Mercedes-Benz Forum banner

PSE Central locking pump overhall

103K views 38 replies 19 participants last post by  trialsntribulations  
#1 · (Edited)
Probably a pretty odd first post, but have used this forum to help sort mates merc's and feel I better stick something back to the community. Hopefully save a few of you a grand!

To start, this is the pump assembly from a W200 S55 amg, after a heavy winter the thing started blowing its fuse regularly, google pointed me here for the suspected fuse and the general consensus that water ingress/condensation was at fault, and a new unit required as it is coded to the alarm.

Now im a mechanic, but also have a degree in electronic engineering, so 9/10 times i manage to repair units like this/ecu's/EPS set-ups and so on and retain the original coding. Thus avoiding the main stealer's. Given this pump was essentially a lost cause, further investigation was warranted.

Electronically the pump checked out, but current draw was excessive, i suspected a stalled motor, either corroded or jammed.

Here is the unit with the main half's pried apart and the pump separate from the valve body and control circuits

Image


at this point you ought to apply 12v through a 5A FUSED!! supply (like a PowerProbe) to the motor directly, then if the motor doesn't stall, block either the outlet or inlet and see if it stalls. (it should never stall). if it stalls your onto a winner, the pump or motor are at fault.
note the white marking to ensure the orientation is correct on disassembly, remove the securing torx bolts but hold the pump together with a spare finger!

slowly remove the metal ring that sat under the torx bolts

slowly remove the top carbon/graphite cover

now the next bit is very tricky, you need to remove the centre part, without dislodging the little blades so they stay in the correct grooves, to do this you need to hold the centre graphite ring, and the part wit the slats as one, once removed vertically place against a peace of paper and you can then lean it on the paper till it can be layed down on a flat surface.

Holding: -
Image


once laid down: -
Image


you should now be left with this: -

Image


gently holding the remaining carbon/graphite piece thats attached to the motor, power the motor up again. if the motor still stalls out, you have a motor problem and will be on your own :( if the motor spins fine then the issue lies with the carbon/graphite pump assembly!

now if you look at the last photo you can clearly see where the pump has "ran" on the surface, its this shiny stuff that's causing the pump to stick, it needs buffed back to be the same as the rest of the surface. this is achieved by (and i cannot stress this enough) VERY!!!!!! gently buffing with 1500grit wet and dry paper, DRY! do this to every shiny part of the pumps internals, be very careful not to accidentally round any sharp edges and be very careful when buffing the individual blades and the contoured edge that makes contact with the centre ring.

once done, reassemble, only nipping all the torx bits give it a wee shoogle to centre it all then final tighten. don't go wild, the pump is made of a light brittle material, you don't want to brake it! and repeat the test with the fused supply, all being well the pump will run like a dream and not stall!

reassemble: -
Image


and refit to the car! .. I hope you marked all those vacuum pipes :thumbsup:


A final note, a few people said to me while i was doing this, "why not squirt in some wd40" simple, this pump is made of a material which is clearly self lubricating, same sorta stuff motor brushes are made off. if you were you introduce a liquid lubricant any deposits would stick to the liquid, before you know it the liquid becomes a paste, to be more precise, a grinding paste ;)

also i wore powder free latex gloves to do this as i feel even the oil from your fingers would hinder the materials lubricating properties!

enjoy :)


-- to the mods/long term members, feel free to host the pics on a local server if there of use and edit the links, as i may not have my photo-bucket forever ;)
 
#2 ·
Not an odd post at all - a great one!! Welcome to the forum!!!

Is it possible for you to save this as a .pdf file, and post that? If not, I'll work on it. I'd like to add this to the W220 S-Class Owner's Encyclopedia, and would hate to lose it if you give up photobucket.
 
#3 · (Edited)
PDF file of post

I have saved it as a compressed pdf, so we won't lose it. I copied text blocks and photos sequentially into a Microsoft Word document, then saved as a .pdf. Hope the quality is adequate. It will require Adobe Reader 7 or Acrobat 7 or later to open.

