8V TDI common rail - interested?

So it is probably safe to say that there is not an OEM dual-rail system for a 4-cylinder engine, yeah?

Just wondering how one would fabricate a dual rail. I mean, that is a bit of an important part having to deal with 1500+ bar of pressure....

Brendan

Maybe we need a hpfp with a variable stroke swash plate like the ac compressor. That would have a smooth output regardless of flow.

What about non-piston-based designs, for that matter? Or are pistons the only way to reasonably get these pressures?

Piston types can have incredibly tight clearances, so leakage is minimal. Any other pump eg gear, vane, trochoid etc are hard to machine with that minimal internal leakage.

Ski in NC said:

Maybe we need a hpfp with a variable stroke swash plate like the ac compressor. That would have a smooth output regardless of flow.

Click to expand...

ALL HPFP's are variable displacement and variable pressure, this goes beyond a swash plate design in that it is not limited to a constant pressure.

There is a quantity controller that cycles at varying rates that determines the quantity of fuel being compressed. This sets up an initial pressure and an initial quantity during the rotation of the pump.

This reduces the load imposed on the motor increasing the efficiency of the system by only displacing the quantity of fuel required to support the load conditions (fuel consumption) at that moment.

If you step on the peddle, the slight reservoir of fuel in the rail permits an instant response as the pump displacement and pressure are increased to sustain or further increase the output of the engine.

Typical HPFP's operate at 500 bar most of the time. During acceleration and certain conditions only will it increase the rail pressure to the upper threshold limit.

Because the system spends 90%+ of the time in this low pressure regime the rails pressure is harder to maintain.

The other issue is that because of the pressures in the system, higher average rpms are required to negate the pumping losses. Despite the DLC coatings and close tolerance plungers there is still a significant loss across the plungers bore. Higher rpm's reduce these losses by virtue of reduced time at pressure and a secondary benefit is the variable pressure output of the system.

A swash plate would in theory turn slower, but this would increase the total leakage across the system, increase load on the motor, increase complexity and for the most part impart significant losses overall that cannot be justified.

Swash plate designs are typically limited to low to mid pressure ranges (3,000-6,000 psi) where higher volume displacement is required i.e. Hydraulic systems with large actuators.

Lord_Verminaard said:

So it is probably safe to say that there is not an OEM dual-rail system for a 4-cylinder engine, yeah?

Just wondering how one would fabricate a dual rail. I mean, that is a bit of an important part having to deal with 1500+ bar of pressure....

Brendan

Click to expand...

A dual rail could be accomplished using two OEM 4 cylinder TDI rails but combined with a single triple plunger HPFP.

The tuner could then build their own pressure control profile with a lower total pressure since there is a greater volume of fuel available at the lower rpm ranges.

Tapping two injectors from Rail 1 and the remaining injectors from Rail 2 would provide a larger reservoir of fuel under pressure as well as permit a more easily regulated system pressure with less reliance on the rails pressure regulating valve.

The two rails are simply tied together using a common injector line and fitting and could in practice be tied to one another on a common support running parallel to the head.

If a swash plate is desired, stick with a VP37 as that's exactly what it is with only semantic differences. Swash plate = cam plate.

TDIMeister said:

If a swash plate is desired, stick with a VP37 as that's exactly what it is with only semantic differences. Swash plate = cam plate.

Click to expand...

What I meant was a variable stroke swash plate. The vp37 cam plate is fixed stroke.

Present hpfp's are fixed stroke with intake throttling for flow control. Pressure spikes would be smoother with non-throttled inlet yet variable stroke.

Not suggesting this is practical, just throwing ideas out there....

In the interests of accuracy, I must point out that not all CR pumps are really "variable displacement". More recent pumps may have an inlet throttle as a quantity adjuster that acts roughly as Pete describes.
However there are many many (millions, at a WAG) vehicles using a three-cylinder pump with NO inlet flow control. The only way to decrease flow is by disabling one of the pump heads. This is unthrottled, and kills the head completely. In normal driving infact, the pump will be driven as 2-cyl most of the time, and the pressure-control valve has no difficulty in tightly controlling rail pressure regardless of pressure pulses or losses...

Ski in NC said:

What I meant was a variable stroke swash plate. The vp37 cam plate is fixed stroke.

Click to expand...

It is in fact "effectively" variable stroke through the quantity adjuster collar in that the extent of the plunger stroke before it spills its load varies.

DPM said:

In the interests of accuracy, I must point out that not all CR pumps are really "variable displacement". More recent pumps may have an inlet throttle as a quantity adjuster that acts roughly as Pete describes.
However there are many many (millions, at a WAG) vehicles using a three-cylinder pump with NO inlet flow control. The only way to decrease flow is by disabling one of the pump heads. This is unthrottled, and kills the head completely. In normal driving infact, the pump will be driven as 2-cyl most of the time, and the pressure-control valve has no difficulty in tightly controlling rail pressure regardless of pressure pulses or losses...

Click to expand...

The problem here is you would be required to utilize the older Solenoid actuated injectors not the fast acting Piezo.

The older pumps ran lower pressures not the newer 1600-1800 bar.

