SIEMENS PD PIEZO INJECTORS (S)

SIEMENS PD PIEZO INJECTOR

These injectors are used on all 2.0 PD TDi Passat engines. (Not 1.9 PD TDi engines)

The electrical connector on a Piezo injector faces upwards and not sideways like a Solenoid injector.

Other than this, Piezo and Solenoid injectors look similar, with the injector nozzle looking like a fat nail and the pump looking like a valve spring. These injectors are always held in by two bolts.

Piezo injector failure seems mostly due to an internal electrical fault in the Piezo valve. The internal piezo wiring seems to burn out and short to the casing. The Piezo valve voltage signals are 100 -200V. Rather than risk damage to the ECU when one “shorts”, the ECU switches them all off.

The engine will cut out instantly and will not restart.

A VCDS or similar scan may give a fault code of:

17448 P1040 - Supply fuel injector(s) A, electrical malfunction in circuit

18578 P2146 - Supply Voltage for Fuel Injector Group A, Open Circuit

18579 P2147 - Supply Voltage for Fuel Injector Group A, Short to Ground

18580 P2148 - Supply Voltage for Fuel Injector Group A, Short to B+

17447 P1039 - Supply fuel injector(s) A, shorted

How to find the dud injector.

Pull the connector off an injector and start the engine. If it starts and runs on three cylinders, you have found the dud injector. Pull off, test and replace connectors until you find the dud injector.

Testing an injector.

You can’t test an injector for function but you can check the Piezo valve wiring. As this is the bit that normally fails, it is worth checking.

The piezo socket on the injector has TWO pins.

Neither should be connected to the casing (Earth or “-“).

Set your multimeter to continuity or Ohms.

Place one probe on one connector pin and the other probe on the injector body.

The meter should read “1” and not buzz.

Repeat for the other connector pin.

E.g.

If the Piezo valve has failed the meter will buzz or give a reading. A failed injector will give a reading of 400 – 500 Ohms or less.

If you are new to multimeters, an Ohm reading of “1” is not less than 400 – 500.

“1” means infinity or super super high resistance, in other words, no connection.

You must test both connector pins.

Don’t check the resistance between the two pins. It won’t tell you anything.

It is possible to remove the piezo valve part of the injector, but it isn’t easy due to the position and size of the clamp nut.

E.g.

The injector is on the left and the Piezo valve is on the right. In the centre of the piezo valve you can see a small (5mm) circle. This is the plunger that operates the injector valve to control injection.

Enlarged the piezo valve looks like this;

And like this:

From the business end it looks like this:

The schematic below shows how the piezo crystal are arranged inside the piezo valve.

The central plunger has a stack of crystals behind it. The crystals are separated by conducting plates. Each crystal receives a current from the ECU. The current from the ECU makes the crystal expand.

The expansion is so small you need a stack of crystals to move the plunger less than 0.5mm.

It is not possible to see the crystals without destroying the valve as it is sealed at the factory.

The piezo crystal stack

The conductors and crystals are housed inside this plastic module. The connector wires can be pulled out. Each connector has two insulating rings.

The connecting wires.

Each connecting wire or pin has two insulating rings because the wires have to pass through the piezo body top and then through the piezo cage top to get into the crystal stack.

It is impossible to be certain but I believe that these insulators are what fails. On this example the two piezo body top insulators (1 bright blue, 1 black rubber o–rings.) look ok in the picture But the black o-ring for pin two looks and feels different as if oil has got onto it.

I believe this is where the current leak occurred, causing the 400 – 500 ohm resistance measured earlier. This may have been enough for the ECU to shut down the injectors.

I cannot prove this as I can't get a short across the o-ring with everything dismantled.

I don’t think the piezo crystal stack fails.

The crystal stack insulation seems far too good for this to happen from within the stack which is why I suspect the connector wire insulators.

DISMANTLING THE PD PIEZO INJECTOR.

Manuals often say that dismantling and repairing an object is not a DIY option.

In the case of these injectors this is the truth.

No spares are available.

Special tools are required.

The picture below shows the complete injector

Before dismantling the valve you need to lift out the lever assembly from the valve body.

If you don’t, it will fall out and you will lose all the bits.

In the picture above I have already taken it out.

These are the bits that make up the lever assembly.

The top lever guide lifts out first with the top lever.

Underneath is the lever plate and bottom lever. These lift out.

A very small guide pin can be pulled out of its hole.

The needle valve can be lifted out.

If you turn the valve body over there is a torx bolt which can be removed to reveal the needle valve spring.

These bits are all shown in the picture below.

The lever assembly is needed because the piezo valve expands less than 0.5 mm and that’s not enough to move the needle valve. The two little levers magnify the expansion enough to move the needle valve.

When removing the pump assembly the top plunger just falls off.

The second plunger pulls out.

The pump spring assembly pulls off. (It is a tight fit)

E.g.

This is the injector with the pump assembly removed.

At this stage I found that both collar nuts are a non standard size and are very very tight.

I clamped the valve body in a vice and first loosened the bottom collar nut and then loosened the top collar nut without taking either of them off.

