fly by wire throttle

Free Open Source Firmware project discussion forum. Post your Free Open Source firmware projects here!
jtw11
TO220 - Visibile
Posts: 12
Joined: Mon Dec 31, 2012 2:02 pm

Re: fly by wire throttle

Post by jtw11 »

Does this mean I'm going to have to concentrate on the 'one-stroke' engine design instead??? (Using a 'double ended' piston).
Haha, 'fraid so! Sorry for the delay, I took two quick pictures in the entrance the other day... Looks to have some pretty series valve trickery!

Image

Image
TheRevva
DIP8 - Involved
Posts: 26
Joined: Thu Dec 13, 2012 1:10 am

Re: fly by wire throttle

Post by TheRevva »

YIKES!!! Looks like someone gave the left bank of a V6 a somewhat malignant / cancerous growth!!! <Grins>
Nevertheless, it's VERY impressive! So thanks heaps for the pix!

It kinda makes me wish I had free access to a decent foundry and a limitless supply of funds.
I'd LOVE to be able to 'tinker' with some concepts like these...

Heck, it'd be enough to just have a MultiAir / Pattokon engine to 'experiment' with.
Imagine changing the crank / cam sprockets to spin the camshaft at crank speed instead of at a 2:1 reduction...
Effectively, it's suddenly a 2-stroke (more or less). When you want it to be a 4-stroke, the ECU simply opens the hydraulic 'pressure relief' valves at the opportune times.
It may still require a regrind of the camshaft since running it at double speed will double the effective acceleration curves of the poppet valves.
If so, it'd probably be easier to grind a 'custom' cam doubling up on the number of lobes per valve, thereby retaining the existing cam / crank sprockets.

Anyway, we're probably a bit 'off topic' here (Sorry Fred).
I'd love to chat more on the subject, but maybe Fred would prefer it in some 'other' thread?

(Minor edits to fix typos)
User avatar
Fred
Moderator
Posts: 15431
Joined: Tue Jan 15, 2008 2:31 pm
Location: Home sweet home!
Contact:

Re: fly by wire throttle

Post by Fred »

Guys, see post on previous page that was held back by n00b restrictions. Just FYI. I didn't even read it yet.
DIYEFI.org - where Open Source means Open Source, and Free means Freedom
FreeEMS.org - the open source engine management system
FreeEMS dev diary and its comments thread and my turbo truck!
n00bs, do NOT PM or email tech questions! Use the forum!
The ever growing list of FreeEMS success stories!
MotoFab
1N4001 - Signed up
Posts: 307
Joined: Thu May 29, 2008 1:23 am
Location: Long Beach CA

Re: fly by wire throttle

Post by MotoFab »

jtw11 wrote:As much as I wish it was, I'm afraid the dyno these tests were performed on does not belong to me, no.
Oh, I didn't mean 'do you own a dyno?'. I was asking about a 'motoring' dyno.
TheRevva
DIP8 - Involved
Posts: 26
Joined: Thu Dec 13, 2012 1:10 am

Re: fly by wire throttle

Post by TheRevva »

manousos wrote:
TheRevva wrote:What's stopping them using another such solenoid on a pressurised fluid supply to OPEN the valve too?
It is more easy than this. You don't need a second solenoid electromagnetic valve. Take a look at the HyDesm (Hydraulic Desmodromic) at http://www.pattakon.com/pattakonHyDesmo.htm
EXTREMELY interesting thanks Manousos... I'm learning a LOT!
In my (perhaps overly simplistic?) mind, I still see it as being more complex than it NEEDS to be.
There's still a camshaft-like device with the sole purpose of running a localised 'hydraulic pump' at each valve (via the eccentric shaft).
Current ICEs traditionally have some form of oil pump (mostly driven from a cam gear in my experience) for the traditional lubrication needs.
Why can't this be a simple electric pump instead? (And if desired, it could be a two compartment pump with one compartment dedicated to a valve train hydraulic supply).
We already have an ECU, so suitable 'safeties' can be incorporated to 'kill spark and fuel' if output pressure drops too low.

