At some point just idling it caused it to die.
There were more subtile problems before that, but this was the final of it essentially.
Engine is a 4d56 2.5l turbo diesel of a 4WD GLX, that has been driven 80.000km so far and is a late 2012 model.
I don't have a replacement EGR valve yet to put in and see if it really is all there is to solve my engine problem (ordered, takes a little) but what I found was basically that it's a design failure which causes the upper ball bearing to seize (the lower one was fine).
The motor in there is a DC motor with brushes, that seems to wear a lot (fine grey/black dust with copper particles are being produced). The motor drives a threaded rod that moves in/out depending on the turning direction of the motor. The whole thing is sort of a servo, with the motor pushing the rod out and a spring pushing it back in and the speed of it depends on the ability of the motor to spin, which depends on the ball bearings to work.
I disassembled the valve and by now it doesn't resemble it anymore as parts have been altered, so I don't have a more 'complete' image of how it looks like, sorry.
The upper image shows essentially the 'top' of the motor in the valve with the copper rings to get the currents to switch direction. There is 5 brushes in there (3 in the image), with the two others (not shown) riding on the outside slotted copper ring, which causes the switching of the current direction to make it all work.
where to host photos
Turning this part around (lower image) you see the top ball bearing that sits underneath it when everything was assembled. It's metal seal shield (it's a 6802Z) was covered with black/grey, copper particle dotted thick grease.
Upon removing the seal the same thick stuff was found INSIDE the ball bearing.
It really didn't want to turn anymore.
You can also see the small brushes on the springs that are worn. The one that has the shortest spring is worn the most, and it's actually just half as high now as the one in the center. At some point this will fail in any case.
Anyhow, the whole motor is sealed - airtight - and the only thing going in/out of the whole valve motor is the threaded rod at the bottom that drives the actual valve. This rod slides against some rubber seal that is greased, so air can't move by there (all down there was also very clean, no exhaust soot etc. visible, so the returned exhaust gas is not the culprit).
Internally in the valve the rod does work like a cylinder compressing/expanding space which means that the air inside has an incentive to move.
And what do you know, the piece that sits between the 'chamber' where above brushes sit (which produce some sort of fine dust) and the 'chamber' the rod travels in is the upper ball bearing.
The grease/oil inside it 'filtered' the dust in the air that was moving between the two chambers and turned the grease into a seizing paste.
As the valve now is slower to react to the PWM control of the ECU, the ECU will become very irritated to what is happening with the exhaust gas re-circulation it sets and how the engine reacts.
This is not serviceable. The EGR valve housing is plastic that has been sealed by ultrasonic welding and it essentially breaks when you try to open it.
Replacement motors on Aliexpress are ~150USD.
Anyone had their EGR valve replaced?
Did you pay for it?
I'm pretty sure that this is a screw up on the side of Mitsubishi (or the OEM that makes the valve) and they should be liable, but I'm not a lawyer.
I think the remediation would be to have a small hole in the external wall of the upper chamber (and probably one in the lower chamber) to not cause airflow over the upper ball bearing, that causes dust to get trapped in it's grease to avoid seizing.
PS: the valve motor will use power, even if you block the exhaust gas re-circulation from actually happening. It's a pretty strong motor, so a lot of wasted electricity.
