Powershift Primary Oil Filter Replacement

[AlanM]

A really poor bit of design: The primary oil filter in the Getrag 6DCT450 (officially) cannot be replaced without removing and completely dismantling the transmission. Even worse, with the valve body cover and the valve body itself removed, you can see and get hold of the filter.

But it can be done. All you need is a bit of determination and a heat gun. The torx screw that retains the filter is where the determination comes in. It's quite tricky to access, you can only move it a fraction of a turn at a time, and it is RIDICULOUS how long it is, given the job it does.

But eventually it'll come out.

Then, to get the old filter out you just remove the magnets, which gives a bit more clearance, and use the heat gun to soften the plastic top half of the filter a bit. Then push it down while levering it out. You have to deform the top half by enough to get the spigot out of the oil pump inlet. It's actually not much, maybe about 10mm.

Getting the new filter in is pretty much the reverse process. A bit of heat (I had the gun set to 150C) to allow the filter top to flex without cracking and a blunt lever, (I used the handle of a small hammer) you push the top of the filter down and in, getting the spigot under the oil pump pick up.

Once it's under you can push it back and forth until the spigot pops up into the pick up. Then a couple of screwdrivers under the filter to push the spigot all the way in.

Now comes the hard part: getting that ridiculously long screw back in. I actually ground the tip of the screw into a bit of a point, since there's no way of seeing when the screw hole in the filter is aligned with the hole in the oil pump pick up. Grinding a point helps it to align itself.

Then it's just another 45 minutes screwing it in a 16th of a turn at a time... it probably wasn't that long but it seemed longer.

So anyway, it can be done, and a lot quicker and less work than removing and splitting the transmission.

[oldgun]

the question is , do you have to replace that main filter which is only for big chunk of particles.

[AlanM]

Originally Posted by oldgun

the question is , do you have to replace that main filter which is only for big chunk of particles.

It's a paper filter, not a strainer, so it would get at least partially blocked by just the material that comes off the clutch plates. The magnets in these boxes get a thick coating of sludge on them, and they certainly wouldn't be as effective once covered like that. So that leaves the other two filters to keep the muck out. But the external one is in line with the oil cooler so only actually works when the oil is hot. All the oil constantly circulates through the internal one so having a clean one in there certainly can't hurt.

These transmissions rely on oil pressure to properly engage the clutches. The internal filter is on the suction side of the oil pump so restrictions here could quite possibly impact on the operation.

[AlCan]

Sounds like a real mission.


Did you happen to take any photos while you had it apart?


I had no idea that filter was paper. I totally agree, really lousy design. No wonder Getrag changed their name. Such a shame, as some parts of the transmission are very smart. But it can't be allowed to last too long, right!?

[AlanM]

The hardest part is getting the screw out and then back in. There's not much access to it.

Actually getting the filter out and in isn't so bad, once you heat it up a bit to get a bit more flex.

The filter element itself is like a bag, with the opening at the intake hole, which effectively doubles the surface area.

Sorry, I didn't take any photos, but there is a YouTube video on a channel called "Life on octane".

[oldgun]

Managed to get the job don, took 3 hours with heat gun.

[AlCan]

Thanks Alan,

This is very helpful, especially the video pointer.

For ref, it's here: https://youtu.be/ULCgqmGZ9tY

Hand held camera which is very busy, but you do get a few glimpses of the paper element inside the plastic and steel filter box. Looks awfully thin!

Originally Posted by AlanM

The filter element itself is like a bag, with the opening at the intake hole, which effectively doubles the surface area.

Sounds very like a disposable vacuum cleaner bag.

I guess this construction explains why the transmission suffers mechatronic faults after 1-200,000km or so of not getting transmission servicing. Still guessing, but my guess is that once that paper bag gets thoroughly clogged and the pressure drop goes up, it tears open...

Hmmm. Is that external filter even in the oil flow path to the control components, or is it just a bypass filter? Maybe that's why the primary, paper bag filter is so much finer than a steel mesh or similar pickup would be.

Like it's designed to punish those who don't pay for regular transmission servicing.

