steps backwards - Is my vacuum understanding correct
 

84SD 617. Vacuum pump replaced with a new pump from whoever the OE manufacturer is. There are approximately 8,000 miles on this vacuum pump.

Recently and intermittently, brakes get hard and car won't shut off. There are 2 small vacuum line branches in the main engine compartment and another larger one behind the false wall that goes to the door locks.

The large branch is plugged awaiting my tracking down leaks and getting the door vacuum working. I measured only 10" of vacuum at one of the smaller ports then disconnected and plugged the other small port. No change. Next up was to disconnect and plug with my thumb the line at the brake booster in case the booster was leaking.

Next was to connect another vacuum gauge to the 2nd port to confirm the 1st guage. No change, still 10" of vacuum.

The car will shutoff by connecting the mity vac to the shut off line.

Are there any other leak sources or one way valves that could be leaking? If not, I'll look into another vacuum pump or rebuilding this one.

Ports that are plugged:
2 small lines in the engine compartment
larger port going to door locks
end that goes to the brake booster.

Is there anything else? Shouldn't vac be 18" or so?

Vac should be >20" if the pump is healthy and the vacuum lines are in good shape without leaks.

Is it possible the check valve at the pump isn't sealing?

As previously stated a low vacuum can often be the Check Valve the screws into the Vacuum Pump. However, a new Vacuum Pump I believe comes with a new check valve so one would think that would make a check valve issue unlikely.

They way to know for sure would be to remove the Tubing where it goes into the Vacuum Pump and use one of those tapered rubber adapters and push that into the Vacuum Pump Check Valve so it seals and measure the exact vacuum that the vacuum pump puts out.
If you are getting good vacuum there thest the Brake Booster to see if it holds vacuum. After that all that is left is the main vacuum line.

I'll work on it tomorrow. There may be some useful fittings in an old compression tester kit.

The door locks work from the vac in the trunk, not the one under the hood. Sounds like your shut off valve (inside the back of the injection pump) is OK. Probably just a leak; check all the rubber Y connections. The plastic lines are probably OK. Mine is a 1981 300SD and the plastic lines are just like new.

I agree, starting at the source and working your way outward is the most efficient way to examine a vacuum system. Pump, check valve, main vac line, brake booster then each branch off the main trunk line.

On an SD I would remove and isolate the the cabin and that horrible mess of pods and actuators under the dash from the system first.

Any way I measure, I get 8" at idle. Vacuum goes up to 20" if the engine is revved and stays there as long as the engine is revving. Vacuum slowly settles to 8" when the engine returns to idle. The one way valve tests good.

I'm going to check warranty but will likely end up replacing or installing a kit. Any suggestions on what kit or what brand of pump?

I won't explore further to fine reasons for failure until parts are in hand because I want to keep the car mobile. Strange that this crapped out with out many miles.

Edit: I think the pump was a Pierburg

If you are measuering directly at the vacuum pump you have some issue in the Check Valve or Vacuum Pump itself.

When I had a low vacuum and I mean way back in like 2008-2009 I removed the Check Valve and found it full of Gunk. I holsed the innards of the check valve out with WD-40 and re-installed it and had no more issues from the Check Valve. That was a long time ago.

Later I took off the the vacuum pump and took it apart and cleaned it and re assembled it.

I have also pulled the pump of for inspection of the Arm at least 2 times. Never found any issues.

Not sure about the avaliablity of Kits as they were getting scarce and never were cheap for the piston types.

There was 2 kits. One for the Arm and Bearing end and the other for the Piston end with the valves.

@ Diesel:
I measured directly at the pump even with the closest metal line removed. Next, I pulled the check valve (22mm socket) and it did have some gunk but easily cleared. I reinstalled the check valve & re tested. Vac was still 8-10" unless the engine revved.

I've read threads where pumps failed for various reasons and am not looking forward to tracking down internal engine parts to repair. I'm contemplating getting a compression gauge as that also needs to be checked as a step in figuring out what is causing an engine shake.

Vac pump repair kits are apparently NLA. I've seen a few on Ebay at ~$170. I'd buy a new Pierburg pump at that price and keep this one for parts. I may have kept the old one somewhere.

Current thoughts are to: 1. run compression test while vac pump is installed so I have compression information on engine health. Right now I only know that it starts easily with 1 glow even at temps in the teens.

2. remove and disassemble the current pump before deciding the next step.

I have a parts car available but want to keep that engine complete without scavenging parts.

