. Not compatible with automotive needs.
So, someone (yes, I know who) invented the prechamber diesel engine.
A prechamber injection system, is practically a two stage injection system.
First, when the piston is rising, and the pressure in the cylinder is realtively low, a low pressure injector injects the fuel in the prechamber. Then, since the prechamber wall is hot enough, the fuel begin to burn, and the pressure spike is enough to project the majority of the fuel in the main chamber, and obtaining a good air/fuel mix, when the pressure in the main chamber is enough to make it burn. First a mechanical injection, then a thermal one.
But there is nothing as a free lunch. In prechamber diesels the head has to be heavy. The larger surface area of the combustion chamber reduces the thermal efficiency. The ignition delay, from the mechanical injection to the combustion of the main part of the fuel, is bigger, so a big part of the fuel burn when the piston is lowering, thus lowering the eficiency of the engine (an engine can have a 22/1 geometrical compression ratio, but if the fuel burn when the piston is middle way down, the gasses are decompressed only two times, and therefore work too little). Moreover, since the engine is designed taking into account the combustion delay, a combustion delay had to be. If the conditions into the cylinder are too favorable to combustion, an so it's too rapid, for example because a turbocharger has heated the room too much, the result is destructive (in a direct injection engine the problem is less acute, since the fuel is injected at the time when you want it to burn).

This is the kind of drawbacks which manufacturers accepted until the '80s, according to you not to use solutions that were already there in the '30s, after which politicians snapped their fingers, and suddenly all the manufacturers have adapted to their desires.
Ok....

However, the problems that a direct injection gasoline engine has to solve, are different. In a gasoline engine, the injection timing, and consequently the pressure, is less important (infact, low pressure indirect injection engines work very well). In a gasoline engine you don't have to inject fuel at the exact time you want it to burn, since you send it in an environment such to make it spontaneously ignite. In a gasoline engine, at any time you inject the fuel, it will burn only with the spark.
But the DB601/605 example is intersting anyway. Since you talk as if developing it's direct injection system had been a walk in the park for DB technicians...

 

[vikingBerserker]

I will say one thing, I've learned a lot about diesels from this thread.

Great info gents.

 

[swampyankee]

I think that the first direct injection (means no prechamber) automobile diesel was produced in the late 1970s or early 1980s; before that all small diesels used prechambers, which both reduce efficiency and demand increased compression ratio. They also permit lower injector pressures, so comparing the injection pressure a typical automotive diesel of the 1980s with a large diesel is invalid: direct injection needs higher injection pressure for sufficient fuel atomization.

As for electronic controls? My experience with gas turbines is that electronic controls -- FADECs -- are much easier to develop and optimize than are mechanical or hydraulic control systems. While fuel control in gas turbines is, in many ways, much easier than that for diesels (gas turbines do not need precisely timed high-pressure injections of fuel), the mechanical alternative still requires precisely machined parts, and in large diesel engines, these are big precisely machined parts.

 

[Trilisser]

Dogwalker, you forget that diesels employed blast air injection for quite a while before solid fuel injection came to be. As for injection pressure, I have driven a 1980s Fiat Ritmo diesel that was of prechamber design, yet the injection pressure was some 150 bars even though some sources claim that as low as 70 or so bars is possible.

As for why wasn't direct injection diesels adopted in automobiles in wide scale until quite late: cost. When a business can get away with lower cost alternative it invariably selects that. With tighter and tighter emission regs set by politicians they could no longer get away with the cheaper alternative.

 

[Denniss]

According to the german wikipedia the first direct-injection Diesels of the new age in passenger cars were:
Fiat Croma 2000 Turbo D i.d. ( 1986 )
Austin Montego MDi (1988 )
Audi 100 2.5 TDI ( 1989 )

Some earlier implementations in small trucks like Fiat Ducato may be possible.

But it was Audi/Volkswagen with Bosch who continuously improved the system for efficiency and mass-production which led to a huge commercial success.

 

[FLYBOYJ]

As far as diesels in today's aviation world...

China buys Thielert diesel engine company out of bankruptcy - Aircraft Owners and Pilots Association

With EPA regulation and FAA/JAA certification processes there are limits on what could be done diesel recips today, but IMO I think they are going to be the wave of the future. It will eliminate avgas and establish one fuel source for all aircraft engines.