So, many thanks for this post, I'm off to buy a primary trans filter asap!

Anyone have any idea of what the oil pressure is at the pump outlet? Maybe that plastic external filter housing wouldn't be up to it.

[AlCan]

Uhh... Can't edit last post, but I see now that the external filter is connected via low pressure hose and hose clips, so can't be much pressure there. Must be a bypass filter.

[AlCan]

Originally Posted by AlanM

The torx screw that retains the filter is where the determination comes in. It's quite tricky to access, you can only move it a fraction of a turn at a time, and it is RIDICULOUS how long it is, given the job it does.

Found a very interesting video on YouTube. It's from Latvia, and it's in Russian, with a few recognisable English terms thrown in. Even not knowing a word of Russian, it's quite enlightening and I learned a couple of Russian words along the way: Boltik and Plastika - which was the subject of quite a bit of discussion... The guy who does the teardown is obviously very familiar with these boxes, and spotted something abnormal right at the start. Only realised that, by following the transcript once translated into English. (I'll try to post the translation as an attachment in case anyone is interested.)

https://www.youtube.com/watch?v=j4PdTiTZ9y4&t=1000s (Link starts where the sump filter is being discussed.)

I was particularly keen to see if I could figure out how the oil maintains a higher level in the Mechatronics section than in the transmission housing, but it's not at all obvious (to me anyway). The Oil Pump picks up from the Sump Intake Filter in the Mechatronic section (which would tend to drain it), and presumably most of the fluid returns through exhaust ports on the Mechatronic slab (back into the Mech section). But I guess a certain amount goes into the Bell housing, via input shaft / clutch ring seals leakage) which I suppose ultimately connects to the transmission section. It seems to me, there must be a way that it keeps the Mech section topped up. Maybe the crown wheel acts as an oil slinger and "throws it over the wall"? I have no idea. Anyone else?

There are what seem to be pump scrolls cast into the bearing end housings but whether that's what does it via hollow shafts, I don't know.

Anyway, the sump filter and pump feature at about 17:00 into the video, and I see that the bolt that holds the sump filter spigot in place has a dual function, hence its length. It also helps bolt the oil pump into the housing, so I imagine it needs to be tightened back up pretty well.

My replacement sump filter / pickups arrived today, but they are much smaller than they look in all the videos I've seen. Maybe it's just the fact of viewing on a large screen, but mine, although looking the right shape etc, are only 192mm long (overall) x 137mm wide (overall). They are a weird shape - no edge is parallel to its opposite and only one corner is 90°.

AlanM, did you keep your old filter, and did you happen to cut it open, by any chance? I'm very curious about how it's constructed but would like to know before I pull mine out.

BTW, I'm now convinced that all the oil, pressure fed to the mech unit etc must pass through the sump pickup filter, so for bits of plastic to get in and block orifices and jam solenoid valves, there must be a breakdown of the filter element. Found a schematic for the full hydraulic circuit, produced by the ATSG, which I haven't found in the Tech Resources section, but it can be downloaded here:

https://godsskill.files.wordpress.co...rvice-menu.pdf

Enjoy!

[AlanM]

Originally Posted by AlCan

Found a very interesting video on YouTube. It's from Latvia, and it's in Russian, with a few recognisable English terms thrown in. Even not knowing a word of Russian, it's quite enlightening and I learned a couple of Russian words along the way: Boltik and Plastika - which was the subject of quite a bit of discussion... The guy who does the teardown is obviously very familiar with these boxes, and spotted something abnormal right at the start. Only realised that, by following the transcript once translated into English. (I'll try to post the translation as an attachment in case anyone is interested.)

https://www.youtube.com/watch?v=j4PdTiTZ9y4&t=1000s (Link starts where the sump filter is being discussed.)