Thanks again!
 

Attachments

#6 ·
Thanks for the post, what test did you use?

So im having almost the same issue with this pneumatic vacuum pump in my 2003 S430 and pulled mine apart too. It looks identicle to the pics btw. I dissasembled before thinking of marking for reassembly but luckly it has grooves that align it up and the little squares seem identicle when measured so im pretty sure its all back together correctly. When apart I found that the motor on mine didnt spin when I tried to move it with my fingers. I put a wrench on it and added gently torque and it broke free and was able to spin again :) I hooked it back up to power supply and ta-da it runs. Im hoping that it was just a frozen moter and that my $500 problem is solved but just in case its not...

How did you test the circuit board to see if it is bad? Also if it is bad does anyone have any suggestions on repairing it or taking it somewhere? I live near Seattle, WA

I called Bosch and they would not provide me w/ wiring diagram for the unit or give me any specs for testing it. Im not a electrical engineer, but know a little, have a 12v power supply and a few electrical testers. Thanks... and I really appreciate the post its exactly what I was looking for :thumbsup: !!!
 
#7 ·
In the event the motor has became a short circuit, the next thing to do would be to connect the unit to the car but leaving the circuit exposed to trace power, the motor was fed through a relay and some sort of ajustable vacuum switch arrangment, all pretty old school stuff, so any failure here should be easily found(burnt contacts, dry solder etc) if it's not that then it's a burnt trace or power feed issue, due to the relay etc the controll circuitry is well isolated from any damage a seized motor may cause! If the controll circuitry showed signs of water ingress, then unless simply drying and washing with alcohol doesn't work I would leave it to an electronics person , as will be a minefield.. Hope it works for you!

Incidently, as the vacuum pump is powered by a relay and not a FET, any 12v vacuum pump that draws <10A could be substituted!
 
#9 · (Edited)
I can only relay my experience with my past W140. Had the "instant blow fuse" situation on the PSE unit under the rear seat. Before expending big bucks, I looked into the cause and it was the electric motor brushes that had worn down so much, that they cut the brass? electrical legs that hold the carbon brush to the armature of the motor. My armature was pretty beat up because of the brass grinding onto the armature. Started looking around and at the local hobby shop, found a 12VDC motor for one of the model cars that was the same size. It even had the same bolt pattern and holes to hold the pump assembly. So I was going to just swap them out, but two things stopped me, (1) the current load of the replacement motor was huge and (2) it was a 4 pole motor where the benz motor was a 12 pole. Eventually I just took the new brush with cap off of the new motor and put it on the old one. Fit perfectly. $20 solution.

However, these things are a B*&ch to break apart as the end caps are cheap plastic and the cans are one tough steel so gingerly pry the retaining bends apart. I actually cut the bends at the mid point (you'll understand when you look at the motor) and bent the tabs back individually.

Reinstalling can be confusing as I found no method to easily retain the brushes when reassembling so if someone has a better idea, chirp. I ended up taking the armature out of the assembly, installing the brush end cap onto the armature first outside of the motor housing, and then reassembled the rest of the assembly. This could easily destroy the bearing cap on the other end if you are not careful.

The whole rebuilt assembly lasted about 1-1/2 years and finally succumbed (I'm guessing here) to the quantity of carbon brush dust generated fouling the circuit board and letting the smoke out. Still $20 for a 1-1/2 years I didn't mind. I would strongly recommend that your turn the old armature brush contact point (on a mini lathe or sandpaper) first to clean it up before reinstallation and you will probably be alright, but I don't know. I didn't have access to a hobby lathe to play with it so I let it be.

Lastly, please don't start badgering me for the model number of the motor, I can't remember because it was over 5 years ago that I did this. Just go to your local hobby shop with the old motor and start comparing what they have in the display case. I wouldn't be surprised that if you get any motor with the same diameter, the end caps can be used.