Some of that gunk you found may have made it inside the pump to the two small internal check valves. That is what happened on mine. If you are getting at least a reduced amount of vacuum, you may have similar.

It would be great to be able to clean and install. That would leave the open question about where the gunk came from.

Now what
 

Brakes are hard at startup and engine doesn't always shut off immediately with the key. Fuel cut off valve tests good and shuts engine down with a mitivac.

Vac at idle measured at the pump with 2 vac gauges measures 8". Vac quickly goes to 20" when engine is revved. Both gauges match each other and also go to 25" when connected to the Dodge Cummins engine.

Pump itself is Piersburg bought new with approx 8,000 mi on it.
Main check valve works - ie lets air into the pump but not the other way.

The attached pics are how the pump came off the car without any cleaning.

Is the next step to disassemble the pump? No kits are available so I'd either have to reuse gaskets or make some with by using a ballpeen hammer.

Edit: I didn't disassemble the ramp but there is no obvious side to side play. The bearing in the back of the pump is nice and smooth.

Doesn't yours have the cover with several small screws? If so, all that is needed is to remove it and the two small check valves are inside.

OK, I'll check on it tonight. I was just uncertain about whether the seal could be reused and whether there was a trick to reassembling. The spring that resists the roller is nice and strong.

this PN. 000-230-13-65

Pelican Parts - European Automotive Parts and Accessories - Porsche • BMW • Mercedes • Volkswagen • Audi • Saab • Volvo • MINI

Have you blocked off and disconnected everything from the pump (including the booster) to test the vacuum directly at the supply fitting on the pump? Sorry if you've answered this already it isn't clear. If you're leaving the booster attached and you have a leak in the diaphragm, low vacuum in the system is exactly the symptom you'd be getting.

The brakes should be holding vacuum even if the rest of the system doesn't. There's a check valve in the brake line to have it hold vacuum. It should have enough vacuum stored to last a pump or two of the pedal before going hard.

Yes, the vac was measured both at the 1st metal line coming off the pump and again carefully at the check valve. I don't see how I could have mismeasured because all of the external vac drains were out of the system.

Brakes are only intermittently hard. I noticed first when backing after starting the engine. Brakes felt weak and didn't grab. It isn't consistent so let's say that the brake booster needs to be confirmed as holding vacuum.

Can the vac pump be reassembled without new seals? If so, I'll make sure that the individual check valves are working.

I'm wondering whether parts are missing. There are 3 check valves under the cover. The 2 on the left had the flat side up and have seals underneath. The single valve on the right was installed upside down when compared to the other 2.

Seal kits are NLA but show 3 of the seals for the check valves.

This was a new pump that worked so I'm expecting that I goofed something up. I intended to simply lift the cover and take a pic but am not sure at this point.
(I hate it when that happens - but need to put it back together correctly.)

Does anyone know:

1 the correct orientation of the single check valve
2 is there supposed to be a seal under the valve?

edit: Figured it out by looking at the old pump. All 3 check valves have a seal under them. The 2 valves on the left in the pic are installed with the flat side up. The single valve is installed with the flat side down.

There are o-rings is the Harbor Freight metric kit that "may work" but I didn't have to try them. The OE seals have flat upper and lower surfaces where o-rings are made of a round material. Still, the o-rings seemed like they would work based on their thickness and diameter.

The old pump that was still in my parts bin may have simply needed cleaning out. The single check valve was full of nasty thick oily stuff. The other 2 were relatively clean. I didn't look at the external check valve but it is easy to remove. I haven't seen it obviously posted but cleaning the valves would be the place to begin instead of replacing the pump.

I reinstalled the pump and ran the engine without installing the fan and power steering belt. Vacuum started out at 8" then climbed to 15" at idle after running a little while. I'm hoping that I'll get more with a trip down the road.

Does anyone have any idea why this may have failed? The pump is a Piersburg and was bought new. Engine oil is full Synthetic and changed without an extraordinary amount of miles. I'll check later. Nothing seems amiss regarding the pump itself. There was a little bit of stuff in the external check valve but not really what I'd call a lot. The ramp and bearing seem good.

Why low vacuum that goes up on acceleration?

If you have a check valve that's being held slightly open, reduced vacuum at idle wouldn't be unexpected. If everything is sealed up and working, you should be looking at 20" or better. The check valves *MUST SEAL* or you won't develop the proper vacuum. The pump itself likely isn't the issue, the check valves are.