The Piezo is much simpler and easier to tune (aside from IMA coding) but there in lies the issue of regulating the pressures at a very high degree of accuracy else you throw off the quantity regulation of the injector (7 injectors per combustion stroke capability).

I am simply of the opinion that using a newer style pump (single or triple) with Piezo is the way to go.

There are now fast-acting solenoid injectors, FWIW - they're being used on the Passat now.

In any case, on an ALH, do you really need 7 injection events? Two of those events will be completely wasted, after all.

Why not using a simple CP1 pump?
It's the simplest CR pump in the market and can obtain 1800-2000 bar easily.
With modded shaft (7,5mm bore and 7mm stroke) this pump can obtain 175cm3 /200 shots, more than most of CP3-CP4 in the market.
I've start a project like this 3 years ago; EDC16C8 0281011511 (JTD) and the relative wiring/sensor but not much time...
Sorry for my bad english: Italian school do it worse

Ah bhtooefr is right! the last solenoid injectors can work as fast as Piezo.
The high pressure sealing ball has been changed with a valve.

TdiDream: Because the CP1 isn't common here, whereas the CP3 is. You can get aftermarket upgraded CP3s quite easily here.

bhtooefr said:

There are now fast-acting solenoid injectors, FWIW - they're being used on the Passat now.
In any case, on an ALH, do you really need 7 injection events? Two of those events will be completely wasted, after all.

Click to expand...

Not sure where you are getting your information but it's 100% wrong.

I have been working with the CRI systems since 2006 with both Solenoid and Piezo. It's also an advantage that 2 of my vehicles are Piezo systems and the other Solenoid and have the diagnostics to fully analyze either of those systems.

Since you will probably want a citation....HERE IT IS!

VW said:

In the common rail system of the 2.0 Liter TDI
engine, piezo-controlled Fuel Injectors N30, N31,
N32, and N33 are used.
The fuel injectors are controlled over a piezo
actuator.

The switching speed of a piezo actuator is
approximately four times faster than a solenoid valve.
Compared to solenoid actuated fuel injectors, piezo
technology also involves approximately 75% less
moving mass at the nozzle pin.
This results in the following advantages:
-Very short switching times
-Multiple injections possible per work cycle
-Precise metering of injection quantities

Click to expand...

If you know of a Solenoid that is as fast as Piezo I am sure Bosch would love to hear about it.

what injectors are in the new US Passat, Pete?

US passat= solenoid. Jetta gets piezo. Jetta might need faster injectors for its emissions system. Maybe they found the solenoids were "good enough" and cheaper too??

Alfa Mito 1.6 MultiJet use a Bosch solenoid tipe injector that is as fast as Piezo.
The limit of the solenoid type is the sphere/cone.
Change this in a valve, can reduce shutter stroke to 20um and You can obtain 8 injection per stroke.

Revisit the VR6 TDI?

CP3 at crank speed should time the injection events like DBW is pushing

Go for it Jon! That's your baby.

You could make a hpfp out of a ve pump. Cap off all four fuel outlets. Machine out plunger barrel center screw hole (where dial ind goes) and install a delivery valve there. Then each stroke of plunger pumps into that center delivery valve. Adjust flow rate with qa and plunger collar.

Don't know what pressure it would go to before it eats its guts, but that outrageous cr pressure is probably only needed for emissions. Normal pressures would do fine with some bigger nozzle holes??? Thinking aloud...

Some of the guys who want to go on their own CR systems might explore this possibility. The QA could then use rail pressure for the feedback loop.

Drivbiwire said:

If you know of a Solenoid that is as fast as Piezo I am sure Bosch would love to hear about it.

Click to expand...

Bosch has already heard of it.

http://www.mtz-worldwide.com/index....3jgnlt1&smart42SID=n1src9jtuc9ag5ktflo3jgnlt1

Hence, the injection rate map is comparable to that of an 1800 bar CRI3.2 piezo injector.

Click to expand...

TDIMeister said:

Some of the guys who want to go on their own CR systems might explore this possibility. The QA could then use rail pressure for the feedback loop.

Click to expand...

In fact, Bosch uses modded inline pumps and single-cylinder jerk pumps as their developing countries (read: crap fuel tolerant) HPFPs.

Good, get to it guys.

In fact, I feel so generous that I'll offer this: Extending Ski's idea, instead of even touching the VE pump, just have all 4 existing lines go to one rail. Then a cross over line to second rail that connects to each injector (thanks, Pete).

just make sure one of the rails has a pressure control valve...

For the VR6 TDI, I think I'd recommend basing it on the 24v VR6, so that you get roller rockers. I'd think that would reduce the valvetrain stresses that other VAG diesels have experienced.

And, I think that'll make injectors easier to handle, because it looks like the spark plugs go in the head at an angle (and straight into the cylinders), so I think you could use normal 4v/cyl injectors? Not sure what the spark plug angle is on the 12v VRs (their setup looks weird and offset), but if it's the same as the angle on a 8v VW I4, then you'll be able to use normal VW 2v/cyl injectors, I think. (You'd also need to find a way to hold them down, though.)