Discussion - engine and fuel system controls

[Genghis the Engineer]

I have no particular reason for raising this subject, it's just one which annoys me from time to time.

Once upon a time I was briefly allowed to earn my living in the back of a fast jet. Taking a typical aircraft, the Hawk, one had

- A throttle
- A start button
- A fuel cut-off

And basically that was it.



Now at the other end of the scale I spent today tinkering with my Raven-X, which is a Rotax 2-stroke engined flexwing. This has:-

- A throttle
- A choke, for starting
- A fuel cut-off / tank selection lever
- An ignition switch
- An electric carb-heat switch

A similar but more modern aircraft, say a Quantum 15-912 would have

- A throttle
- A choke, for starting
- A fuel cut-off lever (only one tank)
- Two ignition switches
- A starter button


Now on the other hand, if the local runway emerges from the floods, I also get occasionally to fly my 1/17th of a PA28, this has:-

- A throttle
- A primer
- A mixture lever
- A throttle friction lever
- A carb heat control
- A combined ignition / starter switch
- A fuel tank select / cut-off device.

If it had a VP prop, add in an RPM governer control. That is 8 separate controls, compared to 5 on a modern microlight, and two (throttle and ignition) on a modern car. So what could reasonably be automated:-

- Carb heat does not need to be as used on the standard Lycontinental installation. On various microlight installations I've seen coolant based carb body heating, coolant based air intake heating, electrical carb body heating - all of which can be left permanently on without any significant power loss and only the electrical system needing any kind of function indicator - the others will work so long as you have coolant.

- Choke / prime. This has been automated on cars for 20 years, why are we still mucking about with them on aeroplanes, and flooding our engines when we get them wrong.

- Fuel tank selection / cut-off. Anybody flown a 4-tank Piper - it's an ergonomic nightmare and totally unecessary. If one must have a multi-tank system the technology to automate balancing and use has been with us for 40+ years and is totally reliable - in most cases it only consists of large bore hose and some baffles.

- Fuel cut-off. Why? Cars simply have a cut-off solenoid connected to the ignition switch. If one wants to make sure there is no fuel in the lines at shut-off (so as to isolate an engine fire for example) a simple timing device would suffice to solve this.

- Throttle friction. Really. Is it really beyond a modern design engineer to make a lever that stays where it's put in a vibrating environment without the forces being excessive? For that matter, was it beyond them 30 years ago?

- Mixture. Cars manage without, using very simple and inexpensive computers programmed to give optimal running based upon measured density altitude and EGT to ensure that isn't exceeded. Link this to the throttle so that if more than 75% power is selected it defaults to fully rich and we've automatically got rid of another control in the cockpit to muck about with. Ensure a failure mode of fully rich (plus a failure indicator light) and the 10^5 reliability issues can be swept away.

- RPM control - same applies surely. Fine is only needed above 75% power, the rest of the time the system should be defaulting to best-for-cruise for a given power and density altitude, so lets combine this with the mixture control in a simple ECU that defaults to fully fine in the event of a failure.

- Mag switches, this is an area I'm slightly unsure, but... The ignition circuit is only functioning for a fraction of each cycle, during running it should be possible to continually test them for system capacitance which should fall within specific values. One circuit out of limits, an automatic warning light. But I may be wrong about this.

So I present you with Genghis' 21st century engine and fuel system controls:-

- A throttle
- A run / stop control.
- And if absolutely necessary, a 1-2-both mag circuit switch for checks.

Eliminating 4-6 cockpit controls that clutter the place up, can be wrongly operated and distract pilots from the serious business of enjoying the view.

Or am I missing something?

G

[BOAC]

Valid, indeed, Ghengis, but...........

what are you going to leave QFIs to nag about?:)

[Tinstaafl]

And by doing most of these you also have to include an alternate independant electrical system. I've had a complete electrical failure in an a/c. No drama at all. The engine didn't miss a beat. It couldn't because it wasn't involved.

I've had a fuel solenoid go tits up in my car but not in an a/c.

