There has been just so much energy put into this one, I just had to post it that way. I did not mean to bring so much pain.

Yes, I did join the federation after much searching on the tdi forums.

Step 1 : Buy this device - Airtrol F4200-X30
Step 2 : Buy this diode from radio shack. I can not find the number. I will have to add it later.
	The diode is IN4004 purchaseable at radio shack. Please report back with results. Failed or success. 
	Sometimes it takes a little tweaking here and there but once you get it you are golden.
	No CEL and I have logged over 2,000 miles.
Step 3 : Set the device to switch at about 19 or 20mm Hg
Step 4 : Snip the last number wire on the maf. 
	It is number 5 or 6 depending on the maf used. Mine was yellow. 
	I snipped it next to the computer under the plastic cover. Just make sure you have the proper one. 
	Also the device fits neatly in there and it is quite obscure.
Step 5 : Hook top (com) terminal to wire to maf, hook middle (NC) terminal to diode, hook bottom (NO) terminal to other end of 
	diode which also has the wire to the computer hooked to it. Just make sure the diode flow is going from NC terminal to 
	NO terminal. All connections must be really good or soldered.
Step 6 : Hook vacuum hose that goes from solenoid to egr to the device instead.

When the computer sends vacuum via the solenoid to open the egr it will switch the device instead. The diode will create the necessary 
voltage drop that the computer is suppose to see from the maf. When not activated the circuit functions as normal.


http://www.airtrolinc.com/components/f4200x.htm

The F-4200-X series vacuum to electric switch is designed for the customer with cost and space limitations, and is available 
in adjustment ranges from 4" WC to 30" Hg full scale. A wide variety of options including gold contacts, factory pre-set 
and special deadbands, make the F-4200-X series an excellent choice for the O.E.M.
	
CONTACT FORM 	
Single Pole, Double Throw
TERMINATIONS 	.187 Quick-Connect
AGENCY COMPLIANCES 	UL/CSA Approved Microswitch
OPERATING SPEED 	25 MSEC
MATERIALS 	Body..Polysulfone Diaphragm..Polyurethane Spring..Stainless Steel Others..Nylon, Carbon Filled Nylon , Acetal
TEMPERATURE RANGE 	40° TO 150° F [4°-66° C]
REPEATABILITY 	< +/- 2% F.S. OR .15" Hg, (3.8 mm Hg) whichever is greater, after 10,000 cycles


http://www.peachparts.com/shopforum/diesel-discussion/168261-got-time-egr-bypass-606-turbo.html

------------------- ------------------- ------------------- -------------------
http://www.peachparts.com/shopforum/diesel-discussion/168261-got-time-egr-bypass-606-turbo-7.html

It's actually a pot. Here's the theory of operation of the EGR operation validation:

The EGR transducer is fed a unique 12v from the k40 relay - hot with ignition. (red/green wire in the picture) The transducer is connected to ECU pin 35 by another wire (green/gray). When the proper conditions are met, the ECU makes pin 35 low to activate the transducer and open the EGR valve.

When the EGR opens, there is a corresponding decrease in the flow of the intake air through the MAF sensor which is read as a voltage drop on the MAF signal feed to ECU pin 21. (yellow/white).

Problem 1: You can't just unplug the transducer because the ECU wants to see a load across the EGR transducer power feed and ECU pin 35 or a code will be set.

Problem 2: You can't just plug the vacuum line going to the EGR valve or plug the EGR feed tube because the ECU won't see the desired effect of the voltage decrease on pin 21 and again, you get a code.

So, to tackle problem 1, I first snipped the ECU pin 35 feed and then simply placed a 1/4 watt, 1k resistor inline between that and the EGR transducer power feed to simulate the load of the transducer. Now, that creates problem 2 because the EGR isn't opening and creating the MAF flow reduction.

To deal with problem 2, I needed to simulate the voltage reduction on ECU pin 21 on cue from ECU pin 35. Conveniently, when pin 35 goes low it makes a great place to shunt off some of the MAF signal voltage. I inserted a variable resistor there to make adjustments and find out exactly how much resistance was required in order to get the desired voltage drop. Now that the circuit has cleared all the codes, all I have to do is measure the resistance at the current setting and then swap in a fixed resistor in place of the variable one.

The reason for the diode is that you have two voltages going to the same pin (ECU 35): The <=12v from the transducer feed and the <=5v signal feed from the MAF. The diode keeps the higher voltage from spilling onto the MAF signal feed and altering it's value.

So that, in a nutshell is how it works.

Update - Here's a preliminary schematic:
http://i3.photobucket.com/albums/y71/kartek/Mercedes/EGRDefeat.jpg