Solenoid Actuated Poppet Valves

Could a solenoid be constructed to mimic the proper valve position vs time curve of a engine cam shaft? Maybe a circuit and/or several coils in each solenoid?

It may be tricky or impossible to build such a solenoid that works well over all rpms, but if it could be built, it could be mass produced very cheaply, maybe cheaper than a cam system.

The other concerns such as force necessary to accelerate valves (are solenoids orders of magnitude more inefficient than cams?) and mangled valves from electronic failure (like other fail safe systems set the default of every valve in a safe position) shouldn't be problems.

Bret Cahill

I brought this up some time ago in some mechanical group, and they seemed to indicate that it's been tried a long long time ago, and abandoned because it's noisy or something. Actually, I still think it's a pretty cool idea, when you consider how many HP a camshaft/rocker/ pushrod/lifter arrangement must consume.

And 100-amp semiconductors are common these days.

Anybody want to start hacking their cylinder head? ;-)

Cheers! Rich

Ooh, snotty!

Fast, powerful solenoids have a lot of copper loss, hence the cooling problem. Not a lot of energy would be recovered.

I'd think that a roller cam could recover at least 60% of the energy dumped into the valve spring. The exhaust valve does some real work against pressure, but that's a different issue.

But it's not worth arguing about, or analyzing, because it doesn't work.

I think about everything. That's more fun.

John

.. . .

I wouldn't be so sure.

A solenoid might be or has been built that could recapture about as much wasted energy as that by a cam. As the valve slowed it would recharge the battery or power the next cylinder's solenoid.

In any event your concern is valuable, even if wrong. I didn't think about it and I run energy balances on everything.

Bret Cahill

"The errors of great men are more fruitful than the truths of little men."

-- Nietzsche

That is interesting. One difference is that motors have magnets (or fields) but solenoids generally don't. A linear voice-coil actuator is sort of half-and-half, and can be reasonably efficient as a mechanical-to-electrical converter. A solenoid only stores energy in its inductance (which admittedly decreases in the dropout direction, which pumps some energy back into the winding) but a millisecond high-power solenoid will have to have very low inductance and run at very high peak current, so will be very lossy. A solenoid is just a linear variable-reluctance motor!

Oh, it can be made to work, like variable valve timing and all sorts of other tricks. But it sounds outrageous to me to replace a nice reliable cam and a belt with two solenoids and a heap of electronics per cylinder. I think the 42-volt boys sometimes advocate electrical valve actuators (and power steering, and air conditioning.) When it becomes commercially successful and commonly done, I will be suitably humbled.

I guess you could do the Cadillac variable-number-of-cylinders trick. You'd have to when a solenoid or a driver fails.

You can't restrict my Freedom of Snip rights.

John

Greetings Bob, I used to shudder at the thought of "drive by wire" systems. Mechanical systems just seemed so robust and forgiving when they do fail, because often the failure isn't complete, but happens over time. But, I had the left front suspension break and fold under a car I was driving. In a parking lot going about 10 mph. There was no controlling the car. Another time, I heard a squeak coming from the engine compartment of an econline van I was driving. I turned on the turn signal to change lanes over to the emergency lane. Before I got there the shaft going throught the water pump let go and the fan screwed itself through the radiator. These were sudden failures with no warning ar just a second or two warning. But many electronic systems are much less likely to fail. And they really don't wear out. My Dad's company makes machines which check the probe cards which check the finished wafers before they are sliced up into seperate dies. He's been an electronic engineer for about 50 years. When I told him I had a chip fail in a 20 year old machine he said that the chip was damaged when new. He said it most likely was not damaged by age or run time. And that the type of chip it is should function virtually forever. So I'm waiting for the all electronic car. ERS

I seem to recall reading in mags like EE Times about work on "electronic valve timing" or words to that effect. I had the impression that they were using solenoids, and had similar concerns to yours.

The "all-electronic car" idea seems to be gaining momentum, but it gives me the creeps, especially when they get to electronic steering. Seems like they are just trying to make everything electronic "because they can".

However, many years ago I worked for General Motors Cadillac Division when electronic fuel injection was just coming on board. For a while I was one of team that went around the plant rescuing EFI cars that had died. At the time it was a really fragile system. There were something like

32 wires to the EFI control box (which was mostly analog!), and almost any of those could disrupt operation if it came out of it's little connector. (We carried a supply of paper clips to poke them back into their shells.) Worse yet, the early EFI system put a massive amount of fuel into the chambers if you pumped the pedal when cold-starting, as the drivers were used to doing on carbureted cars. This would foul the plugs so badly that they had to be replaced in order to start the car. (Simple manual cleaning of plugs didn't help.)

I recall thinking that this was a terrible step backward for overall reliability. There were tons of things that could completely disable an EFI car, whereas almost nothing bothered a carburetor short of somebody dismantling it an throwing in a handful of dirt.

