Take a look at pattakon for VCRs :







Thanks
Manousos

How do you keep the correct cam timing with that?
I quite like the idea though otherwise.

Read in the KeyAdvantages how the pat-head VCR manages to keep the timing correct (unchanged) for all compression ratios. It is quite simple.

Manousos

Fun website, wish I had that much time on my hands to play, I'm actually quite jealous, he must be a happy and content man

It's all junk but I'm still jealous

A fully functional and drivable variable compression ratio engine from almost ten years ago

Boy! Why don't we just stick with having the pistons go up and down?

http://www.fs.isy.liu.se/Lab/SVC/

Problem with this VCR engine is that you get a sore arm after a while.

Oh and then i found this...

http://www.iwsti.com/forums/off-topic/1570...on-engines.html

Heath Robinson is dead.
But his spirit lives on. And on. And on..

The Ehrlich system is the most promising VCR I have seen, plus it features variable displacement. Moving the pivot left and right varies the displacement. Up and down varies the CR.

Some history http://en.wikipedia.org/wiki/EMC_Motorcycles

Didn't Yamaha have a hydraulic version on their two stroke 500cc race bikes? I could not find a picture of it but I remember reading it years ago.

As in the multilink VCRs (like Nissan's, Daimler's, Honda's) similarly in Ehrlich's VCR the strong bending loads on the VCR parts make them heavy and increase the inertia forces and the friction. The force on the crankpin bearing is significantly heavier than the piston force (friction). Because the control mechanism takes strong forces, the response of the VCR is not so fast and the crankcase has to be stronger (and heavier) than conventional in order to support the VCR mechanism.

It is also the complication and the cost.
For instance, take the case of Nissan's multilink VCR (ready for production in 2010), or Honda's multilink VCR , or Merchedes' multilink VCR (Ehrlich VCR is similar to them) and think how many complication it adds to the conventional engine. For instance, for a V-8 engine it needs eight rockers, eight rods, two control shafts, i.e. 18 heavy, expensive and difficult to make and support parts.

In comparison take the application of patcrank VCR on a V8 conventional engine:



All it takes is four slim and lightweight secondary connecting rods (the blue ones) and one thin and lightweight secondary crankshaft, i.e. 5 lightweight, and cheap to make, moving pieces in total, plus one slow moving control frame.
The patcrank VCR mechanism deals with only a tiny part of the piston forces (friction reduction, response, lightweight construction).
When you do more with only 6 cheap, lightweight and of low friction parts, what is the reasoning of using the 18 parts solutions?
Note that in patcrank VCR the main crankshaft is conventionally supported onto the conventional crankcase and drives the flywheel and the gearbox in the conventional way.

You can find the animation for the V-8 patcrank VCR (it shows how the compression changes in a V-8) and instructions at the bottom of patcrankVCR or directly at patcrankV8.

Finally a look at the following plot is quite explanatory (excluding the pattakon VCR shown at the sides, the rest plot and comparison is from mce-5 web site) :



Thanks
Manousos

Lately Lotus presented its omnivore VCR for two stroke engines.
They have a video animation in youtube.

Manousos

The basics of V8 patcrank VCR are shown in this simple gif animation:



The main crankshaft (red) is conventionally supported on the crankcase and drives conventionally the flywheel and the gearbox.

If you do not want to open the "exe" animation for more, you can alternatively open the:

http://www.pattakon.com/patcrank/patcrankVCR.wmv

WMV video animation.

Manousos

Are the benefits of VCR really of enough significance to warrant the extra complexity, weight and expense of these mechanisms? Some of these systems look like they would almost double the complexity of a conventional engine, and I wonder about the cost:benefit ratio...

The benefits are considerable, especially with downsized, high boost engines. The challenge is to reduce the cost side of the ratio.

The number of patents and patents applications of Nissan, Honda, Mercedes etc for their multilink VCR says a lot.

Nissan, after putting their VVEL Variable Valve Actuation in their top models, now plans to put, during 2010, their multilink VCR in mass production.
They do know the cost:benefit ratio and the complication:benefit ratio of their VCR, yet they proceed with mass production.