I was particularly keen to see if I could figure out how the oil maintains a higher level in the Mechatronics section than in the transmission housing, but it's not at all obvious (to me anyway). The Oil Pump picks up from the Sump Intake Filter in the Mechatronic section (which would tend to drain it), and presumably most of the fluid returns through exhaust ports on the Mechatronic slab (back into the Mech section). But I guess a certain amount goes into the Bell housing, via input shaft / clutch ring seals leakage) which I suppose ultimately connects to the transmission section. It seems to me, there must be a way that it keeps the Mech section topped up. Maybe the crown wheel acts as an oil slinger and "throws it over the wall"? I have no idea. Anyone else?

There are what seem to be pump scrolls cast into the bearing end housings but whether that's what does it via hollow shafts, I don't know.

Anyway, the sump filter and pump feature at about 17:00 into the video, and I see that the bolt that holds the sump filter spigot in place has a dual function, hence its length. It also helps bolt the oil pump into the housing, so I imagine it needs to be tightened back up pretty well.

My replacement sump filter / pickups arrived today, but they are much smaller than they look in all the videos I've seen. Maybe it's just the fact of viewing on a large screen, but mine, although looking the right shape etc, are only 192mm long (overall) x 137mm wide (overall). They are a weird shape - no edge is parallel to its opposite and only one corner is 90°.

AlanM, did you keep your old filter, and did you happen to cut it open, by any chance? I'm very curious about how it's constructed but would like to know before I pull mine out.

BTW, I'm now convinced that all the oil, pressure fed to the mech unit etc must pass through the sump pickup filter, so for bits of plastic to get in and block orifices and jam solenoid valves, there must be a breakdown of the filter element. Found a schematic for the full hydraulic circuit, produced by the ATSG, which I haven't found in the Tech Resources section, but it can be downloaded here:

https://godsskill.files.wordpress.co...rvice-menu.pdf

Enjoy!

I don't know how either, but oil must get transferred back from the clutch and gearbox section to the mechatronic section for the plastic bits to get there. And they definitely do, I've seen them.

I didn't keep the old filter, but we did cut it open. It didn't seem to have any holes in it, but this car doesn't have the problems usually associated with the clutch spring retainers breaking up either. It was just heavily contaminated with thick black sludge.

Best of luck with your filter, as long as you're patient and persistent you'll get it done.

[AlanM]

Originally Posted by AlCan

BTW, I'm now convinced that all the oil, pressure fed to the mech unit etc must pass through the sump pickup filter, so for bits of plastic to get in and block orifices and jam solenoid valves, there must be a breakdown of the filter element. Found a schematic for the full hydraulic circuit, produced by the ATSG, which I haven't found in the Tech Resources section, but it can be downloaded here:

https://godsskill.files.wordpress.co...rvice-menu.pdf


Enjoy!

I asked a guy who reconditions these transmissions how the plastic gets past the filter. It's actually pretty simple. When you look at the front of the valve body there's quite a few ports. When the valves are moving they can vent oil out of these ports, but also suck some in. Because the plastic is suspended in the oil, the bits can get sucked right into the valves through these ports.

[AlanM]

Thought I should mention, something I've never seen on a video about working on these valve bodies, is to make sure you clean the solenoid contact pads on the TCM. They can get a varnish like coating on them, which can prevent the solenoids from getting current.

Quite often you'll see a pair of shiny dots on the contact pads, where the solenoid was, but once you disturb them, they may contact in a different spot.

[AlCan]

Originally Posted by AlanM

Thought I should mention, something I've never seen on a video about working on these valve bodies, is to make sure you clean the solenoid contact pads on the TCM. They can get a varnish like coating on them, which can prevent the solenoids from getting current. Quite often you'll see a pair of shiny dots on the contact pads, where the solenoid was, but once you disturb them, they may contact in a different spot.

Hi AlanM,

This is a very good point, thanks. I will try to keep it in mind.

Actually, I don't know how I missed these posts from July - for some reason I don't get all notification emails, not even in Spam.

Anyway, I have long wondered about those valve exhaust ports (before I saw your posts), regarding the plastic bits... However, I won't be satisfied until I've pulled my own Mech Unit apart and had a good look.