Nowadays, though, EFI is the norm and they certainly seem to have gotten their act straight. Overall, seems far more reliable than carburetors. So, maybe someday we'll feel that way about electronic valve trains and steering. OK, so maybe a few catastrophic failures in the learning curve to stimulate new ideas....

Bob Masta dqatechATdaqartaDOTcom D A Q A R T A Data AcQuisition And Real-Time Analysis

I never actually said you were wrong, just that even if you were wrong, the issue was valuable.

It's interesting electric motors can be 95% efficient but something linear is only 0.1% efficient.

Sounds reasonable.

There are some web sites with solenoid engines that supposedly run.

That's what we want to hear. Just don't delete my Nietzsche quote next time.

Bret Cahill

Actually, existing engines are awfully good. Their only problem is that most of them are designed to accelerate 5000 pounds of ugly steel surrounding one 110-pound soccer mom. We don't really need more efficient engines as much as we need higher gas prices. No major technological breakthroughs are needed to make that happen.

John

A lot of people seem worried about a computer glitch causing the valves to get mangled, etc. but the drivability of a motor vehicle drops to zero anyway without the computer. For the past 20 years automotive computer failure hasn't been that big problem.

Besides there are various fail safe protections that could be implemented.

Bret Cahill

Eric R Snow snipped-for-privacy@whidbey.com >

I hate the way every microwave oven is different. Just put a dial on the #@!&%! thing and forget being cute with a lot of buttons.

Good story.

Maybe someone knew what he was doing.

In politics we're talking trillions of dollars down the drain.

An unreliable piece of electrical or mechanical junk is positively charming in comparison.

Where do I sign up?

That's what started it. Instead of one engine, I'ld have 2. A low power high efficiency engine for 50 mpg and a high power low efficiency engine in case you needed to haul something or get on the freeway fast.

Someone mentioned a convertable 4 stroke/

2stroke engine and either you had to slide the cam shaft off the rocker arms or you had to operate the valves some other way.

Bret Cahill

i seem to remember hearing long time ago that the amount of instantaneous power required for a solenoid to operate normal valves with return springs is so large that the resulting inefeciencies leads to a very large power consumption making it rather ineficient. maybe with latest magnets or posibly even high temperature superconductors it wld be advantagous. maybe a stepper motor could be used instead, with a similar arangement used inside disc head actuators.

however there are alternatives, depends what you are trying to acheive, variable valve timing has been acehived with hydrolics on lotus and now other cars i think, however why restrict yourslef to popet valves, consider desmodronic valves wich have no return spring but rely on a 2nd mechanical or potentialy electromechanical force to return the valve, and the presure inside the cylnder finshes the job.

alternatvly the cam shaft could simply be driven by a steper motor, or the cam shaft cld be driven in the normal way but a mechanism driven by a steper motor could alter the angle of the cam shaft relative to the drive drive puley.

there was talk about an all ceramic engine being 10 years away but i heard this wel over 10 years ago i think, i always wondered about the posibility of sliding valves, maybe just a rotating disc, posibly ceramic just like in some taps, they are very efiecient in the amount of energy required to operate them and give good life.

Colin =^.^=

Anything is possible, but if you just want variable valve timing try this:

Convert the lifters to have little thin rollers on the end... sort of like tiny inline skate wheels but made from steel and running on a bearing. Now the lifter is only useing a narrow section of the circumference of the cam lobe.

Next, make each cam lobe segment very narrow and stack several different shaped lobes right next to each other.

Finally, shift the cam shaft in and out as it turns.

I personally did a VERY crude version of this years ago on my first "hot rod" as a teenager. I wanted to grind the cam shaft as several of the other "gang" had done, but I knew my dad would kill me if he ever heard the car running rough so I ground HALF of each cam lobe, added a crude lever system to shift the cam in and out about 1/4 inch which was all the play I could get out of the cam journals. It was enough, and I could idle quietly out the drive way, reach under the dash, pull the lever and be "one of the cool ones."

Of course I totalled the car about 6 months later and my Dad did the worst possible thing to punish me: He cried. Later though, when he noticed the "modification" he helped me start a rebuild on the car. Didn't finish it, but I took it as a sign he had been impressed.

Anyway, I've never seen such a thing since, but I've been out of cars for a long time.

Someone suggested a 2-stroke/4-stroke convertable engine. 4-stroke if gas was expensive and four stroke if you needed more power.

The sliding cam was the only way other than solenoids.

Bret Cahill

In message , James Newton writes

Snip, Fiat use a scheme like this. Their system uses two meshing cross cut gears, one internal and the external one is rotated to push/pull the camshaft in and out to a different portion of the profile

--
Clint Sharp

===================

Thats' been done already, they're still ironing out the bugs.... not ready for domestic use 'yet'.

I must say I just 'love' the possibilities this will introduce. Imagine an engine that could putt around in city traffic & still go like a Ferari on the open road.

--
Australia isn't "down under", it's "off to one side"!

stanblaz@netspace.net.au
www.stanblaz.customer.netspace.net.au
www.cobracat.com (home of the Australian Cobra Catamaran)