In comparison, patcrank VCR makes more for less, keeping way lower the cost:benefit and complication:benefit ratios.

Manousos

So the CR will be basically linked to MAP to give better performance at no/low boost, and allow higher maximum boost pressures to be used?

Yes. The main shortcoming of turbo downsized engines for production cars is the poor part-throttle/off-boost economy of the low compression engine.

Think why the hybrid cars are so better (in CO2 emissions, i.e. fuel economy) in down town traffic compared to conventional cars.

Having a 200 PS conventional engine to pull your car in urban cycle, the engine operates most of the time at light and very light load (where the consumption of fuel, Kg, per generated KWh of energy, is way higher compared to the case the same engine runs at heavy load).
In order to keep constant 100Km/h speed on a flat road you need no more than 20bhp of power. In down town traffic you need way less.

In the hybrid car the thermal engine operates from time to time (but at optimum – as regards economy and emissions – conditions, like 3500 rpm and 80% open throttle) to charge the battery.
Despite the energy lost in order to change the mechanical energy into electrical energy by the electric generator, the energy lost in order to store the electrical energy into the battery, the energy lost into battery until discharge, the energy lost to change the electrical energy of the battery back into mechanical energy on the electric motor to finally move the car, the total efficiency (Kg of fuel per Km of distance covered) is way better than in the case of the conventional spark ignition engine running at light load.
In other words, the thermal efficiency of the conventional engine at heavy load is way better than the thermal efficiency of the same engine at light load.

Alternative to the hybrid technology is the Variable Compression Ratio (VCR) for both naturally aspirating and turbocharged engines.
At partial loads the VCR increases the compression ratio as much as possible (just before knocking) making the engine as efficient as if it were working in heavy load.

Manousos

Not quite, because pumping and friction losses at part throttle still reduce efficiency significantly, but VCR does help a lot. The high boost downsize engine fares better (than the NA) in this regard since its pumping and friction losses are lower.

There is the Desmodromic VVA (or DVVA) to minimize the punmping and friction losses ( read at Desmodromic Fully Variable VVA for more). The DVVA controls independently the valve duration and the valve lift, is rid of valve springs and other restroing springs.

The high boost downsized engines do take advantages from a VCR, but this is also true for the naturally aspirating engines.
In town traffic, a big conventional naturally aspirating V8 is more like a CO2 generator than a mechanical energy generator. This can change by a good VCR. A good VVA in the cylinder head can improve things further.

Manousos

Agree on the VVA. My point on the high boost downsized engine is it has a greater dynamic range -> greater range of cylinder filling -> wider range of TDC cylinder pressure (without VCR) -> greater benefit from VCR.

You asked for comments, I'll bite. Every incarnation of VCR mechanisms adjust the compression ratio by changing the clearance volume,

where CR = (Vclearance + Vswept) / Vclearance

In the overwhelming majority of VCR incarnations, including the all the ones illustrated above, the change in compression ratio is achieved by altering the clearance volume via adjusting relative piston position at crank TDC.

It is well known that even minute changes in the "quench" clearance drastically worsens the in-cylinder charge motion and flow turbulent kinetic energy. What this means is that the improved full-load knock resistance achieved from a numerically reduced compression ratio resulting from increasing the quench clearance is not proportional in comparison to if the geometric compression ratio was obtained though optimal shaping of the clearance volume in the first place. In simple terms, an engine with a calculated geometric CR of, for example, 9:1 achieved though increased quench clearance, has worse knock resistance than one with the same 9:1 CR but with combustion chamber geometry so designed and optimized from the outset.

This is the exact experience observed in a project in which I am involved for a highly-boosted SIDI turbo engine (>25 bar BMEP). While not equipped with VCR (although my employer is deeply involved with VCR), the geometric CR was reduced from 9.8 to 8.5:1 using a thicker head gasket (greater quench clearance) for the purpose of expediency. When proper CFD-optimized dished pistons with the same 8.5:1 CR minus the thicker head gasket were obtained and installed, we were able to run significantly greater ignition timing advance up to the knock-limit and also moved the 50% burn point and point of maximum cylinder pressure to a much more advantageous crank angle. Furthermore, we were able to maintain stoichiometric operation at full load to a higher RPM. Both effects manifested in a reduction of BSFC. Incidently, the 8.5:1 engine with compression ratio "done the right way" did not suffer in BSFC compared to even the original 9.8:1 concept at full-load.