The reason I say this is that I'm becoming more and more convinced that the German Engineers who designed this box thought of *almost* everything. As per other threads, they clearly designed it to rigorously control clutch slip, to the point of apparently reducing the engine power to almost zero to prevent burnout. Also, as per my thread about it skipping 2nd gear and jumping down into 1st when it can't engage 2nd (presumably, in the required time) - they also thought about, and took care of that possibility. And as Crazy Dazz https://www.fordforums.com.au/member.php?u=2104378 says, other transmissions seem to take similar actions to workaround hardware issues...

Having done quite a bit of studying of the ATSG document's diagrams, I'm also very convinced the engineers thought about contamination* getting into the valve body via the exhaust ports - and I believe, they took sensible steps to prevent that too! If you look at the diagrams or the real thing, you might agree that they seem to have combined almost all the valve exhaust feeds together, and vented them back to the sump (via port "133U" on the 470 casting drawings). This is repeatedly identified on the 450 Hydraulic Schematics as "Combined Drain Ball" which is a simple Spring Check valve. FWIW, it has Part Numbers 350 & 351 on Figures 47, 48 & 49, and as far as I can see, it can only have one purpose - the prevent Reverse Flow back into the valve body. I know there are other valves that vent directly, but I'm guessing the Engineers figured reverse flow would not (should not?) occur with those.

I'd also wonder whether hydraulic valve exhaust ports really do suck back in during operation - the only time I can imagine that is following shutdown, when there could / would be slow drain back through the hydraulic pump, but would drain back occur through a port which is already below the fluid level? OTOH, I think the Combined Drain Ball valve is located at the top end of the valve body so drain back is likely there.

When I get to this in the flesh, I'll update - - -


Last thing:
* I use the word contamination deliberately, because it's clear too me that they wanted to avoid rubbish getting back into the valves. The question is, did they already know that bits of chewed up plastic would be among the contaminants? Obviously, wear metal is bad enough and would no doubt be what they wanted to stop in the first instance, but maybe they already knew.....

[AlanM]

I don't know whether they considered it or not, but the plastic bits definitely get in there.

[AlCan]

Originally Posted by AlanM

I don't know whether they considered it or not, but the plastic bits definitely get in there.

Yep, I'm not disputing that they get in. I agree, there's lots of evidence for it.

My question is still, exactly how? With that Combined Drain Ball Check Valve in there, it seems unlikely it's by that route, and my feeling is that the plastic slides in the clutch packs probably start to break up the year they leave the showroom. I'm about the change the fluid on my '11 and expect there will be plenty of plastic coming out, as it's had clutch shudder for years, but so far (touch wood), no mechatronic issues... So, if it gets in via those valve drains, it should have happened by now. Hope I'm not jinxing it!

[AlCan]

Originally Posted by oldgun

Managed to get the job don, took 3 hours with heat gun.

Originally Posted by AlanM

The internal filter is on the suction side of the oil pump so restrictions here could quite possibly impact on the operation.

Hi Guys,

Just wondering how you have got on since changing your primary filters? Any problems, or worked a treat?

The reason I ask is that I've pulled my filter out and carefully disassembled it. These filters are very cheap and nasty internally. They appear to be seriously designed to fail.

I'm concerned that the process of heating, softening and bending the top downwards during installation will end up damaging the filter, effectively destroying it.

The membrane in mine is not actually paper, it's like some kind of non-woven fabric, kind of like face mask material. No complaint about that. My filter was made by Filtran, who specialise in this type of filter. They have a patent material called "V Pore", which seems fairly effective for filtering high flow-rate oil / transfluid.

The first problem is that the membrane is installed basically flat, and is simply crimped into the edge seam between the metal base and the plastic top. That would be okay if it was supported adequately internally, but it isn't.

The real problem is that there is a series of "supporting" ribs moulded into the inside of the plastic top, but these ribs are not continuous, they are segmented. The gaps between ribs and segments are quite large, so that as the filter clogs up, it gets sucked down (ok, up) into the gaps between the ribs. Imagine a powerline running across the countryside on poles. It dips down between each pair of poles. So, a membrane which starts out flat ends up becoming a series of dips and peaks.