 

[Dogwalker]

Dogwalker, you forget that diesels employed blast air injection

"Forget"? Blast air injection diesel was the tecnology that didn't work well (even for economic reasons obviously) for small engines, and led to the predominance of the prechamber for about 60 years.

As for injection pressure, I have driven a 1980s Fiat Ritmo diesel that was of prechamber design, yet the injection pressure was some 150 bars even though some sources claim that as low as 70 or so bars is possible.

And so? It was a prechamber diesel, it worked with low pressure injection.

As for why wasn't direct injection diesels adopted in automobiles in wide scale until quite late: cost.

Right. Apply a sufficient amount of money, and problems will be solved. Today as in the '30s. But that not means that there are/were no problems to adress.

 

[swampyankee]

Even about 30 years ago, with NASA's GAP program, the spark-ignition aircraft engine seems to have been considered to be to the point where only quite minor performance improvement was possible (there's been a lot of effort into improving engines for automobiles; the best of them do not exactly overwhelm the O-200 in any important measure of performance, like sfc or power/weight): GAP engines were a highly supercharged two-stroke diesel and a gas turbine. Williams Research tried to get a commercial derivative of the gas turbine, the EJ22 turbofan, into service but didn't. I don't think Continental even tried with the diesel.

 

[Trilisser]

I think one major problem with many small aircraft diesels today is that they attempt too much with odd solutions like that Zoche design. Another question is that where is a modern (=not decades old) small 250-300 hp class simple low-cost turboprop? I don't refer to Allison 250 and likes.

EDit: There is the RR300, but its sfc figures are not too attractive.

 

[swampyankee]

I think one major problem with many small aircraft diesels today is that they attempt too much with odd solutions like that Zoche design. Another question is that where is a modern (=not decades old) small 250-300 hp class simple low-cost turboprop? I don't refer to Allison 250 and likes.

EDit: There is the RR300, but its sfc figures are not too attractive.

The market for that size engine is very price sensitive, and even small engines cost a lot to develop. Small efficient engines cost even more. There are several reasons: manufacturing tolerances do not scale, so the tip clearances on a 100 mm diameter compressor are, proportionately, much larger than a 500 mm compressor, smaller blades are more fragile, so FOD and erosion are more serious issues, which is a major reason why small engines usually have centrifugal or axial/centrifugal, vs all-axial, compressors, even though axials tend to be more efficient (although the advantage to axials gets smaller as size decreases: the efficiency of axials falls off at a rate roughly proportional to the fifth root of the Reynolds' Number, that of centrifugals, roughly, by the tenth root), it gets harder to design and fabricate cooled blades (even with superalloys, the turbine inlet temperature of some large engines exceeds the material melting point), and, since stage pressure ratios depend on tip Mach number, bearings get very difficult (iirc, most modern engines run with their compressor tips moving at about 500 m/s, regardless of size).

So, small engines run at lower pressure ratios (a single centrifugal stage can generate a pressure ratio of over 10:1, but these may have problems when operating off-design, and may have quite small surge margins), and lower turbine entry temperatures, both of which reduce achievable efficiency. The compressor and turbine are going to be less efficient, which reduces the efficiency further. The square-cube law will mean that heat losses through the combustor and turbine housings to the outside will be proportionately greater than a large engine, etc.

So, no high-efficiency small turbines because of cost and fluid mechanics. Money is just a social convention, but fluid mechanics isn't.

 

[FLYBOYJ]

Don't forget the certification process. The Feds in many cases make profitable engine development very difficult.

eCFR — Code of Federal Regulations

 

[swampyankee]

Don't forget the certification process. The Feds in many cases make profitable engine development very difficult.

eCFR — Code of Federal Regulations

Well, there are ways of saving money. GE skipped the bird shots on their CF6, certifying it by similarity with the TF39s, which was a massive cheat (as in, imho, fraud), and ended up with a loaded DC-10 losing power on take-off at LGA. This ended up with all the DC-10s with GE engines being grounded. The feeling at Lycoming (I worked on the ALF-502; my signature is on some of the certification paperwork) was that we'd have lost our type certificate for that sort of crap (most of the people thought Pratt would have, too). Yes, the certification costs aren't cheap, but the regs are there because of past manufacturer screw-ups, not just because bureaucrats are being obstructive.

 

[FLYBOYJ]

Yes, the certification costs aren't cheap, but the regs are there because of past manufacturer screw-ups, not just because bureaucrats are being obstructive.