If someone comes up with an engine mechanism to change the compression ratio by changing the BDC piston position (in effect changing the swept volume component of the CR equation above) while leaving the TDC piston position unaltered, that would REALLY have my attention. Mr. Pattakon, you have shown some pretty smart VCR ideas in past years; here is my challenge to you.

Otherwise, there is still significant resistance of the added cost/complexity/fuel consumption benefit trade-off preventing widespread OEM deployment of VCR in series production.

Altering the effective compression ratio by valve timing (e.g. Miller/Atkinson cycle) gives most of the benefits of VCR and is robust, proven, cost-effective technology.

By the way, the above concept with an eccentric crank journal is exactly new... Gomecsys (working with ProDrive) has a similar approach to yours, since both vary the eccentricity at the con rod big-end. In the scheme of generic VCR concepts (your own picture below again), it falls closest to "2-C". However, almost ALL of the illustrated concepts except "E" suffer from the same fundamental problem I've explained in my previous post above of changing the TDC piston position and altering the quench clearance.



IMO, a whole new class of VCR concepts that alters the BDC point but not the TDC remains shielded in secret development labs... Would you be so kind to give credit where it's due when it comes to the light of day.

FEV also has a Type "3-C" concept, although I'm aware that they have EVERY main type covered in various concepts:
http://www.fev.com/content/public/default.aspx?id=498

No it doesn't, the big Guys patent anything they think is even slightly worth probably trying to protect - maybe something there that they can't quite see themselves that they can scream is theirs later.

I love searching through patents and especially the Japanese have huge amounts of useless stuff patented. The amount of Toyota 2 strokes I have been looking at recently is amazing.

Actually some of the multi-rod-crank mechanisms (including the Ehrlich system I mentioned in a previous post) are capable of variable displacement and VCR, so therefore should be capable of VCR with constant clearance volume.

Interestingly our friend Feliks who posts here, has designed a secondary-piston-ported engine which would be capable of operating as a type "E" VCR when varying the valve timing.

Let's suppose we built your ideal VCR system, having constant Vclearance (some call it dead volume) and variable Compression Ratio (CR).

At 10:1 compression ratio the necessary Vswept1 volume (i.e. the volume swept by the piston during a stroke) is 9*Vclearance.
This is so because:
CR =10= (Vclearance + Vswept1) / Vclearance, i.e. 9* Vclearance= Vswept1.

At 19:1 compression ratio (for instance at light load operation) the necessary Vswept2 volume is 18*Vclearance, i.e. double than the Vswept1 volume.
This is so because:
CR =19= (Vclearance + Vswept2) / Vclearance, i.e. 18* Vclearance= Vswept2.
The only way to double the Vswept volume is by double piston stroke.
Having a 90mm piston stroke at CR=10:1, the necessary piston stroke (in order to increase the compression ratio to 19:1 keeping the Vclearance constant) is 2*90mm=180mm (imagine the face of a car maker the moment you propose to make the stroke of his engine twice as long, for the sake of a VCR with optimised combustion chamber).

At 7:1 compression ratio (for instance at heavy overboost operation for peak power) the necessary Vswept3 volume is 6*Vclearance, i.e. 2/3 of the Vswept1 volume.
This is so because:
CR =7= (Vclearance + Vswept3) / Vclearance, i.e. 6* Vclearance= Vswept3.
Having a 90mm piston stroke at CR=10:1, the necessary piston stroke in order to decrease the compression ratio to 7:1 is (2/3)*90mm=60mm.

Now think what your ideal VCR engine does:
When you need as much as possible air to be trapped into the cylinder, in order to get as much torque (and power) as possible, that very moment you decrease the piston stroke to 60mm, i.e. to 1/3 of the maximum available piston stroke of the engine (180mm at CR=19).
To increase the intake pressure (provided by the turbo) in order to compensate for the reduction of the Vswept, is worse nonsense.