If you had a suitably reliable solenoid such as that described in the HyDesmo system, it could completely control both valve opening AND valve closing.
The solenoid would have a reasonable return spring that would 'default' to pushing the valve closed as a safety precaution (to protect an 'interference' engine from self destruction if a solenoid failed)
Additionally, a weaker than normal valve spring might be used (which would pull the valve closed in a more traditional sense).
The 'modulation' of the solenoid is under direct control of the ECU thereby allowing INFINITE valve timing control. (And yes, this would also therefore allow dynamic 2-stroke / 4-stroke switching).
Like the HyDesmo system, the valve closing can be regulated with 'hydraulic braking' by progressively closing vent ports in the opening oil chamber.

Assuming the above:
  • If the oil pump were to fail, the valve springs would maintain the valves closed (And the ECU might even disallow the starter to energise until sufficient pressure is available?)
  • If a solenoid coil went 'open circuit', the associated solenoid return spring would make sure the valve stays closed.
  • There is a definite ISSUE if a solenoid 'sticks' in the open position, or if it's directed to open the valve at the wrong time.
Systemic requirements:
A decent source of hydraulic pressure (which could easily be routed through a dedicated valve train oil gallery in the head)
A single solenoid (operating akin to a SPDT hydraulic switch) per valve
A fairly large / complex amount of code in the ECU <Grins>

And finally...
When a modern 'green' engine deactivates a cylinder, it's logical to simply leave the intake valve closed. (Obviously, any associated injector would be inoperative on a shutdown cylinder)
However, my 'opinion' is that the exhaust valve should be held OPEN for a significant angle of crank rotation. It should only be closed to alleviate interference issues nears TDC piston travel.
(While it's not essential, the additional 'breathing' of an inoperative cylinder might help to limit associated losses, and our Cray XMP ECU can surely handle it?).
The potential downside of this is that it might cause 'thermal issues' on that 'shutdown' cylinder. It depends upon the temperature of the 'intake charge' that it would suck in through the exhaust valve. Normally, this would consist of the exhaust gasses of a nearby cylinder and, with careful 'tuning' of the exhaust manifold, it could probably maintain the cylinder at a 'normal' temperature?
Heck, our Cray ECU could even monitor the temp of the cylinder liners and 'adjust' the exhaust / inlet valves in order to maintain a suitable temperature?

Perhaps the Cray ECU is mild overkill, but I'd have reservations of this being achievable with the FreeEMS CPU on something like a V8. Perhaps dedicating an MC9S12XDP512 to each cylinder and a 'supervisory' CPU sitting above?
(I would NOT like to be the one having to maintain THAT code!!! <Grins>)
manousos
TO220 - Visibile
Posts: 7
Joined: Thu Jan 10, 2013 5:35 am

Re: fly by wire throttle

Post by manousos »

TheRevva wrote:
manousos wrote:
TheRevva wrote:What's stopping them using another such solenoid on a pressurised fluid supply to OPEN the valve too?
It is more easy than this. You don't need a second solenoid electromagnetic valve. Take a look at the HyDesm (Hydraulic Desmodromic) at http://www.pattakon.com/pattakonHyDesmo.htm
EXTREMELY interesting thanks Manousos... I'm learning a LOT!
In my (perhaps overly simplistic?) mind, I still see it as being more complex than it NEEDS to be.
There's still a camshaft-like device with the sole purpose of running a localised 'hydraulic pump' at each valve (via the eccentric shaft).
Current ICEs traditionally have some form of oil pump (mostly driven from a cam gear in my experience) for the traditional lubrication needs.
Why can't this be a simple electric pump instead? (And if desired, it could be a two compartment pump with one compartment dedicated to a valve train hydraulic supply).
We already have an ECU, so suitable 'safeties' can be incorporated to 'kill spark and fuel' if output pressure drops too low.
. . .
It is not the pressure of the hydraulic circuit that opens and closes the intake valves in the MultiAir of FIAT and in the PatAir and HyDesmo of pattakon. The oil pump supplies oil in a relatively low pressure to just fill the oil chamber during the period the valve remains closed. Then the trapped (into the oil chamber) oil behaves as a “solid” / "steel" pushrod untill the electromagnetic solenoid release valve is actuated by the ECU, the hydraulic "push rod" collapses and the intake valve - under the action of the valve spring - closes independently of the cam.

