Failed MDS lifter autopsy

Disclaimer: Links on this page pointing to Amazon, eBay and other sites may include affiliate code. If you click them and make a purchase, we may earn a small commission.

Xsen

Member
Joined
Oct 25, 2021
Posts
80
Reaction score
180
Location
Urals, Russia
Ram Year
2014
Engine
Hemi 5.7
Hey everyone,

Last week I had a bit of spare time on my hands, and my gaze fell on the box with the failed lifters that I took out of my 2014 5.7. So I thought why not try and disassemble the failed one to see what had started the whole adventure. I've learned quite a few things out of it, I think, they're worth sharing. I haven't seen similar info on the forum, but if exists, maybe it would be a good idea to merge the threads.

Oh, and I also have a similar description of the hemi lubrication system with pictures and labels, so I guess I'll post that later.

Buckle up, this will be a long one. :D

Here are all the components of the MDS lifter in an approxiomately correct positioning:

Lifter_overview_v2.png

On the very top of the lifter there are two snap rings (1) and (2) that hold the internal assembly in place. Underneath the snap rings is a precisely machined cartridge, that holds the lower half of the hydraulic part (4), that rests against the bottom of the cartridge on a small spring (5). The top half of the hydraulic part (3) resides inside the cartridge (6) as well. When the engine is running, the oil is supplied to the orifice in (3) via the oiling channel in the pushrod and fills the chamber inside (4) and (3), thus adjusting the lash.

EDIT: I have messed up a little here, so adding an impotant detail. At the bottom of the lower half (4) there's a spinged ball valve, that allows the oil to escape the chamber, as more oil is supplied.
Lifter_valve.png

This valve can also become a failure point, i.e. if it is stuck open the pressure in the chamber would not be sufficient for lash control, leading to the ticking noise at the valvetrain frequency.

If we were looking at an ordinary lifter that would be it, however this contraption is a bit more sophisticated.

By the way, the lifter never gets the full oil pressure that is developed by the oil pump, as there is a throttling orifice in the head gasket that limits the amount of oil that can get up into the cylinder head, plus the oil gets into the pushrod via the rocker arm, and mostly when the rocker is in a specific position, which allows overlap between the orifices on the rocker and on the rocker shaft, further limiting the amount of oil in the pushrod.

Back to the autopsy.

Let's look at the cartridge (6) in more detail.

Part6_overview.png

In the lower part of the catridge (6) there's an orifice that hosts two pins (6A), that are pushed out by the spring (6B). The pins' travel is limited by a wire ring (6C), and this ring will play a role further.

Inside the lifter body (8) there's a groove inside that the pins stick into when the engine is running on all 8 cylinders and the lifter does open the valve. Simply put, in regular operation the cartridge (6) is locked with the lifter body (8).

Part8_inside.png

At the bottom of the lifter is a roller (11) that runs on a needle bearing (10) and the axle (9). We'll look into those further on.

When the ECM decides to switch to V4 mode, the MDS solenoid opens the oil flow into the oil galley in the block that runs through the lifter bores. There are four galleys in the block, one per each block corner. It's important to remember that each galley unites the four lifter bores that correspond to the neighboring cylinders (1-3, 5-7, 2-4 and 6-8), one of which can be disabled (the magic hemi sequence of 1-4-6-7) and the other one is always working. The cylinders are shut off after (or rather during) the power stroke, so there is no exhaust event and all the gases stay in the cylinder working as a pneumatic spring, though of course their cyclic work is slightly negative due to friction losses.

When the solenoid is open the oil from the galley flows into the lifter bore and pushes the pins (6A) into the cartridge body (6), thus breaking the lock between the cartridge (6) and the lifter body (8), and enabling the lifter body (8) to move up and down in relation to the cartridge (6), which is held in place by the valve spring (via the rocker arm & the pushrod). To prevent lifter bouncing on the cam lobe, under the cartridge (6) is a spring pack (7), that maintains some of the push on the lifter body (8) and therefore on the roller (11).

When the ECM decides to go back to normal operation, the MDS solenoid cuts the oil flow to the galley almost completely, making the oil pressure inside the lifter bores drop. Therefore the oil pressure inside the lifter body (8) also drops, the spring (6B) pushes the pins (6A) out and they lock in the groove on the lifter body (8), establishing the link between the cam lobe and the valve stem again.

Here are two intermediate conclusions:
- All lifter bores are oiled well only during the MDS Active phase (V4 mode). Only during this phase there's additional oil flowing from the bores to help lubricate the rollers, their bearings, etc. This was very well illustrated in one of the Hemi Garage Hack's videos on YouTube. In all other times the amount of oil in the galleys and lifter bores is minimal (although not zero).
- Additionally the MDS lifter body (8) has an orifice at the bottom that allows the oil, that has entered the lifter body (8) to escape, providing lubrication to the roller.

photo_2026-11-09_21-27-48.jpg

This is it for the autopsy. Now on to my failed engine.