Similarly, trying to operate the engine at light load (CR=19), the mean piston speed is three times higher than the mean piston speed at CR=7. This means that the friction is way higher.
This also means that the inertia loads become 3 times stronger, i.e. the vibrations will be unaffordable (provided the engine is well balanced at CR=7, because at low CRs is where the engine operates at high revs – full load).


Looked from another viewpoint:
Let's suppose we have the patcrank VCR and want to apply it to your engine that has, let say, CR=12 and "optimised combustion chamber" at that specific compression ratio.
We keep the combustion chamber unchanged. So, when the patcrank VCR engine operates at exactly 12:1 compression ratio, "the in-cylinder charge motion and flow turbulent kinetic energy" etc, etc, is exactly as in your conventional engine (without VCR).
When a different compression ratio is desirable, one can decide whether the advantages of the variable compression prevail to the "optimised combustion chamber". I.e. the variable compression just provides additional "modes" of operational to the engine to choose from and optimise the overall engine operation (other moments for peak power, other moments for flat torque, other moments for easy cranking, other moments for warming-up, other moments for good response, other moments for minimizing the emissions and so on).

To optimise a specific characteristic of an aeroplane engine (for instance the combustion chamber) is important because the engine will operate at 95% of its “life” at the condition for which the optimisation was done (for instance 3500 rpm, 80% open throttle, specific altitude).

But a car engine has to operate in continuously variable conditions of revs, load etc. The optimisation of any of the characteristics of the car engine cannot be as good at 1200 rpm and at 6,500 rpm, neither at 20% open throttle and at 90% open throttle. As the rest “subsystems” of the engine, similarly the combustion chamber has to be good in a wide range of conditions, i.e. it cannot be just "tuned" for specific conditions.
Let me guess: the effect of the trottle position (i.e. how much it is opened) on your "optimized combustion chamber" is heavier than the CR, but because you can do nothing for this, you simply accept it.

You worry about the "cost/complexity/ . . . " the patcrank VCR introduces. If you study it, you will realize that it adds less "cost/complexity/ . . ." than a "decent" VVA (read at Idle Valves in pattakon web site).
By the way, the VVA systems, like the fully variable Desmodromic VVA (or DVVA) in pattakon web site (independently variable valve duration and valve stroke, no valve springs at all etc), can improve “the in-cylinder charge motion and flow turbulent kinetic energy” and, more important, can keep this improvement in a wide range of revs and loads. Alone the DVVA improves things a lot, but cooperating with a good VCR (to keep optimised the compression ratio, too) is the best choice we have now.

Manousos

The scheme you refer to and the comparisson is of mce-5, with the difference that the three pattakon VCR have been added at the sides.

Quoted from KeyAdvantages of pattakon web site:

"Gomecsys VCR versus patcrank VCR
To change the compression ratio, Gomecsys VCR changes the phase of an eccentric ring interposed between the big end bearing of the connecting rod and the crankpin.
A gearwheel around the crankpin holds the eccentric ring. By some other gearwheels the phase of the eccentric ring, relative to the crankpin, is controlled, and so the compression ratio.
The robustness of the crankshaft degrades (multi-piece "assembled" crankshaft with cuts and passages for the gear wheels) as well as the supporting of the crankshaft on the crankcase (by omitting some main crankshaft bearings).
Adds balance shafts of 1st and 2nd order.
The 720 degrees cycle adds "new" inertia vibrations of half order.
The gearwheels on the crankpins generate and undergo heavy centrifugal forces that increase friction and flex the gearwheels.
Is incompatible with many typical cylinder arrangements.
Difficult to cope with high revving.
On the other hand, Gomecsys VCR provides over-expansion and 720 degrees cycle. It is also fast responding due to the small loads on the control mechanism.
patcrank VCR resembles to Gomecsys VCR in that it is also based on an eccentric ring interposed between the big end bearing of the connecting rod and the crankpin.
But patcrank VCR keeps the original single piece robust crankshaft and the main bearings as they were, is applicable in any engine arrangement, does not add vibrations and stresses, is rid of gearwheels, is rid of heavy (i.e. creating strong inertia loads and friction) quick moving parts.
All it takes is one secondary thin and lightweight crankshaft, one slim and lightweight secondary connecting rod per crankpin and a lightweight control frame that moves only when a different compression ratio is desirable."