Certainly there are some good points in what you've said. Fuel storage particularly. There's no difficult reason why tanks can't be directly connected, effectively making one or two large ones.

Countering that is when fuel is used as ballast for balance. It's damned convenient to be able to isolate an advantageous compartment.

Ditto throttle/prop. interconnects. They would simplify things. Until you hop in a piston twin when having fine manual control is needed to sync. the props.

Mixture could be fairly easily automated I think. Perhaps with a barometric/temp. device?

Personally I like having a reasonable degree of manual control. I regret the loss of my car's manual choke & prefer manual gearboxes.

Perhaps my arguments reflect more my personality than anything logical!

[Fat Dog]

Well just fly a B737 then...:p

[FlyingForFun]

I think there are two issues here:

First of all, pilots need to have a huge degree of control when things go wrong. Take the combined fuel tanks for example - there's no reason they couldn't be connected together with a pipe and some baffles, until you get airborne and discover that you've got a bad batch of fuel in the left tank.... suddenly you wish they weren't connected together any more.

But the most important thing, I'm sure, is pilots' egos - the more knobs we have to play with in the cockpit, the more important we feel! :D I know I miss the mixture and carb-heat controls on my Rotax - it's just boring being able to control the engine with just one lever! (Oh, plus the mag switches, and fuel selector, and choke, and prop pitch!)

FFF
-----------

[411A]

Aside from the fuel tank/selector, which oddly enough, on some light aircraft, approaches the complexity of the 10 tank DC-6B I flew many years ago...
I believe both Continental and Lycoming have developed a FADEC type of control for some of their engines. TCM (I think) recently had theirs certified.

However, it's gotta be expensive, I would think.

[JEP]

Last year I saw a PA28 with a Thielert Diesel engine.
It was fitted with a VP propeller and a turbo charger. The engine controls was:

1 lever (electronics do the injection timing and pitch control)
1 starter key (like any car)
1 fuel selector (the standard PS28).

The engine has been approved by JAA for PA28 and C172 - so we are getting there. Furthermore the fuel consumption is reduced by apprx. 40% and it runs on dieselfuel or JET-A1.

[Blacksheep]

So there you go Ghengis; us engineers like to fix things and make the world a better place to live in, but folks actually prefer the horse and cart, candles, open coal fires and sailing ships. They don't want all this new-fangled jiggery-pokery. Leave the pilots to twiddle their knobs and fine tune their propellors by hand. Meanwhile us engineers can live in warmth and comfort and enjoy the view.

Where would we be without engineers? In caves, dressed in animal hides thats where. And a much better place the world would be too, no doubt. :rolleyes:

Keep on inventing Isambard ;)

**************************
Through difficulties to the cinema

[LOMCEVAK]

Ghengis,

The more the automation, the greater the COST, the greater the chance of something going wrong, and flexibility of operation is reduced. However, workload and the probability of human error SHOULD be reduced. Just a few thoughts:

1. If you have the facility to select magnetos individually, you can deselect a faulty one and perhaps restore smooth running to an engine that is running roughly due to a mag problem.

2. If an engine has a tendency to overheat at high power settings and low speed, you can reduce CHT by running the engine with a richer mixture. Also, you may be able to squeeze a bit more power out of an engine which is losing power for certain failures by selecting fully rich mixture.

3. If you link MAP and RPM such that you only need one power lever, you will always have low RPM for low MAP. This will increase the time that it takes to achieve full power from idle/low MAP during a go-around/roller landing/stall recovery which could be unacceptable. Also, there are times when you might want max RPM/prop blades on the fine pitch stops at low MAP to give some extra drag and help slow down. And if the prop blades fine off too slowly when increasing power you may overboost the engine.

It is interesting to note that the Merlin engine has far less controls than the Lycomings to which you refer and is simpler to operate! The DB605 in the Bf109G (and similar installations in other German WWII aircraft) had an automatic function for matching RPM and MAP to maintain an optimum relationship, but it could be, and was, operated in a manual mode for take-off and landing. All engines are different and you need to match cost, reliability and operating flexibility and then see what controls are needed. BTW, the Hawk also has switches for the ignition, and electric fuel pump and engine starting, but never let the facts cloud a good arguement!!

