Armed with my mity vac and vac/pressure gauge I decided to try and find the source of my hard shifts, knowing full well that the vacuum mess was probably involved somewhere.

The '85 and later cars do not have the switch over valve contraption on top of the valve cover, in it's place is a micro switch. All the EGR and air recirculating plumbing appears to be down behind the passenger headlamp. One of the first things I did was remove the pipes feeding that mess from the mix as something down there appeared to be leaking.

I confirmed that I have 20" Hg at the primary vacuum connection on the feed coming from the pump. At idle I have about 4" showing to the line that feeds the modulator on the transmission, dropping to about 0" when the throttle is depressed. I replaced all the rubber vacuum connectors and confirmed that everything holds vacuum apart from the control valve on the injection pump.

From trying to follow the vacuum diagrams it looks like the control valve connects to a switchover valve, which then connects to the vacuum amplifier to which the transmission modulator attaches. It looks like what happens is that the control valve is designed to bleed off vacuum at an increasing rate as the throttle linkage is advanced, emulating the inlet of a gas engine. I have a feeling that mine is bleeding off way too much, resulting in the low reading seen after the amplifier. If I disconnect the control valve feed from the amplifier while the car is running, it makes no difference to the reading at the modulator feed. If I induce a vacuum on the control valve line using the mity vac, I immediately see an increase in reading at the modulator feed.

As a short term attempt to improve the quality of shift I have adjusted the screw on the underside of the blue vacuum amplifier, and wound it all the way out. At the maximum extent of it's adjustment I can see about 8" at the vacuum modulator line, and the shifts are noticeably better.

Have I come anywhere close to understanding how this stuff is really supposed to work? From reading other posts it sounds as if there is an adjusting screw under the white domed cap on the side of the control valve, presumably that would be the next thing to play with?

Kevin

__________________
'85 300SD

Kevin,

I've recently spent some time with the vacuum system on the 300SDL. It sounds like the system is similar to yours. The modulator feeds the vacuum amplifier. However, the vacuum amplifier also is fed with straight vacuum. It is also fed with manifold pressure. Somehow, it combines the modulator feed with the manifold pressure and feeds the transmission a "regulated" vacuum.

The SDL was thoroughly tested and here are the results:

When checking the output from the modulator, on the line to the amplifer:

The output will start at about 15 inches and will drop proportionally with throttle position. I had it down to about 5 inches with a quick blip of the throttle. Since it was in the driveway, in neutral, I cannot confirm if will go all the way to zero, however, it is likely.

Now, the output to the transmission is quite different. It starts out at about 10 inches and remains there until the pedal gets to about the 1/2 way point. Then it drops steadily with increasing pedal to about 5 inches. To get it below 5 inches requires the turbo to be in boost. It will then go down to about 2 inches.

My first advice is to check the output of the modulator. Connect a gauge to the output and drive the vehicle. See if you start off with something above 12 inches and see if it drops proportionally with pedal pressure. If you get this, then any issue must be with the vacuum amplifier.

I have not completely sorted out the amplifier and have not done anything to adjust it, because the shifts of the SDL are fine the way they currently are. However, I'd sure like to determine how that amplifier exactly works.

Thanks Brian.

I've just experimented some more and have a much better understanding of how things work:

The vacuum control valve on the injection pump is not a source of vacuum, it is a valve that admits air to reduce vacuum pressure. The further the innards of the valve are rotated the more air it admits, thus reducing the vacuum present in the system.

The control valve is hooked up in a rather weird looking circuit to the crossover switch which seems triggered by the coolant temp sensor. The hose on the middle of the crossover feeds to the bottom of the vacuum amplifier. Since both sides of the switch connect to the same Y piece eventually there must be a restrictor somewhere that biases one side more than the other. I found a scan called "egr-california.jpg" on the forum which looks exactly like my Federal car so far as this area is concerned, see 300SD diagram. I believe that the purpose of the crossover is to switch between the less vacuum controlled branch and the branch connected to the regulator valve once the engine is warm - presumably to give more vacuum = softer shifts on a cold engine.

This vacuum circuit, which TEEs off to the control valve, is fed from a nipple on the main vacuum trunk just before the check valve where the 4 way rubber junction hangs off - marked white and then black where it connects to 140 on the diagram. My nipple was completely blocked, so this entire circuit had no vacuum feed at all. I carefully, I thought, cleaned it out with an appropriately sized (I thought) drill bit. Bingo, all of a sudden there is vacuum going to the crossover switch and suddenly the output to the modulator valve jumped to 16" Hg

Clearly the nipple that the circuit is fed from incororated a restriction of some sort, which I had "cleaned" with my drill bit. I went through all the pieces left over after eliminating the EGR and air recirc stuff, and found a couple of color coded connectors - both of which featured an internal restriction. Think of them as color coded resistors in an electrical circuit. I enhanced one of them with a safety pin until the reading on the line going to the modulator valve was around 10" at idle, I can make it go higher if I open the restrictor more.

At this level of vacuum there is a tiny flare under certain load conditions going from 2-3, but the shifts are incredibly smooth. I would not have believed it possible, it's right up there with the electronically controlled transmission in my wifes Accura. Yesterday I had shifts which produced a marked clunk from the rear end going 3-4, right now it is honestly difficult to tell when the shift occurs.

Once I got the hang of using a pressure gauge with the vacuum system it became just like using a voltmeter to track down bad connections. There appears to be no reason at all that a car with 174K on it shouldn't shift beautifully.

I also replaced my ignition tumbler, the old one fell apart in my hands when I pulled it out, and replaced a previously missing engine shock. A pretty smooth weekend all in all.

Kevin

__________________
'85 300SD

A further update, in case this helps someone else down the road:

It appears as if the side of the vacuum circuit that is used until the coolant temp hits 50C is in fact a lower vacuum than the side connected to the control valve. I say "appears" because I had consistently firm, but not jarring, shifts through all gears until the crossover switch kicked in and then instantly had smoother shifts.

Now that I have functioning vacuum to the control valve I will run through the process of adjusting it's range and see what effect that has.

Kevin

__________________
'85 300SD

This appears to be in conflict with the generally accepted function of the 50C switch, which is to soften the cold shifts. Without the 50C switch in the loop, the shifts are noticeably firmer when cold. I have disconnected the switch completely on the 603 and I cannot confirm its function, with certainty, one way or the other.

You are quite correct Brian, I had the pipes reversed.

I sat down with a clear head and the vacuum diagram and followed everything through. Having previously removed the plumbing to the EGR and air recirc I now reinstated the restrictors which had formed part of that circuit, and double checked all the remaining connections.

I now get vacuum everywhere I should, and the valve above the IP works as expected. I now have no flaring, no clunking, just proper shifts at all engine speeds and temperatures.

Kevin

__________________
'85 300SD

Kevin,

Dave and I seem to have some question on the transmission vacuum when the blue amplifier is utilized. It appears that the vacuum to the trans stays at nearly a constant 10 inches until you get well into the pedal. Then it drops down to about 5 inches until you get into heavy boost. Then it will drop to about 2 inches.

Can you check the output of the amplifier (line to the trans) while driving the SD?

We would be curious if the vacuum starts out at 10 inches or so and remains there for about 1/2 the pedal travel.

The amplifer output is in direct contrast to the output from the VCV, which is almost dead linear with pedal position.