Diesel, are all 3 internal check valves the same? If so, have 6 to choose from. I'll check vacuum again before putting fan etc back on.

Theoretically, could the piston seal be a problem?

Edit: measured at the pump with the 1st metal line attached, I get 22" with a fluctuation. I think one of the check valves may not be perfect but couldn't identify 1 as being really bad. They are all clean, all installed correctly but still have the fluctuation so something isn't holding. I may also have some vacuum leaks which aren't critical for testing because they can be blocked off but will/do affect vacuum and could affect trans shifting etc. I thought all of that was handled.

I still can't account for the pump having issues with this little amount of mileage.

All of the external leaks are now holding vacuum when applied. Brake booster, VCV, trans line. All others are blocked off.

The external check valve on the vac pump leaks when vacuum is applied with a hand pump. The external valve on the pump in the parts box is coming apart.

The pump is now only putting out 15" at idle when measured at the pump and still goes up when revved.

The local parts house has a Pierburg pump available for $275 which is about what is available on line. I'll probably order just to get this thing back on the road. I cannot account for the failure but will test the external check valve to measure how much difference there is between it and the old one.

The check valve alone is ~$90 and may not fix the problem. I wish the repair kit were available.

That's not a bad price. I think I paid close to 350 for one a couple years ago. My pump worked fine but I'd read all the horror stories over the years about pumps ruining engines and couldn't rest until I installed a new one.

At the first sign of trouble out of the newer pump I'll change back to the old one and start saving for another. It's one of the few things that can destroy these engines. Imo, by buying a new VP you are doing yourself a favor, biting the bullet instead of having it fired at you from a gun. :)

The parts that destroy the engine are perfectly fine. The parts that create and hold vacuum are not. I ordered today and will report back once installed. The local chain store had it, not the zoo or the leprerchaun or no-acceptable-parts-anywhere, the one where parts advance.

edit: store shipped the wrong part. Only the kit with the roller was available.

Apparently, Pierburg pumps aren't plentiful. Genuine Mercedes pumps are nearly $400. Pierburg pumps were available a few years ago but I haven't found any yet. This is turning into a more expensive problem than I expected.

Send me a PM if you've seen a new pump in the $300 range. I'll look a little more.

Edit: Pierburg pumps are apparently NLA. The only thing I found was genuine delivered for $360. That included a core charge and shipping but was still the lowest delivered price found for a new pump. There are threads where some are converting to electric vac pumps. I decided to no not screw around figuring out how to do the conversion and get it working.

Perhaps add snagging these pumps to your yard trip list. At least open the front cover and get the internal check valves along with the external valve.

I ended up with a new OE pump. Vacuum at the pump at idle is 22". Vacuum measured at one of the small lines on the main line going to the brake booster is 18". Apparently some leaks are still in the system.

The only repair parts available is a kit for the bearing and the main external check valve. Perhaps the bearing can be bought from a bearing house. The kit was too expensive to be worth trying to repair. The kit with the 3 check valves is NLA. Only OE pumps are available new. Stock is drying up and prices are rising.

I didn't find any reason why the pump failed. The bearings were intact & the ramp looked fine.

I would like to see how an unmolested main vacuum line is installed. Mine doesn't look like the factory would have designed it that way. For now, everything is working. I'm going to snag a pump from the yard since the supply is drying up.

The external check valve/fitting is expensive new ($90), which is why most 300D/SD you see in the junkyard have it removed. I bought the rebuild kit (check valve, piston seal, gaskets) a few years ago for $110. Not comforting that it is NA today or $$$. The earlier pumps w/ rubber diaphragm have cheap rebuild kits. I have seen VW pump rebuild kits on ebay which have internal check valves which look the same, though the other parts differ. I recall those kits are ~$25 so might be one way to get new check valves. You can buy the pump to block gasket on PP cheap.

One of the local parts stores had the gasket. The 1st price quote was .80 then they charged $2.50. I suspect the 1st price was cost. It is a Victor Reinz. You could make one by using a ballpeen hammer and stock material.

I imagine that people would be selling the kits if VW valves would work. Apparently at least 1 of my valves is bad in the old pump. I couldn't tell by blowing through them. They all held to a point then leaked. The new check valve did too. The rubber gasket under the cover and o-ring are also NLA.

I got tired of R&R only to have the same results so just bought new. Still no idea why it failed.

@ Bill, do you have any info on what VW kits you were looking at?