The following pictures don’t always show the exact order I worked in as it was easier to take some pictures at a later stage.

Ignore the fact that the valve body is missing in the following pictures. I started dismantling from the nozzle end.

I next removed the bottom collar which allowed the nozzle and nozzle needle to be removed.

A spacer was in between the nozzle spring chamber and the nozzle assembly.

The spacer has two alignment pins through it and a central hole for the nozzle needle plunger.

The nozzle assembly looks like this;

And like this with the collar nut removed

This is a view showing the spacer, alignment pins and plunger. The nozzle needle was pulled out of the nozzle for the picture.

The nozzle spring chamber was left looking like this.

It has two metal collars with holes and two o-ring seals.

The o-ring seals stop fuel from leaking.

The lower collar is for the fuel feed to the injector and the upper collar is for fuel return.

I pulled off the fuel feed collar and bottom o-ring. (It has stripes)

Then I pulled off the fuel return colour and o-ring. (no stripes)

It was at this stage that I removed the top collar nut to reveal the assembly below.

The valve body and nozzle spring chamber just pulled apart at the joint.

Pulling the valve body and nozzle spring chamber apart revealed a spacer with two alignment pins.

The top of the spacer and the top of the nozzle spring chamber both had o-rings in little grooves.

You can just see the black o-rings in the picture below.

At this stage the whole injector was in pieces apart from the nozzle spring chamber.

I couldn’t encourage the spring and plunger in the chamber to come out for photographing so I left them inside.

The complete disassembled injector looked like this;

The bits are more or less in order.

On stripping this injector I did not find any wear or damage which again suggests that it is the piezo part that failed.

The way the injector works is very complicated but boils down to a few basic functions in terms of the parts.

Pumped fuel enters the injector via the fuel feed collar and excess can return to the tank via the fuel return collar.

The camshaft follower presses the plunger at the top to pressurise the fuel in the injector.

The piezo valve controls the release of this high pressure fuel through the injector nozzle into the combustion chamber.

Here the fuel goes bang.

Without an electronic valve the fuel would pressurise and squirt into the combustion chamber.

Control of timing, volume etc would be very poor.

With a solenoid valve the timing, volume etc can be controlled fairly accurately.

With a piezo valve the timing, volume etc can be controlled more accurately.

The piezo valve can open and close so fast it is possible to have a variable number of injections from one charge of fuel. This greatly benefits fuel economy and pollution control.

IS A PIEZO REPAIR POSSIBLE ?

Please note the following is pure speculation and may not be possible.

I am not suggesting you try this on your car. – If you think this idea is bonkers/impossible, let me know and I will remove it.

It will only possibly work if the problem is a short circuit with one of the piezo connections and even then it may not work.

Don’t bother to try unless you know, one of your pins is shorted to earth.

My idea is based on this picture.

I destroyed the connector to have a good look at the construction.

As you can see the plug end of the connector, passes through the body and ends in TWO V-shaped plates.

You can’t see these plates without destroying the connector.

The connector plate has a V-shaped end which pinches onto the pin that goes into the Piezo unit.

It was seeing this that gave ne an Idea. I have no idea if this will work or should even be attempted.

This is the MAD idea.

The inside of the cam space will need protecting from plastic swarf to stop bits getting in the oil.

There may not be enough room to even try this.

Use a Dremel or similar or even a hand file to grind away the back of the piezo valve connector until the conductor strips and wires are exposed.

See Picture;

THIS PICTURE IS A FAKE TO SHOW WHAT I THINK MIGHT HAPPEN

I MADE THE PICTURE TO SHOW WHAT I THINK MIGHT BE POSSIBLE.

With the rear of the connector ground away you will first see TWO forked connector strips about 2 mm wide.

The strips will need bending up out of the way.

More plastic will need removing to expose the Piezo case.

If the wires get damaged it’s game over.

If you make it to the piezo valve metal case you will see two wires sticking out that pass through TINY rubber grommets. (one BLACK and one BLUE.)

These will need prising out and cleaning. (This is where you cry because you have dropped one and lost it).

Assuming you don’t lose the grommets, they can be dried and put back.

Use a multimeter to test the resistance between each wire and the body of the injector.

If both readings stay at “1” or “no buzz” you may be ok to continue

The forked connector strips will need bending back into place and pinched onto the wires again.

DON’T LET ANYTHING TOUCH THE PIEZO CASE

Use a multimeter to make sure you have good connections.

If you have managed to do all of this, this is where it becomes total guess work.

I am assuming that ARALDITE two pack glue will be oil resistant and a good insulator.

This may not be the case.

My idea relies on the theory that you can cover all the mess you have made, including the wires and joins with ARALDITE or similar.

I would have several goes to build up a thick covering.

When everything is dry/solid, you can test the resistance between each pin and earth as before.

If the connectors now seem not to short to earth you can try reconnecting the injector loom and starting up.

If it starts you are a genius and if it doesn’t I am an idiot for thinking it might.

Injector courtesy of FAT CONTROLLER