If I had to do it again, I would install the larger current motor with an appropriately sized PTC to limit the current draw of the motor. That way it would shut off in the future just like the old headlight wiper units if a problem developed, but I knew nothing about PTC's when I did the first mod.

Good luck and its your nickel (okay $20). Your success may vary so tell me how your experiments work out.

Ron
 
#11 · (Edited)
Welcome to theforum.

It is above the left rear wheel well in the trunk, behind the trunk liner. The first photo (on left) is from a '00 W220. The second (on right, with white arrow showing the pump) is from an '05.
 

Attachments

#14 · (Edited)
Just wanted to say thanks very much for this post, it was a great help when I rebuilt the pump from my R230 just now.

Just a few things to add:

The pump mechanism is made from ceramic, which is basically a form of amazingly hard glass. It's harder than any steel, which is why the blades barely wear, but it's very brittle. I'd not recommend using an abrasive on it, it's so hard you're probably only removing crud when you rub it anyway, but if you do manage to scratch the surface, you may well add a problem you didn't originally have as ceramic (like glass) can break along scratch lines.

My problem became apparent because the soft close, or close assist stopped working on my boot (trunk) after a few closes. What I discovered after much pain, was that the ceramic pump mech had some crud in it, which was slowing it down slightly. That meant that the pressure wasn’t high enough to close the boot assist within 10 second, which alerts the ECU to a possible pipe leak and the circuit shuts down after a few closes.

I fixed it by meticulously cleaning the all the internal components of the ceramic mech. I used something called Profi Dry lube, which is a PTFE powder in a volatile solvent which evaporates off very quickly leaving the components dry and with a fine coating of PTFE. However, after spaying, I wiped all the parts dry with a microfiber cloth so I don’t think the PTFE was helping much, rather that the solvent simply removed all the crud.

I then reassembled the pump mech. You can see where the little pump blades have worn slightly, so make sure you put them back so that the same edge is in contact with the pump housing, because they will have bedded-in in that position. Also, make sure that you don't over-tighten the six bolts that hold it all together. I think that’s very important, not just so as to avoid breaking the ceramic, but if the bolts are too tight, it could add extra friction to the mechanism, you’ll see why when you open it.

There is an easy test to see if it’s all working properly. With the boot open and the pump visible, use a screwdriver to carefully trip the catch in the boot lid, basically you trick it into thinking it’s time to close. The catch shuts and you can clearly hear the how long the pump is on for. It should stop after no more than 5 seconds, if it runs on to 12-16 seconds, it’s logging an error and the close assist system will stop working after five or six operations.

After my rebuild, mine took only 3 seconds to close, before it was taking 16 seconds.

Hope that helps someone else.
 
#20 · (Edited)
Skylaw, I'm opening up an old thread here.

Hopefully Evolotion or someone can help me here. I was working on the pump and as I pulled the off-white "C" connector from the pump motor itself (the clip that holds the vacuum and pressure line onto the pump) several small small o-rings fell out. Now, I 'THINK' I know which ports and which direction they go in, but we all know what that will get us.

Is there anyone that can go over that?? There are two small manufactured plastic spacers with 'fingers', a small filter screen and what appears to be a black rubber sealing ring that I 'believe' goes there too. This is the one I'm having problems with.

I've established that the filter screen goes on the vacuum side of the pump which makes sense, (the left side of the pump if you are lookng at the complete assembly with the circuit board towards your body). The cleanest of the plastic rings will go on that side with the fingers pointing in towards the pump. The plastic ring that has all the carbon looking dust on it will naturally be the pressure side and again, fingers on the ring pointing in towards the pump.

Now, the ports that plug into these two holes appear to already have small, sealing O rings in place in an recess . . . . so where does the other black rubber ring I have came up with go? And it's not exactly like a completely round o ring, it's more flattened like the rubber tires on toy slot cars. I'm wondering if this even goes in line somewhere (and if it does, most likely on the vacuum side), or if it inadvertently fell from some place else.