It is the camshaft that "mechanically" controls the opening of the intake valve. The oil in the oil chamber is the "uncompressed" means that only transfers the force from the cam lobe to the valve.

The necessary force that opens and accelerates / decelerates (ramp) an intake valve is strong.
Even stronger it is for an exhaust valve because of the pressure inside the cylinder during the "valve opening".

Quote from http://www.pattakon.com/pattakonKeyAdv.htm#dvva :
Most VVAs control exclusively the intake valves, either because of cost, or because they cannot control the exhaust valves. The following is a copy from engine-technology-INTERNATIONAL:
. . . Valeo refers to the design (i.e. their electromagnetic VVA) as "half camless". The reason for this is that the forces required to operate the exhaust side are so high, the performance benefits are off set by the increased parasitic losses . . . The spokesman adds: The main benefits are on the inlet side as pumping losses can be reduced. We concluded that if 80% of the benefit is on the inlet side, then this is the best balance. . . . the two concepts together (the electromagnetic "half camless" and a mild hybrid based on StARS+X technology) have been given a dramatic boost by the award of US$108 million (a hundred and eight million US dollars) from the French Industrial Innovation Agency
. If you need to apply this force by high pressure oil entering into an oil chamber and pushing a plunger at the top of the valve stem, you need a high pressure hydraulic system (providing oil at, say, 50 bars).


The electromagnetic system of Valeo offers - theoretically - full control over the intake valve lift. The same is offered by the electromagnetic VVA's made and tested by Siemens, by Lotus etc.
The side effects in practice (like the need for an additional safety system to avoid the posibility of "valve piston collision" in case of malfunction of a sensor and/or electric cirquit and/or ECU, like the low rev limit, etc) and the high cost (nearly US1,000$ per valve, according Lotus, if I remember correctly) keep those systems out of mass production.

If I understand it correctly, your proposal for a fully variable hydraulic VVA is similar to the electromagnetic VVA's.

Compare it to the MultiAir of FIAT.
In the Alfa Romeo Mito MultiAir there is an exhaust camshaft (for the actuation of the exhaust valves as in every conventional engine). In the exhaust camshaft they have been added four additional intake cam lobes, one per cylinder of the engine.

Image

The conventional intake camshaft is "gone".
The "built in" safety is not worse than that in the best conventional / mechanical systems: a "valve piston collision" can happen only due to a broken timing belt (or due to an extreme high revving far above the red line, as in the conventional engines).

When the MultiAir system operates at full power (WOT, high revs), the intake valve follows the motion dictated by its intake cam lobe (like having a solid "push rod" disposed between the valve and the cam).

Image

Without the need for a phaser (or VVT) between the cranks
haft and the camshaft, the moment the intake valve opens - and closes - varies: early closing mode, late opening mode, deactivation, full lift, multilift:

The MultiAir of FIAT is in mass production for over three years. Chrysler already uses it in their cars in the USA.
MultiAir / TwinAir / UniAir was voted as the best engine of the year 2011.

The PatAir:

Image

is an evolution of the MultiAir of FIAT.
Since neither the long duration intake camlobes of the PatAir adds any cost, at all, nor the ECU reprogramming adds any cost, at all, to double the available modes of the MultiAir / TwinAir of Fiat has no manufacturing cost, at all.

The HyDesmo at http://www.pattakon.com is an evolution of both: it is the Desmodromic version (no valve springs are necessary). The solenoid valves can be even simpler and slower (the basic control is mechanical; only the microadjustment between different cylinders needs the ECU and the feedback control).