Effectively, this stretches a membrane which started out flat. Of course, it doesn't appear to be elastic or stretchy, so basically, it's put under tension, and that tension is retrained around the edge seam - which is where it fails. In mine, a section of the membrane has pulled out of the seam, with obvious consequences. This is how plastic bits get into the mechatronics, I'm in no doubt about that now.

My concern is that when heating and softening the plastic top and pushing it down to get the spigot into the pump throat, you're doing exactly the same thing - pushing the ribs down against the membrane and stretching it. Possibly to the point of pulling some of it out of the edge seam.

I do have a bunch of photos to post, but no time at present.

Anyway, I'm very keen to hear if your filter replacements have been successful.

[AlCan]

...retrained... I meant retained, or restrained, but maybe detained would be better.

[AlanM]

Worked fine. A mate has the car now still drives fine.

[whitelion65]

Following

[AlCan]

I promised some photos so here they are:

First, some pretty little hedgehogs! https://photos.app.goo.gl/2WwNu5LAM87V23y67 (Photo inverted to show similarity)


I used hot nichrome wires to cut through (most of) the filter spigot - to preserve the evidence as much as possible. Turned out to be by far the most difficult part of the job, due to the brass insert around the back (and maybe the bolt, which I left in to stabilise the operation). Broke 4 wires but got through about 90%, then snapped off the last bit.


Filtran OEM Info: https://photos.app.goo.gl/svjGW92phSpaca5F9


Opened up: https://photos.app.goo.gl/aeZPmrwUwZFJKq6t9 The oversize gap on the left side allows fluid coming through the lower section of membrane to flow around to the outlet, to the pump.


A tell-tale bulge, not what I expected, but has the same effect of letting dirt, filings and plastic through into the control valves etc: https://photos.app.goo.gl/YeyCmenYoktkgmA6A


In case you missed it: https://photos.app.goo.gl/dKpscd5FuGi5Xbp76


A close up: https://photos.app.goo.gl/t8TwBhfwuhugZ4N88


The most important details about this:


1. The failure occurred at 135,300km. Very early, compared with many, but probably just a result of being driven very hard, and maybe partly due to extra wear from the sudden stalling and cutting out problems described previously.


2. Even more significant, IMHO! The failure occurred very soon after performing a Shift Fork Learn. Like, about two days, maybe less than 50km, came on quickly, and rapidly lead to "TFL*" and DTCs, only 2nd & Reverse gears, very rough clutch operation and Limp mode, with an idle speed increase to about 1,000RPM.


I want to say more about this.

I've read all over the Internet people reporting Power Shift transmission failures very soon after a Ford Service involving Transmission Fluid replacement.


There is no obvious reason for this to be the case, but here's my theory...


As Ford has the equipment and the knowledge, I believe that after correctly replacing the fluid and hopefully the secondary filter, Ford Service then frequently runs a Clutch Learn and maybe even a Shift Fork Learn. Then a quick test drive before returning the car to its hapless, soon to be poorer owner. Huh? They set a ticking time bomb with a very short fuse? How? Did Ford do a very risky transmission flush as well? I say No, not at all.


Clearly, these Primary Transmission Fluid Filters were never intended to be replaced prior to a full teardown / overhaul. This could have been deliberate, built-in, designed to fail obsolescence, or it could just have been wishful thinking.


The fact is, these filters are put there for a purpose - to filter out any contaminants that get into the trans fluid during normal operation, to stop them getting into the Valve Body. This includes wear metals, clutch friction material and of course, those dreaded plastic bits from the disintegrating clutch damper spring separators.


So, normally these wear materials are trapped by the filter, slowly building up with milage and kW.hours. At some point, the filter becomes "significantly clogged", and as a result, increased suction is required to drag the fluid through the clogged membrane.