While regulation is required, there is no doubt that because of the way its interpreted (and the way its enforced) it limits development and in my 35 years in this business I've seen dozens of bureaucrats being extremely obstructive as there are many people in MIDOs that don't know what they are doing and know little about new technologies; that's why you have seen little in recip engine development during the past 40 years.

The only reason why we're even seeing recip diesels being certified is because the feds know that leaded avgas will soon be a thing of the past and they have no choice but to get off their lazy butts and work with manufacturers.

The FAA - the compassion of the IRS combined with the efficiency of the US Post Office...

 

[Shortround6]

The FAA - the compassion of the IRS combined with the efficiency of the US Post Office...

How dare you slander both the IRS and the US Post Office.....

 

[Trilisser]

I wonder who dares to dry a small turboprop with a recuperator...

 

[FLYBOYJ]

Well, there are ways of saving money. GE skipped the bird shots on their CF6, certifying it by similarity with the TF39s, which was a massive cheat (as in, imho, fraud)

Just an after thought on this - who from the FAA or what DAR/DIMR signed the final certification paperwork? I've seen certifications like this happen and off the cuff I don't think there's a whole lot of difference between the two (I do know the CF6 was developed from the TF39 and they are very similar). If the Feds allowed this to happen its on them and my previous point well proven. If GE did this without FAA approval, the ones responsible for this should be in jail - just wondering!

 

[swampyankee]

Just an after thought on this - who from the FAA or what DAR/DIMR signed the final certification paperwork? I've seen certifications like this happen and off the cuff I don't think there's a whole lot of difference between the two (I do know the CF6 was developed from the TF39 and they are very similar). If the Feds allowed this to happen its on them and my previous point well proven. If GE did this without FAA approval, the ones responsible for this should be in jail - just wondering!

The TF39 has a completely different fan assembly than the CF6: the TF39 has a two-stage fan; the CF6 has a single stage fan. Trying to get it certified by similarity was a bald-faced lie; as I said, the feeling at Lycoming was that if we or Pratt (a lot of people I worked with there were ex-Pratt) had tried the same thing, we'd have lost the type certificate, not just have the engine grounded.

In general, the FAA seems to do remarkably little vetting of manufacturer claims, until something goes wrong. We had our designated engineering rep, but he was a company, not FAA, employee. I don't remember hearing of an FAA employee ever witnessing a test. Certainly, none did on my watch.

 

[FLYBOYJ]

The TF39 has a completely different fan assembly than the CF6: the TF39 has a two-stage fan; the CF6 has a single stage fan. Trying to get it certified by similarity was a bald-faced lie; as I said, the feeling at Lycoming was that if we or Pratt (a lot of people I worked with there were ex-Pratt) had tried the same thing, we'd have lost the type certificate, not just have the engine grounded.

In general, the FAA seems to do remarkably little vetting of manufacturer claims, until something goes wrong. We had our designated engineering rep, but he was a company, not FAA, employee. I don't remember hearing of an FAA employee ever witnessing a test. Certainly, none did on my watch.

Well apparently someone sold this to the Feds when they issued the TC, lie or not. It shows the incompetency at MIDOs. I've seen it flow the other way as well where a simple field approval wouldn't be approved because of a typo on the TC data sheet. Why fines and criminal charges didn't fly on this one beats me, probably because someone at the agency bought in on it and MIDO would get a bunch of egg on its face, the best thing to do is to just ground the fleet and make the mfg do a "do-over," the easiest way to make it go away.

I do know that there seemed to be a lot of CF6 failures due to hydrogen embrittlement, many ADs written against CF6s (from what I remember from my short DC-10 days)

 

[swampyankee]

Well apparently someone sold this to the Feds when they issued the TC, lie or not. It shows the incompetency at MIDOs. I've seen it flow the other way as well where a simple field approval wouldn't be approved because of a typo on the TC data sheet. Why fines and criminal charges didn't fly on this one beats me, probably because someone at the agency bought in on it and MIDO would get a bunch of egg on its face, the best thing to do is to just ground the fleet and make the mfg do a "do-over," the easiest way to make it go away.

I do know that there seemed to be a lot of CF6 failures due to hydrogen embrittlement, many ADs written against CF6s (from what I remember from my short DC-10 days)

At the time (late 1970s to late 1980s), we could tell if a GE-engined commercial aircraft crashed: the engine maker was never mentioned on the news.