Ever ridden a bike with flat slide carbs? If you want non linear and sometimes inverse throttle response, you'll get it!
I've got a T140 Bonneville with Amal Mk2s. It seems OK.
Vince - virtually all the digital EFI bikes from 1988 until the mid-2000s had the same issue you experienced with jerky throttle response. Some of the Japanese manufacturers went to similar double throttle solution used on your Honda single, but for a different reason. The smaller of the two throttles per cylinder is used to soften the jerk from small throttle openings and then the large throttle opens at higher mass flow. A four cylinder bike has eight throttles. For an existing EFI engine, one solution I and others have used successfully is to set the off-idle mixture (very small alpha at all rpm) a little over-rich. If you do it carefully, it doesn't impact fuel economy too much and helps.
Going down those kinds of complex development roads with aircraft engines, as I believe Thielert and others have found, is like opening a can of worms. The end result is huge development budgets spread over very limited numbers of units. And after that, the customer has accept (or being coerced into accepting) that he's buying hardware with extra weight, planned obsolescence, and a finite service life after which it "isn't supported". Given that the big budgets for aircraft piston engine development faded when commercial air transport shifted to jets, and production volumes are not going to rise to big numbers, and the economics for light aircraft dictate multiple overhauls, I don't see it happening.
"Sophisticated" Porsche engined Mooneys needed a bigger plane to carry the engine, had significantly worse performance and if I'm not mistaken all of them ended up reengined with Lycomings. Some technical director at Porsche was quoted as saying "maybe aircraft engines are just industrial engines after all, and we're not interested in that kind of business". He undoubtedly meant it as a snide Germanic-style sideways insult, but he was actually correct that they did not understand or develop the best technical solution.
I had a Honda VFR that was EFI equipped. It had a jerky throttle response
Possibly the worlds worst example of how to do fuel injection on a bike. The VFR's were reknowned for bad mapping, something Honda never really sorted out (although I think they did eventually). There are many better examples.
Ever ridden a bike with flat slide carbs? If you want non linear and sometimes inverse throttle response, you'll get it!
The only real place where fuel injection has been a struggle is in two strokes. Arguably this is because by the time injection came around, the motorcycle manufacturers had thrown the towel in (arguably very prematurely) on two strokes. However, Evinrude use it on their two stroke E-TEC outboard motors with immense success. They are more fuel efficient, cleaner, more performant, require less maintenance and are more reliable than their four stroke counterparts. In some ways, boat engines have a very similar mode of use as aircraft (long periods at fixed power), so I'd be amazed if there weren't some transferable technology. However, never let this detract from never mentioning the words "2" and "stroke" in the same sentence to a pilot - you think FADEC is controversial? 
My experience with motorcycle EFI (1500-2000 people use my work now) is that its very hard to produce results that better carbs, including fidelity of control.
I had a Honda VFR that was EFI equipped. It had a jerky throttle response just where you need it to be smooth on a bike; rolling on power exiting a bend. When I upgraded it for a Blackbird, I got one of the last carbureted ones - smooth as a babie's whatsit!
The three bikes in my garage right now all have carbs! One's a big single, and it has 2 carbs! It has immuculate throttle response.
My experience with motorcycle EFI (1500-2000 people use my work now) is that its very hard to produce results that better carbs, including fidelity of control. The reasons its used in that application are (1) more power without affecting idle quality, due to high pressure delivery and no venturi in the intake, and (2) the ability to tune a narrow digitially defined band of the operational range to pass emission tests without affecting the broader range of operation.
In the motorcycle application, it might surprise you to know that when well equipped, high volume manufacturers switched over to EFI, economy typically dropped 10-20%. That's not to say it can't be done well, but EFI is not a panacea for engine control.
The biggest potential advantage of EFI for aircraft engines is automatic altitude compensation. Certainly for me that is inconsequential compared to the complexity of owning and maintaining a certified EFI system over decades.
these are still treated with derision by a significant proportion of the GA community
I don't think they are treated with derision. What I see happening is that, on the old UK pilot forums, there is a very much "them and us" positioning, and the "Rotax crowd" (I use that term loosely) carries along a small number of individuals who are very vocal in pushing their position, often adding inflammatory stuff e.g. the demise of Avgas and the closure of GA airfields would be just fine (because "I" have my own farm strip, with mogas), etc etc etc.
The way I see it is that there is room for everybody.