As you'll see in the photos, there is clear evidence of this increased pump intake suction in the subject filter. You can clearly see how the membrane has been sucked hard against the ribs in the upper filter housing, and stretched into the spaces between them.

https://photos.app.goo.gl/sicd3LX28wSTVnxYA



(For those who haven't seen one, the membrane in a new primary filter is perfectly flat, as seen through the intake hole underneath.) Although at only 135,000km, this filter was already significantly clogged. Then, at some point shortly before the transmission control failure occurred, the suction became too much, and a section of membrane pulled out of the edge seam, as per the photos. And crud then sailed on through into the control valves, causing one of them to jam.


So, why did this happen soon after a service and brand new, fresh, clean transmission fluid?


My theory:


Driving these cars moderately for a few thousand km (in Drive mode), I note that it is rare for the engine RPM to much exceed 2,500r/min. Only if I push it do the revs approach 3,000.


Here, we need to understand a bit about the transmission's internals. Fed directly by the Primary Transmission Filter is a Hydraulic Gear Pump. This pump is gear driven by gears directly off the main input shaft, which turns at Engine speed. FWIW, the gear ratio is about 1:1 (not sure exact numbers, but about 100 teeth on each gear).


Ignoring various losses for this discussion, the ideal flow through this pump will be in direct proportion to RPM. So, let's say the pump delivers 10 litres/minute at 1,000 RPM. At 2,000, that will be 20 l/min, at 3,000, 30.


The pressure drop across the intake filter will form an exponential curve versus the flow rate of fluid through it. For example: https://www.youtube.com/watch?v=ChkwzU-nH3c


With a new, clean filter, the pressure drop might be just 5kPa at 1,000RPM and 10 l/min and say 25kPA at 4,000 RPM with 40 l/min. However, when clogged, these pressure drop figures will be greatly increased.


With a clogged filter, we might see, for example, 15kPA at 1,000RPM, maybe 40kPa at 2,500RPM and 80kPa at 4,000 RPM. Now lets assume for the sake of this example, that the filter in question fails at 70kPa pressure drop. This is likely to occur at around 3,500RPM.


Now, is the theory becoming clearer?


As the milage increases, so does the amount of crud clogging the Sump Filter. The RPM Pressure Drop will be increasing, but if your engine never Revs over 2,750 RPM, the filter might survive until 250 or 300,000km.


But, at 240,000km, you take it to the Ford Dealer for its regular service, and they change the trans fluid etc, but the old, nearly clogged Primary Filter stays.


Now, being thorough, they also hook up their OBD Unit and run a Clutch Learn then a Shift Fork Learn...


When I ran my Clutch Learn, the RPM never exceeded 3,000RPM. And my filter apparently survived. However, when I ran the Shift Fork Learn a week or so later, I noticed that during the Initialisation phase, the Engine RPM went to 3,600 before settling at 3,000. I didn't think too much about it, and figured maybe I hadn't played my part correctly. But maybe that's not uncommon. Or, maybe 3,000RPM is just survivable by a badly clogged filter.


But at 3,500, it fails, and pulls a seam. You get the rest...


This is my theory for how the dirt, filings and plastic bits suddenly get into the mechatronic unit typically after reaching 150-300,000km. The filter has clogged and suddenly fails, probably at a time of high engine RPM. But it may take a while for a valve to jam, so you probably won't connect the two - especially if the high RPM occurred in the Ford workshop, without your knowledge.


MY ADVICE!


Based on my own experience, I would recommend anyone with a Power Shift transmission to get the Primary Trans Fluid filter replaced every 120,000km to be on the safe side, especially if your car has an unknown history, or gets to work hard.


If you know your car's history and have always diven it kindly, 180,000km is probably okay. Over 200,000km and you might be pushing your luck. Just like the cambelt.


If I get more time, I'll do a write up on what I found to be quite an easy job, though I was very nervous the first time.And as confirmed by AlanM and maybe a few others including now, myself, heating and compressing the plastic filter housing seems safe enough...


Let's keep our PowerShifts running for another 200,000km!


Lastly, a little puzzle for everyone.


What is this?https://photos.app.goo.gl/uhcSXAyiLGawQCRx6



*The dreaded "Transmission function limited" display message, for those wondering.

[AlCan]

I think that's 'Transmission limited function', but I'm out of time...