Did we ever fly the Hawk together? Best regards. L.

[Genghis the Engineer]

Looking at your profile I suspect that if not that, maybe a Hunter, Tucano and/or Jag, probably roundabout 93-97 at BDN ? Would I be right in guessing that you were a tutor on 20FTE/55FW course? I hope that having joined Pprune you'll come over and contribute to the FT forum occasionally.

I accept that a degree of control is certainly a good thing, but I think there is so much that the increased risk of miscontrol is far greater than the risk of failure would be due to automation in many aircraft.

What I suggested in my first post was an extreme view - I'd not personally wish to do away with a manual fuel cut-off for example (even if the codes would let me !). But carb-heat for example is successfully automated with near 10^5 reliability for a very low cost on many microlights, so I can't see why it can't be fitted onto a light aircraft. Ditto choke/prime and friction without any reduction in safety - I can think of many problems caused by having a primer control on a Lycoming and none prevented by one.

G

[Shaggy Sheep Driver]

The Yak 52 has automated mixture control so no cockpit lever - and I certainly never missed it. I would miss the prop lever, however. That big paddle prop in fully fine with the throttle closed makes a great airbrake and allows *very* steep approaches. And if you need to go-around, you're already in fully fine.

And the '52's aeros power setting is full power power at 82 percent RPM (why 82, I wonder??).

Cars have complex but reliable electronic control systems with fuel injection which eliminates carb ice and ensures the mixture is always optimum and the tickover rpm constant regardless of ancilliary load. But aero engines are simply are not produced in sufficeint volume for such developments to be commercially viable.

SSD

[Blacksheep]

Hmmmm, if LOMCEVAK's opening comments are correct, why do we civilian heavies have FADEC, the 'Quiet Dark Cockpit' philosophy [with two crew operation too!] leave most of the engine monitoring to computers - to the extent that the flight crew don't even have access to most of the performance data - and still have the highest reliabilty and lowest operating costs in the history of aviation? Should we have stuck to the well tried methods, with armies of flight engineers twiddling and tweaking away at their massive banks of levers and dials? or, as a simple systems development engineer, am I missing something, somewhere? :confused:

**************************
Through difficulties to the cinema

[LOMCEVAK]

Blacksheep,

My comments related to GA piston-engined aircraft, which was where Genghis started this thread. Obviously, for commercial turbine-engined aircraft the advantages of FADEC are very worthwhile from both a commercial and a safety point of view. However, good point to raise to keep this all in context. Horses for courses.

Rgds

L

[Blacksheep]

LOMCEVAK,

While Ghengis was twiddling his mixtures and adjusting the carburettor heat he mentioned sitting in the back seat of a Hawk somewhere - hardly General Aviation - but I suppose that depends on your point of view. ;)

My main point and surely that of the aviation hordesman himself, is that the benefits of engineering development seem to have passed much of the general aviation world by. While it is true that eliminating carburettor heating, mixture control and other contraptions for manual fine tuning the powerplant increases system complexity, the significant reduction in cockpit workload more than compensates for this. As Ghengis suggested, it reduces the number of things that the generally less experienced light aircraft pilot can operate incorrectly and leaves more time for enjoying the view.

I admit that being in avionics development biases my opinion towards automation, engine mapping and the like, but similar engineering developments have made all our lives so much easier over the last century. Resistance to change, even beneficial changes, is a part of human nature that needs to be overcome - please allow engineers like Ghengis to show the way. :)

**************************
Through difficulties to the cinema

[Genghis the Engineer]

A Hawk is arguably not GA, until it's retired from service and operated by over-moneyed PPLs anyway. But, people like BAe have the money to spend on R&D so we'd be daft not to steal their ideas.

Engineer leading the way? Village bar is on the right as you go in.

G