The valve on top of the IP costs about $180 new. At least on W123s or similiar W126s they do.

__________________
Jeff M.
Mercedes W123 DIY pages are now located here.
1983 / 1984 300D Sold
2000 CLK430 Cabriolet ~58k Sold
2005 Avalanche 4x4 ~66k

This is the one that bolts to the side of the IP on a 603 in an '86 300SDL.

Sounds like a completely different animal.

__________________
Jeff M.
Mercedes W123 DIY pages are now located here.
1983 / 1984 300D Sold
2000 CLK430 Cabriolet ~58k Sold
2005 Avalanche 4x4 ~66k

The first and second picture in Brian's posting over on the first page show it. Little metal critter with two lines in the top and one vent line out the side. I pulled it off teh pump and crudely tested it. Could not tell any difference in flow thru it in any throttle position. Seems to be just running straight thru in all positions. Figures it will be the expensive part that is bad. I could not believe how much of an effect the vacumm had on shifting.

It would, effectively, be impossible to test on a bench without a vacuum pump.
The reason is that it bleeds air constantly and you can't squeeze the Mityvac fast enough to compensate.

You can test it on the vehicle with the vehicle running. Just hook up a vacuum gauge to the OUTPUT hose that heads over to the transducer (big blue valve). Make sure that you don't put a T into this line. Connect the line directly to the gauge. Drive the vehicle and watch the gauge at various pedal positions. This will be a valid test to determine if the VCV is the culprit.
Make a note of the gauge readings at various pedal positions and report back.

Here are the updated vac readings per my late shift problems ... I need help finding the source of the problem here ... (thanks) ...

Main Vac Supply from Booster line at amp input: 22" at idle.
Tranny Line at amp output: 17" idle, stays at 12" under acceleration all the way up tach to 4K RPM, will spike down to 5" if I stomp on the gas, and I mean stomp.
VCV line from switchover valve: 14" at idle dropping to 7" as soon as I press on accelerator and staying that way thru acceleration.
Boost line from IP area: Tops out at 9" floored in probably third (40 MPH). Usually sub-9" though; only pops there every now and then.

Apparently my setup is more like a 722.3, which can be viewed here ...

www.ultrahot.com/vacuum/722_vacuum_func.jpg

Is that not enough Boost to reduce vac enough for proper shift? Shifting update: car only likes to shift between 4k-5k RPM, but shifts well there. But freaks out the girlfriend when floored through residential neighborhoods.

BTW, to do the tranny line vac test, i never really got up to 40-50 MPH. I floored it to like 30 then released pedal. Might vac drop further farther up the line?

Well, you have confirmed that you have proper boost. It probably could be a bit higher than 9 psi, but, that's not the source of your delayed shifts.

Furthermore, nothing in the vacuum system to the trans will delay the shifts to 4K rpm as you currently are experiencing. It would be nice to see the vacuum to the trans drop down to 2" on maximum power, but, it's not that big of a deal.

There have been several members who have suffered with a stuck kickdown switch. This switch is directly beneath the pedal and when you mash the pedal to the floor, the switch closes and creates a kickdown. Of course, if the switch gets corroded and stuck closed, the transmission will never upshift except at maximum rpm's.

So, see if you can find this switch and disconnect the wires to the switch.
Then drive it and see if the problem goes away. If it does go away, you simply need to replace the switch.

Your problem is not in the vacuum system or the boost. Both of these systems are OK. The boost could use some adjustment upwards and you can tweak the wastegate when you get some extra time.

Thanks, I will look. I'm not sure why the vac is not the problem though. If the tranny vac under hard but not insane acceleration only drops to 12", and only drops to 5" for a second (not sustained) when I floor it to the risk of all on the road, this doesnt seem to be the specs described elsewhere. Can you explain?

Again, thanks for sticking with me on this.

My system on the SDL is similar, but not identical.

It starts off around 11 and stays there most of the time until I give it heavy pedal. Then it drops down to 5"-6" or so. If I want to get it down to 2", I have to floor it and wait for full boost.

It sounds like yours would remain at 5" under full boost if you had the road to do it. I agree that 5" is a bit high, but, like I mentioned above, it won't delay your shifts.

Remember, that the vacuum on these vehicles controls the quality of the shift. It has nothing to do with the timing of the shift relative to rpm's.
You have been diligently chasing this vacuum issue in the valiant attempt to reduce your shift points. Unfortunately, you must abandon this approach for awhile and pursue a direction that will cure the problem. The vacuum system is not perfect, however, if it were perfect, and the vacuum dropped to 2" at full pedal, the problem with the late shifts would remain.