Personally, I would never trade my TB20 for a smaller machine because I need that mission capability for the trips that I do (mostly with Justine, on holidays, and carrying a load of stuff) which make flying really worthwhile for me.
the average old aero engine spends more time, a very long way from its optimum efficiency, that it does operating at it's optimum efficiency.
It probably does. A 30 minute burger run is just that.
Using modern engine management removes all doubt - you get an engine that operates at it's peak efficiency all the time.
It's an interesting Q but I would be suprised if you could run say an IO540-C4 (250HP) at 250HP in all phases of flight, at stochiometric, with any kind of engine management. You would need water cooling, and a turbo to substantially pump up the MP. Then you are looking at a different engine. Would it be more reliable? I am certain not.
Does anyone think that having automated engine management has potential greater benefits for the longevity of an engine?
It must have, but will it be delivered in a reliable form and at a price people will pay?
How much innovation has there been in biro manufacture over the past 30 years?
Good point! Actually I was thinking as I submitted the last post several examples in the natural world where evolution hasn't taken place for a long time. Crocodiles haven't changed at all - in evolutionary terms they clearly have little need for improvement. Dinosaurs also. But what these things have in common is that whilst they are around for a long time, when they go, they go very quickly. Evolution (as opposed to revolution) is a good thing.
The original point was made in regard to CofA aircraft. Permit/experimental aircraft (and their Jabiru/Rotax etc powerplants) have come along leaps and bounds, and are a perfect example of what I argue hasn't happened in the CofA world. And like most innovations, these are still treated with derision by a significant proportion of the GA community, even though (notwithstanding their limited mission profile) many quietly agree they are vastly superior to their older CofA counterparts.
I'd be inclined to agree that if you can run a old aero engine in a steady state for a long period, you may well get great efficiency. However, I would suspect that as an ensemble, the average old aero engine spends more time, a very long way from its optimum efficiency, that it does operating at it's optimum efficiency. You can argue that's operator error, but you could argue that's just reality. Using modern engine management removes all doubt - you get an engine that operates at it's peak efficiency all the time.
Quick question: Does anyone think that having automated engine management has potential greater benefits for the longevity of an engine? Poor engine management can certainly reduce an engine's life expectancy. And presumably, the closer you run an old aero engine to it's optimal efficiency, the finer the margins for error before you unknowing start causing damage.
Hard to prove unless you conducted a controlled scientific experiment, but it would be interesting to know what effect it might have for an "average" aero engine in the hands of an "average" pilot. As with fuel efficiency, this is really the performance that matters.
I think Rotax was motivated towards EFI by the unavailability of better carbs than the CV Bings for their specific engine. The Bings could in principle be OK but they were not designed for the same level of reliability as a simpler aircraft carb: a vacuum controlled moving piston, a thin rolling rubber diaphragm to seal it, hose clamp mounting and spring return cables are fine for a 1972-96 BMW motorcycle (the main application for these carbs), but aren't optimum for aircraft. Plus the Rotax engine by design needs two carbs versus one, and they must be synchronized, which is a lot of 'monkey motion'. The constant velocity carb does provide a useful measure of auto lean (as do some aircraft carbs BTW, see http://en.wikipedia.org/wiki/Bendix-Strombergpressurecarburetor) and I can imagine Rotax looking at their options, seeing nothing else available in side draft aircraft carbs with or without auto lean, and thinking that EFI was the only way forward despite it having no advantage in power plus complexity and component obsolescence issues.
I like the simplest possible setup myself, and admire the original designers of the flat aircraft engines in coming up with a very simple carburation solution, positioning a purpose-designed single carb under the engine to minimize leakage issues and fire potential.
Apparently its effectiveness depends on the altitude/atmosphere. The carb creates the air/fuel mixture and that needs to be exactly at the optimal stoichiometric ratio to reach book SFC.
I don't think this can be done accurately enough over the full range of altitude. One needs to be measuring mass flow to do it properly - or else do what they do in cars which is to use an oxygen sensor. But oxygen sensors on cars pack up fairly frequently...
I recall that Conti used altitude compensation in their fuel servos. There is a nasty failure mode in that whereby the pressure sending diaphragm bursts at altitude. The result is an engine stoppage - until you realise what's happened and lean massively.
Well, Rotax is basically without competition so why innovate?
Same goes for most things. Improvements are at most very marginal.