It had developed a very strong and pronounced tick in a sudden manner. This tick would stop whenever the MDS activated, so it was obvious that the root cause was related to one of the MDS lifters, but only during the lifter disassembly I've managed to find it. Looking into the side orifices on the failed lifter I saw a bit of the metal wire ring (6C) that was already out of the groove on the cartridge (6) body. So at some point, maybe during cylinder reactivation, the wire ring (6C) broke free from the groove on the cartridge (6) and got caught between the cartridge (6) and the lifter body (8). This locked the cartridge at some intermediate position, letting it open the valve enough to not have misfires during idle and light load.

Aside from that there was a very light tick on a hot engine, but this one was standard for the hemi - the roller (11) developed some radial play due to the axle (9) wear:

photo_2026-11-09_21-27-43.jpg

This damage is on the loaded side of the axle, that takes the pressure during the valve opening events. I guess that's what you get with poor lubrication.

My final conclusions:

1. The MDS lifter can die in multiple ways, and not all of them might be related to the insufficient roller lubrication. As far as I know, the case described above is relatively rare, which is a good thing.
2. This exact type of failure can be avoided by turning the MDS off, however this would also worsen the lubrication conditions for all lifters/cam lobes, so it's still very not recommended.

If I were rebuilding the engine now, I would have gone with the aftermarket lifters that have the roller axle oiling channels, and would eliminate the MDS entirely. But it's a bit too late, the truck is already back on the road, so we'll see how it goes.
 

Attachments

  • Lifter_overview.png
    Lifter_overview.png
    1.2 MB · Views: 4
Last edited:

Ken226

Senior Member
Joined
Mar 12, 2017
Posts
2,388
Reaction score
6,309
Location
NE Oklahoma (for now).
Ram Year
2013
Engine
Hemi
Excellent writeup. Well written and very descriptive.

Thanks for posting it.


Just to add a little to your description of operation, specifically regarding this point:
1794926112963.png

From an article I read that Chrysler published back when the 2009 Hemi got MDS and VVT first came out. I remember reading it while sitting in the waiting room at a Dodge Dealership waiting on an oil change and new tires.


I've tried to find the article since, but havn't been able. I remember the following though.

The camshaft has a different profile for the lobes that MDS lifters ride on. The lobe has a slightly elevated profile at the beginning of the lobe, that lifts the lifter, just a little early to provide a little additional dwell time, so that when the PCM shuts off pressure to the MDS lifters, the pins have a little more time to seat before the lifter starts moving again. The PCM shuts off oil, then that little extra ramp profile gently lifts the lifter body enough that the pins engage, right before the main part of the lobe hits the roller.

Also, the PCM does a little trickery by doing something to the fueling and unlocking the torque converter for a small duration of time, to coincide with MDS reengagement, so the whole system works together to keep the system working.
 
OP
OP
Xsen

Xsen

Member
Joined
Oct 25, 2021
Posts
80
Reaction score
180
Location
Urals, Russia
Ram Year
2014
Engine
Hemi 5.7
Thanks!

A very good point there, indeed the lobes on the MDS camshaft have bit of a different profile for the MDS lifters, to assist with the reengagement, and as far as I know this is precisely why using the MDS cam with non-MDS lifters may result in sloppy idle and vibrations - this little ramp would cause the valves to be open a little longer, when the airflow is small it would affect the torque management during idle. Nothing that can't be fixed by making changes to the ECM though.

As for the torque converter - aside from unlocking it during return to V8 mode, it most probably operates in controlled slip mode while being in V4 mode, to smooth out the torque pulsation. I base this on the TCC control algorithms used in GM's 6L80 transmission, for which this is a proven fact, and also because neither GM nor FCA would offer cylinder deactivation with manual transmissions. And this is one more thing that makes me dislike MDS/AFM/DoD technologies - I just can't imagine controlled slip being a healthy mode for the TC clutch. Oh, it will hold out for long, but still - this introduces unneeded wear.

Then again I'm not a transmission engineer, so I could be wrong.

EDIT: I've looked into a log file, that I've recorded a bit ago, and specifically at the points when the MDS has been re-engaged - there is no change in TCC lockup solenoid current, so maybe in fact it does not unlock. ZF sure likes its TCC applied. But what I do see is that when the MDS cylinders were starting to fire again the injector timing on cylinders 1-4-6-7 was +25% against the others. Probably to make their restart easier.
 
Last edited:

Forum statistics

Threads
213,354
Posts
3,092,143
Members
174,575
Latest member
DITRIO
Back
Top