If you wish to confirm this (and I believe that you have already done this), simply disconnect the vacuum line to the trans and drive it with 0" vacuum. The shift points won't change a bit. It will, however, jar your teeth when it shifts.

For the love of God. For whatever reason I had become convinced that this was a Vac problem. Weeks ago I had given the kickdown switch a little inspection and thought it was ok, but did not disconnect it. Turns out, it was the kickdown switch. Although I feel like an ass, I learned alot about Vac system from Brian and others. You guys rock. Thanks alot. I owe you one.

Help is what we do here.

Having read thru this thread, I'm suspecting that my VCV is misadjusted or bad.

I have an '87 300D Turbo, build date 7/86, and from the photos Brian posted on the first page of this thread, the vacuum setup for the tranny is identical to that on his '86 300SDL.

I've checked the throttle linkage per the FSM, everything looks to be within spec.

Main vacuum supply at idle to the amplifier is 15".

Vacuum supply from the amplifier to the modulator at idle is 9".

The part that has me wondering, the output of the VCV to the amplifier barely reads 1" at idle. This was measured by disconnecting the hose at the Y connector, and connecting the vacuum gauge directly (no tee).

Revving up the engine to around 2000 rpm produced no change whatsoever in any of the above readings.

Should the output of the VCV be something other than a consistent 1" at idle and 2000 rpm?

Right now the car is in the body shop getting accident damage repaired, just wondering if there is a potential problem i should investigate once i get it back.

__________________
Just say "NO" to Ethanol - Drive Diesel

Mitchell Oates
Mooresville, NC
'87 300D 212K miles
'87 300D 151K miles - R.I.P. 12/08
'05 Jeep Liberty CRD 67K miles
Grumpy Old Diesel Owners Club

You've got some problems here and I suspect that it's some sort of measuring error.

Firstly, the main vacuum supply to the amplifier must be over 20" if the vacuum pump is working properly and the lines are plumbed properly. It's a direct feed from the pump.

Secondly, if you have 9" at idle to the modulator, the supply vacuum from the VCV must be more than 1". With the VCV at 1", the amplifier is going to drop that vacuum to the modulator down to 5" or so and the shifts are going to be extra firm.........to the point of banging.

Did you make these measurements with the engine running? You really can't use a Mityvac on a 603 due to the design of the VCV.

The data doesn't seem to jive.........and I suspect a measuring or plumbing error.

Found the procedure for checking and adjusting the VCV on a 124 with a 602 and 603 engine under the diesel injection section of the combustion manual in the FSM.

Have the engine idling and warmed up to at least 80 C, all checks on accelerator control already performed.

Connect vacuum tester to green damper on output of VCV.

At idle, should get the following readings -

602 - 420 +/- 25 mbar
603 - 385 +/- 25 mbar

Move the accelerator to the full load stop, reading should go to 0.

To adjust the VCV - loosen the mounting bolts, run the engine at full throttle, and rotate the VCV clockwise until resistance is felt. (IMHO, they forgot to list earplugs as required equipment for this part!)

If unable to obtain vacuum readings within spec after adjustment, replace VCV.

Doing a check in my engineering reference book, 1 bar = 29.53" Hg, so 100 mbar should equal 2.95" Hg. Doing a quick fudge calculation, 385 mbar should be around 11".

Under the same section, they also listed the specs for a go/no go test to determine if the vacuum pump needs replacement.

Hook the vacuum tester to the small port on the vacuum pump with the engine at idle. Tester should register 700 mbar after 30 seconds. If unable to meet the spec, the vacuum pump requires replacement.

On mine, you were quite right - it does hard banging shifts at light throttle.

Connecting a vacuum gauge directly to the small port on the vacuum pump, it immediately goes up to 22", around 750 mbar if my math is correct.

One potential problem which might explain the 15" reading elsewhere in the system - the filter which connects the small supply hose from the vacuum pump to the 5 way connector, is less than a snug fit in the 5 way connector, slides out quite easily. That will be the first item I try when I get the car back from the body shop - do a temporary sealing job on this connection and check the results, then start moving downstream in the system. Time to invest in a set of golf tees.

I've no idea if only having 15" vacuum supplied to the VCV would cause it's output to fall down to 1" instead of 11".

__________________
Just say "NO" to Ethanol - Drive Diesel

Mitchell Oates
Mooresville, NC
'87 300D 212K miles
'87 300D 151K miles - R.I.P. 12/08
'05 Jeep Liberty CRD 67K miles
Grumpy Old Diesel Owners Club