And regarding FEV's VCR:

Quoted from KeyAdvantages of pattakon web site

"FEV VCR and the similar versus patcrank VCR[/b]
To change the compression ratio, FEV's VCR displaces the rotation axis of the crankshaft relative to the crankcase. The frame that holds the crankshaft needs to be very strong. The crankcase that holds the frame needs to be even stronger. The additional transmission between crankshaft and gearbox adds friction, noise and reliability issues. A system to keep synchronized the camshafts with the moving crankshaft is also necessary.
SAE awarded FEV's VCR for being a smart breakthrough.
patcrank VCR resembles to FEV's VCR in that it displaces the rotation axis of a crankshaft relative to the crankcase, too. However, instead of displacing the main crankshaft (that stays at its conventional position firmly connected to the flywheel and through the clutch to the gearbox), patcrank VCR displaces a thin and lightweight secondary crankshaft that carries a tiny part of the piston forces, making the response way faster. The control frame that holds the secondary crankshaft is proportionally thinner and lightweight. The sprockets and the timing belt (or chain) that synchronize the camshafts to the crankshaft remain the conventional."

FEV's VCR, with all its problems, was installed and succesfully tested in a VW, confirming the potential of the VCR technology.

Manousos

At what cost? There is and has been an amazing amount of odd and unique engines etc tested for viability and variable compression ideas have been around longer than variable valve timing and yet we have all manufacturers running VVT now so they have decided that that's viable.

Just last week Greg mentioned a "fleet" of Sarich 2 strokes, not one, a fleet - less is better any day of the week in the eye of a manufacturer.

Manousos, there is a small flaw in logic. There is absolutely no need to have a range of geometric CR adjustment from 7:1 to 19:1. SI engines attain peak fuel conversion efficiency at compression ratios between 14-16:1. Furthermore, with gasoline direct injection and variable valve timing (cam phaser in the simplest form), you can run obtain >20 bar BMEP in a forced-induction engine with minimal full-load mixture enrichment at compression ratios approaching 10:1. I know because I've done it. I take for granted that any future engines that might incorporate VCR will also incorporate GDI and VVT.

In reality, you only need a range of adjustment between 9.5-14:1 for a forced-induction SI engine to cover the optimum range of BSFC over a wide load and speed range, and maybe 11-14.5:1 for a naturally-aspirated engine, the respective lower limits being for knock control. And when one performs detailed simulation (validated with testing) and cost analyses, the implementation-cost-to-fuel-consumption-benefit ratio still trails behind other solutions (e.g., variable valve timing, cylinder deactivation, downsizing, full-load cooled EGR), etc.

To sacrifice everything for the sake of the "optimized combustion chamber" is not reasonable.

Even for the 9.5:1 to 14:1 compression ratio range you mention (which is narrow), an engine having at light load 2600cc capacity (compression ratio 14:1, constant dead volume) operates as a poor 1700cc engine at heavy load / high revs (compression ratio 9.5:1). I.e. your ideal VCR does exactly the opposite it has to do: at peak power it decreases the engine capacity. It is also the balancing issue (vibrations), etc, etc.

The analysis explains what I know and you know: such an approach is not good, not even theoretically.
And the analysis was based on the assumption that you can realize your ideal VCR by a compact, light, reliable and efficient mechanism.
In practice there is no mechanism capable to keep the TDC "constant" and to displace the BDC, keeping at the same time the rest characteristics of the piston motion and of the engine acceptably good (unless you have one).

I propose to get back to the patcrank VCR which is based on a compact, light, reliable and efficient mechanism.

In any case thank you because your reply was the only - until now - "strictly technical".

PS.
They locked me out of Eng-Tips forum and deleted my posts there.
I know you are in.
If you please, write to "JCReynolds79" (thread "Wankel Rotary Engine Geometry") that a program that generates the Wankel rotor "envelope" is prepared and is at http://www.pattakon.com/tempman/WANKELN.exe and the code at http://www.pattakon.com/tempman/WANKELN.txt If he has any questions, he can e-mail me at vva@pattakon.com .