Thanks
Manousos Pattakos
MotoFab
1N4001 - Signed up
Posts: 307
Joined: Thu May 29, 2008 1:23 am
Location: Long Beach CA

Re: fly by wire throttle

Post by MotoFab »

Best of luck with your last few months of classes, jtw.
jtw11 wrote:[Keeping] all valves closed on deactivated cylinders, but this is not what decreases pumping losses.
Pushing a piston up against closed valves, is the same as a pulling a piston down against closed valves. The two are the same in that neither activity uses any power over a full crank rotation. That's the thing, the mechanics of air flow are linked to the mechanical devices that cause the air flow. In this case that's the 'circular lever arm' of the crank throw. Air moving through the motor is not a separate phenomenon.

jtw11 wrote:Please, using the provided dyno data, explain what else is causing the lower pV loop to decrease in size with an opening throttle?
A pressure/volume diagram does not show the amount of power required to move air through a motor. A PV graph shows pressure, not airflow. The intuition is that a PV graph is a graph of 'voltage', and since a PV graph doesn't show the airflow 'current', the 'power' consumed in moving the air cannot be derived.

jtw11 wrote:f you allow the air to pumped through the cylinder more easily, pumping losses go down.
Lower resistance equals higher pumping losses. In the same way that electrical current through a lower resistance is higher. The mechanical formula is the same as Ohms law, and electricity is mechanical. Less power is consumed across a high resistance, not more.

For example if intake resistance is infinite then flow is zero, and if flow is zero then pumping losses are also zero because no air mass is being moved. The formula is the the same as Ohms law. In the Ohms law power equations, substitute air flow for current flow.


jtw11 wrote:Freeing up airflow through the intake system reducing pumping losses, get rid of the horrendously inefficient butterfly, and you do exactly that.
An intake design that doesn't use a throttle butterfly, and instead controls intake airflow solely with intake valve control, doesn't reduce pumping losses. Decreasing or stopping the air flow is what decreases or zeros the pumping losses. Resistance decreases the power consumed.

The benefit of controlling the intake flow with the intake valves instead of a throttle butterfly is that it relocates the point of maximum pressure and density differential. With a throttle-controlled intake the maximum differential point is across the throttle plate. With valve-controlled intake the maximum differential point is across intake valves.
manousos
TO220 - Visibile
Posts: 7
Joined: Thu Jan 10, 2013 5:35 am

Re: fly by wire throttle

Post by manousos »

MotoFab wrote: Pushing a piston up against closed valves, is the same as a pulling a piston down against closed valves.

Think the following case:

You keep the intake valves of a cylinder open from the overlap TDC to 30 degrees after the overlap TDC. Then you close the intake valves. The volume of the air (or mixture) entered into the cylinder is about 1/8 of the displacement of the cylinder. As the piston goes to the BDC the air expands and its temperature drops.

Theoretically, with adiabatic "walls" the absolute temperature (Kelvin) of the air at the BDC would be nearly half of the initial (with air at about 50 degrees Celcious before the intake valves, at the BDC the air temperature - in case of adiabatic walls - would be some -120 degrees Celcious). The pressure drops to, nearly, 5% of the pressure just before the intake valves.

Practically the walls (cylinder wall, piston crown, cylinder head, intake valves, exhaust valves) are anything but adiabatic. The air during the expansion towards the BDC warms a lot by the walls; at the BDC its temperature is more than 100 degrees Celcious.

During the following "compression stroke" of the piston, the force on the piston is quite different than it was during the "intake stroke" of the piston (which means: you take back less energy than you offered during the previous "expasion").

For instance, the pressure on the piston at 100 degrees before the BDC and 100 degrees after the BDC is quite different.
At 30 degrees before the combustion TDC (i.e. 330 crank degrees after the overlap TDC) the temperature of the air is expected to be more than 300 degrees Celcious (same reasoning as before).

So, a significant amount of mechanical energy needs to flow from the flywheel / crankshaft towards the working medium (air or mixture) during this "innocent" transition from 30 degrees after the TDC to 30 degrees before the next TDC.