Edited to add:

Also, as a wrencher since I was 12 and an EE, I'm taking a different engineering approach to reassembly of the pump vanes here. I do think that as the author states. . VERY LIGHTLY buffing the shine off of the surfaces of the upper, lower and both sides of the inner middle ring (spinning ring that holds the small vanes) is a good idea. Also, lightly buff the inside diameter of the the outer middle ring. . . . however if you look at the small vanes you will notice that any 'shape' it has that may be different from one side to the other has been caused by wear.

I'm one that thinks it may be best to turn these 180 so there is a new surface riding against the freshly buffed inside diameter of the outer middle ring. At the risk of 3 or 4 hundred bucks for a used replacement unit should I be wrong here. . .I'm going to try that.
 
#22 · (Edited)
???



Ok, I answered my own question on the O ring, it was one that had inadvertently stayed on one of the outlets with the yellow retainer from one of the lines itself. (each line coming off of the unit has both a hard plastic O ring and a rubber sealing O ring). Also, I was correct on the arrangement of the plastic rings (with fingers) and the filter screen on the pump itself. Again, after buffing the internal surfaces of the pump itself I rotated the vanes on the internal pump itself. So far so good, although I have yet to do any pressure/vacuum checks.

I have repaired my problem, which was a shorted power supply to the motor that was riding next to the shaft support on the motor itself. I don't know whether it simply shorted over time and the short caused the insulation immediately surrounding it to melt, or whether I had a motor runon problem that caused exessive heat which then melted the insulation and THEN caused the short.

I've read the informative pdf article on this system and there's a wealth of information regarding vacuum checks.. . .but what about runtime checks. i.e. . . . systems is energized for x seconds after a door softclose function, or x seconds for a full 4 door unlock function. . . . Anyone seen anything like that? I haven't so far.

Further, I have seen a video on YouTube showing the complete replacement of the motor here along with part numbers, etc.

https://www.youtube.com/watch?v=Fb7eRa63Z0I

And here's the website referenced which seems to have a wealth of interesting parts that I have not seen available elsewhere.

http://www.hongmei.com.tw/
 
#23 · (Edited)
Also, I just found this in another thread . . . shaft diameter on the motor is different though. However specs are nice to have here.

I have a 2004 S600 and my central locking pump motor burned up. I have done a huge amount of research and in fact found the motor for these pumps however the company that produces them in China will only sell them 1000 at a time at around $5 each. However I did find a company here in California that "currently" has 22 motors that are extremely close to the motor used. In fact the only difference I can find is the shaft diameter is 6.35mm instead of 6.0mm. Other than that they are identical in all aspects including how the pump bolts on to the front. After I opened my unit I discovered the only item that would need to be modified to use this motor would be the small drive collar that slips onto the drive shaft and runs the pump vane unit. It should be extremely simple to bore out the hole on this collar from 6.0mm to 6.35mm then everything else is a simple bolt up and play. The motor costs around $56 but they said that after these 22 are gone they will not be ordering anymore. The company is Anaheim Automation - Your source for Stepper Motor, Brushless DC Motor, DC Motor, and Planetary Gearbox Products and the motor number is (BDR-52-85-13.0V-6280) @ $56 each. To go directly to the motor spec page go to, 52mm - Small DC Motors
 
#25 ·
Thanks Ex, That at least provides some baselines there.

I'm noticing that I have a slow drivers rear door lock. When locking or unlocking all 4 the pump wants to run for about 9s (as opposed to just 1 or 2 when unlocking/locking the drivers door only) . . . My concern is the door soft close function though. Devoid of what door, the runtime on the pump (and hold time on the doors) is about 13 to 14s. This seems a bit high to me.

But alas, it sounds like I'm going to have to put more work into this than I really want to and check the pressures according to the PDF document that's floating around. I'm sure anyone who has researched this system on here has seen that, but I'll find a link and reference in this thread for posterities sake in a bit.

Thanks again Ex!!