Manousos

9.5:1 to 14:1 is not a narrow compression ratio range. Friction and heat losses increase with the compression ratio, so there is no benefit with very high compression ratios. Saabs SVC engine made use of a 8:1 to 14:1 ratio range, but that was also a port injected engine which adjusted the TDC volume by tilting the cylinders. That engine was functional, the test cars were even driven by journalists (the picture I posted above) and they had a patent on it. But in the end GM stopped the development for cost reasons. There are in other words, as explained by TDIMeister, cheaper ways to improve the fuel consumption. To be competitive you need to spend the money where it gives the biggest return in the form of lower fuel costs.

****************
Thanks GregLocock
****************

No.
When the engine operates at light load, let say 1/4 (common situation in urban cycle) the in-cylinder pressure and temp is low.
What the patcrank VCR does?
It simply displaces the piston stroke a little (let say 4mm) closer to the cylinder head to reduce the dead volume.
At these last 4mm of the stroke is where the cycle differs than the cycle without VCR.
You do spend a little more energy to compress the mixture at the last 4mm, but the mixture now burns as if it were in the engine running at heavy load: the flame propagation is fast and the pressure rises to values comparable to the pressures met at heavy load operation.
A greater part of fuel's thermal energy changes into mechanical energy, that means lower consumption, less emissions etc.
This is because the expansion of the mixture - after TDC - returns much more energy than the additional energy spend during the last 4mm of compression.


As for the SAAB SVC (it is the case "A" in the scheme of generic VCR concepts), this is one more proof of VCRs potential.

Quoted from pattakon web site, KeyAdvantages:

SAAB SVC versus pat-head VCR
To change the compression ratio, SAAB's SVC displaces (rocks) the cylinder head together with the cylinder block relative to the crankcase.
Its architecture (the cylinder head together with the cylinder block are pivotally mounted on the crankcase) generates heavy bending loads on the cylinder block and on the crankcase.
It spoils the shape of the combustion chamber for some compression ratios (the combustion chamber becomes narrow at one side and wide at the opposite side of the cylinder).
The kinematics and the inertia vibrations of the engine depend on the selected compression ratio.
It makes difficult the sealing (oil and noise) of the crankcase.
It makes difficult the connection of the exhaust system with the cylinder head.
It degrades the quality of operation of the engine by additional noise and vibrations.
The pat-head VCR resembles to SAAB's SVC in that it also displaces the cylinder head together with the cylinder block relative to the crankcase.
But the pat-head VCR displaces the cylinder head and the cylinder block linearly (i.e. parallel to themselves) and keeps untouched the kinematics and the balance of the engine, keeps the shape of the combustion chamber for all compression ratios, supports the forces in the proper way without creating heavy bending loads neither on the cylinder block nor on the crankcase, seals efficiently (oil and noise) the crankcase, allows conventional exhaust system conventionally secured on the cylinder head and maintains the operational quality (smoothness) of the engine.

SAAB's SVC (and Toyota's copy - they use two crankshafts at the two sides of the cylinder block instead of one of SAAB SVC) has a few hidden side effects. If you study their patent applications, after the original idea, they mostly deal with the reinforcement of the crankcase and of the cylinder block, in order to receive the heavy bending loads generated by the support of the cylinder block-cylinder head on the crankcase.
The pat-head VCR does more with less: it separates the heavy forces from the light ones and minimizes the bending loads.
Read the http://www.pattakon.com/pattakonVCR.htm and the http://www.pattakon.com/pattakonKeyAdv.htm.

Manousos

Whaaaat!! Who wrote this nonsense! The cylinder bores tilt along with the head.

You are right.
I was wrong about the shape of the combustion chamber of SAAB’s SVC.
Thank you for the correction.
Manousos

This:



is the continuously Variable Compression Ratio (VCR) version of this:



unconventional engine.

The angular displacement of the toroidal cavity about the center of the two concentric Cardan joints, controls the compression ratio.