This is not a "pumping loss" (because the air is already inside the cylinder), but it is a significant loss of mechanical energy.
The lighter the load, the more important (as percentage of the provided by the engine energy / power) this loss.

The previous are not theoretical; this is the way the MultiAir of FIAT and the valvetronic of BMW operate (they call it: early intake valve closing).


To get deeper in the subject, think the way the PatAir operates (Outgoing Air Control):

The intake valve is open from the overlap TDC to 30 degrees before the combustion TDC (i.e. for 300 crank degrees). The air or mixture enters the cylinder (no sub-pressure inside the cylinder) and exits from the cylinder back to the intake manifold during the "compression stroke" of the piston.
At 30 degrees before the combustion TDC (wherein the intake valves close) the temperature of the air is way lower than the temperature of the air in the MultiAir FIAT / valvetronic BMW engine, which means that the following combustion - expansion will be at significantly lower temperatures (lower thermal losses, fewer emissions etc).
And there is no consumption of mechanical energy (at low - medium revs, only the mechanicall friction from the piston rings and the crankshaft bearings is significant).

The new model Dart "Aero" / MutliAir of Chrysler needs premium gasoline to avoid the knocking.

Don't be confused by the fact that the engine operates according a lower effective "compression ratio". The efficiency depends on the "expansion ratio" (Miller cycle).
Doesn't a limited Miller cycle (combined with high "expansion ratio" of 13:1 and 14:1) is used in the most fuel efficient modern engines (like PRIUS of Toyota, SkyActive of Mazda, DIG-S of Nissan)?


At operation with a very little quantity of fresh air, say 10% of the capacity of the cylinder, the “Ingoing Air Control” mode of MultiAir FIAT / valvetronic BMW runs as a refrigerator (or heat-pump) working for cooling purposes and indicating negative indicated power output, whilst the same engine running according the “Outgoing Air Control” would indicate its best indicating fuel efficiency.
Against the common sense, at “extremely light loads” the “Ingoing Air Control” provides negative indicating power, whereas the “Outgoing Air Control” mode delivers its best indicating efficiency.

Thanks
Manousos Pattakos
MotoFab
1N4001 - Signed up
Posts: 307
Joined: Thu May 29, 2008 1:23 am
Location: Long Beach CA

Re: fly by wire throttle

Post by MotoFab »

Thanks Manousos, for the short treatise on some of the differences between the two modes of operation. That is an interesting method of essentially eliminating the compression stroke, and taking good advantage of high expansion ratio. During 'outgoing air control' which crank angles are the fuel and spark?
MotoFab
1N4001 - Signed up
Posts: 307
Joined: Thu May 29, 2008 1:23 am
Location: Long Beach CA

Re: fly by wire throttle

Post by MotoFab »

manousos wrote:[Example regarding MultiAir early valve closing.]

You keep the intake valves of a cylinder open from the overlap TDC to 30 degrees after the overlap TDC. Then you close the intake valves. The volume of the air (or mixture) entered into the cylinder is about 1/8 of the displacement of the cylinder. As the piston goes to the BDC the air expands and its temperature drops.

[During the down stroke, heating of the 1/8 cylinder fill occurs.]

During the following "compression stroke" of the piston, the force on the piston is quite different than it was during the "intake stroke" of the piston (which means: you take back less energy than you offered during the previous "expasion").

For instance, the pressure on the piston at 100 degrees before the BDC and 100 degrees after the BDC is quite different. At 30 degrees before the combustion TDC (i.e. 330 crank degrees after the overlap TDC) the temperature of the air is expected to be more than 300 degrees Celcious (same reasoning as before).

So, a significant amount of mechanical energy needs to flow from the flywheel / crankshaft towards the working medium (air or mixture) during this "innocent" transition from 30 degrees after the TDC to 30 degrees before the next TDC. t is a significant loss of mechanical energy.

Some would not call that a loss. And would say that the process of heating the air of that 1/8 partially-filled cylinder is a heat recovery process. In that the compression pressure is higher due to pre-combustion heat recovery, even though the cylinder is only partially filled. It is a novel method of heat recovery.
Post Reply