For more : http://www.pattakon.com/pattakonRotary.htm

Manousos

You won't get burns similar to high load by increasing the compression ratio, and increasing the compression ratio WILL increase heat and friction losses and when those losses are greater than the gain seen due to the increase in compression ratio, there is no point to increase the compression ratio further.

There is also simpler and better methods to get around the low load operation you mention. Displacement on demand is one, the other is downsizing and turbocharging. It's also possible to handle low load operation using the Miller cycle with late or early intake valve closing. Can you control the intake valve lift profile it's also possible to get rid of the throttle as the engine torque can be controled by valve timing instead.

As you know J, torque reduction without throttling is also being achieved using direct injection. By producing a stratified charge the fuelling and therefore the torque can be reduced by up to 50% (maybe more in future?) without throttling.

This has the dual benefits (normally seen in diesels) of eliminating the pumping loss due to throttling, and maintaining full effective compression ratio at part load.

“You won't get burns similar to high load by increasing the compression ratio, and increasing the compression ratio WILL increase heat and friction losses and when those losses are greater than the gain seen due to the increase in compression ratio, there is no point to increase the compression ratio further.”

On the same reasoning, the increase of the compression ratio of an engine from 6:1 (typical value many years ago) to 12:1 “WILL increase heat and friction losses”; yet the fuel consumption of a modern engine having 12:1 compression ratio, is way lower.
Unless you have another explanation, this simply means that the inevitable “increase of heat and friction losses” (caused by the increased compression ratio as you mention) is way lower than the increase of the mechanical energy provided on the pistons.
This also means that the point you mention “where those losses are greater than the gain seen due to the increase in compression ratio” is away.
Please visit the Mce-5, Gomecsys, FEV, SAAB, Nissan web sites and read their lab and road test data concerning the gains of running at high compression ratio (just before knocking) the engine at partial loads. These gains are anything but marginal.

“There is also simpler and better methods to get around the low load operation you mention. Displacement on demand is one, the other is downsizing and turbocharging. It's also possible to handle low load operation using the Miller cycle with late or early intake valve closing. Can you control the intake valve lift profile it's also possible to get rid of the throttle as the engine torque can be controled by valve timing instead.”

These methods are not alternatives to VCR; they are complementary.
For instance, the pumping loss does decrease by proper control of the valve lift / valve timing / valve duration. Yet the VCR comes to change dramatically the thermal efficiency at partial loads.

VVAs and VCRs are useless for engines that most of the time work at constant load - revs, like an airplane or a boat engine, because their valve lift, valve duration, valve timing, compression ratio etc are factory optimised for their “single mode of operation”.
But for the car engines, where the load and the revs combinations are infinite, the engine has many reasons to need continuously variable VVA and VCR. Let alone the optimisation necessary for different fuel quality, for the transient operation like cold-start etc.
Alternatively, instead of optimising the breathing and the compression ratio, the typical car engine leaves many “safety margins” (like choosing an “acceptable compression ratio” for the various fuel qualities the car will meet, like choosing an “acceptable” breathing for good torque at low revs and acceptably good peak power, etc) and “pays” for them with more fuel and emissions.

Manousos

Primarily real benefits to variable compression would be complements from variable cam. At really low to medium engine operation, throttle, detonation would be far away so compression ratio can be pushed very high to obtain good fuel economy. It should be relative to how much air to compression ratio goes. At low to medium throttle, i think in most cases the combustion chamber pressure is pretty much negative at bottom dead center to probably even 90 degree towards TDC so pumping friction should not be so critical.

Better fuel?

The higher compress charge offers better combustion from closer particles so they become more effective for a given intake so less charge is required for the same performance. Petrol engine makers have resorted to direct injection to obtain higher level of compression through lower level detonation design combustion chambers. Generally it takes away the volatile stuff from danger (a very technical term). I read that Honda, the anti-diesel manufacturer jumping from petrol to electric(ironically produced a really good diesel engine with this philosophy), are researching homogeneous petrol combustion, a petrol engine without spark plug, self igniting. I googled it and found that they call it HCCI and the core to this is both, to obtain high compression for efficient burning and claiming homogeneous charge offer a high charge for the same volume although they found that the petrol charge is too unpredictable resulting in the need for a variable compression ratio method to control the point of combustion.

Courtesy of greencarcongress.com

Yamaha had hydraulic system on one of their 500cc two stroke race engine but it seemed to have vanished. I'll try and look for it.

The pumping loss is the energy per cycle the engine pays to suction the next charge. With the correct valve lift profile (i.e. valve lift, valve duration, valve timing) the pumping loss can be significantly reduced at partial loads. This is one of the things a good VVA (Variable Valve Actuation) system does.

After the suction cycle, it comes the compression cycle and the power cycle.

A better VVA has to do with these cycles too, because the turbulence, the swirl and finally the flame propagation rate (i.e. how fast the combustion is completed) depend on the way the intake valves open and close.
For instance, during idling the early closing of the intake valves (like in BMW's valvetronic and in FIAT's multiair) degrades the initial turbulence / swirl and provides time to the charge to "calm down" before the combustion at TDC.
Take the case of the electromagnetic idle valves described at http://www.pattakon.com/pattakonIdleValve.htm and at the end of the http://www.pattakon.com/DVA_files/pattakonVVAs.pps (6.5 MB). They can open the right time (for instance 90 degress before TDC) and close the right time (for instance 60 degrees before TDC) to minimize the pumping loss and furthermore to optimize the turbulence and the swirl (i.e. the flame propagation rate and the emission control) during combustion. It is the ideal application for electromagnetic valves, because the idle valves are lightweight, their stroke is short and they operate exclusively at low revs.

But the dominant factor for the compression and the power cycles is the compression ratio.
Increasing the compression ratio to the knock limit during the partial load operation of the spark ignition engine, the friction is a little increased by the increased pressure at the top end of the piston stroke, while the thermal efficiency is significantly improved.
This is what a VCR (variable compression ratio) system does.
It keeps permanently the engine near its instant knock limit, lowering fuel consumption and emissions.

The HCCI process needs a continuously variable VCR to control it.
The two mode VCR of HONDA (an application of their VTEC on the piston crown) is not adequate to control HCCI.

Better fuel? (Tony Matthews)

A VCR is adaptable to the fuel quality.
Many drivers, in order to protect their engines from destructive knocking, use premium unleaded (100+ octane).
Using the expensive fuel, instead of the normal cheap one, is, as regards the fuel cost per Km, like having a car with some 15 to 20% worse mileage.
In an engine equipped with a VCR, the compression ratio is adjusted just before knocking for the fuel into the fuel tank. The car uses the available fuel quality, without reliability issues, while the efficiency of the engine is optimized for the specific fuel quality, reducing substantially the cost per Km. I.e. a VCR reduces the running cost of an engine in two ways: it can use any fuel quality (i.e. it uses cheap fuel) and it consumes less fuel because of the improved thermal efficiency.

There is not conflict between a VVA and a VCR system.
On the contrary, to keep the surface of the combustion chamber as small as it gets (in order to reduce the thermal loss and improve the thermal efficiency), the VVA is necessary for a VCR and the opposite: at high compression ratios (i.e. partial load) the small valve lift prevents the valves from hitting on the piston crown, while at full load and high revs the small compression ratio protects the valves - that now have long stroke and more overlap - from hitting the piston (and all these without excessive valve pockets).

Besides the fuel efficiency and the emissions, it is also the power density.
A better VVA that can operate reliably at high revs (with long valve duration, high valve lift, extreme valve overlap and true free breathing) allows racing power density, with the VCR in supporting role.
The same VCR - VVA engine at normal life conditions (like urban cycle) operates as a smooth, green, decent family car engine.

Manousos

Here is another "packaging" of the patcrankVCR for V8 engines:



It has two small / independent "secondary cranks" (yellow), the one for the "left" four cylinders of the V8, the other for the rest cylinders.

There are also the "partial" gif animations:

http://www.pattakon.com/patcrank/patcrankV8CR.gif

http://www.pattakon.com/patcrank/patcrankV8highCR.gif

http://www.pattakon.com/patcrank/patcrankV8lowCR.gif

and the full-size and controllable "exe" animation:

http://www.pattakon.com/patcrank/patcrankV8.exe (11MB)

Manousos