The Decel Valve Hose Diagram for the 2.0L Djet is Wrong, It doesn't work unless it's hooked up differently Options

[pbanders]

I made some postings before the "great split" about the decel valve where I asked questions on how it worked. I figured it out, and now I understand why I was confused.

If you look at the hose diagram Dave Darling made for the 2.0L, and I also believe it's the same way in Jeff Bowlsby's diagram, they show the "side" port to the decel valve connected to manifold vacuum, and the larger "end" port connected to the air box. George Hussey of Automobile Atlanta posted a diagram in a thread on the '74 hose configuration that showed it connected oppositely. Which one is right?

Turns out George's diagram is correct. As I complained in the earlier thread, I couldn't see how the valve worked when hooked up the way Dave and Jeff's diagrams showed. Not their fault - it's shown that way in the Factory Workshop Manual (see the FI manual, page 0.1-1/3, page 0.1-2/1) and all the other references I've seen. The problem is that if you pull the same manifold vacuum on both the side port (which is on one side of the internal diaphragm) and the skinny end port (which is on the other side of the internal diaphragm), there's no pressure differential to open the valve.

It didn't occur to me that the majority of diagrams might be wrong until I saw George's diagram. Now it makes sense - the valve works to limit vacuum if connected so that the large end port is on manifold vacuum and the side port is connected to the air box. Why? Because in this configuration, only the control side of the diaphragm is initially connected to vacuum (actually, it's slightly different from that - a very small area on the other side of the diaphragm is under vacuum where the valve seat for the large end port is located). Once the vacuum is high enough to overcome the internal spring resistance, the valve opens and the intake manifold vacuum is limited.

I tested three different decel valves on the bench, two used 914 valves, and a NOS valve I picked up on Ebay for a Volvo D-Jet application. I used a universal plastic tee connector to connect my hand vacuum pump to both the control port and either the large end or side ports of the valves. In each case, if I connected vacuum to both the control port and the side port (same vacuum across both sides of the diaphragm), the valve never opened up to 25 inHg of vacuum. If I connected the vacuum to both the control port and the end port (vacuum differential across the diaphragm), then at an onset vacuum, the valve opened, and I could not pump to a higher vacuum level - the vacuum was limited, as the valve is supposed to do.

Both of my 914 decel valves had been "adjusted", so I had no idea of what the correct onset vacuum should be. I recently had my motor rebuilt and did some manifold vacuum level testing. Fully warmed up, at idle, I am running about 10 to 12 inHg of manifold vacuum. If I rev the motor to 3500 rpm and snap the throttle shut, I see a maximum of about 22 inHg of manifold vacuum. The decel valve for the Volvo still has the factory paint mark on it and hadn't been tampered with. Measurements on that valve showed an onset of about 15 inHg. Seemed like a good number, so I adjusted my valve to that level and reinstalled it.

Now, when I test it as shown in the Factory Workshop Manual, where you open the throttle to 3500 rpm and snap it shut, you can feel it pull vacuum when the valve opens. I haven't had time to drive it around much since then (working on other problems), but as soon as I do, I'll report on any issues. I may increase the onset vacuum to 18 inHg, as when my motor is cold, it develops about 15 inHg at idle, about the same as the decel valve, and I don't want it to act as a leak.

[McMark]

It doesn't lean out the mixture as much as you're thinking because D-Jet is vacuum based. So this 'controlled air leak' will lower the vacuum from really-high-overrun-vacuum to somewhat-normal-idle-vacuum.

I wonder if disabling the deceleration valve would cause high overrun vacuum to hit the MPS and cause the MPS to fail prematurely... (IMG:style_emoticons/default/idea.gif)

[arvcube]

It doesn't lean out the mixture as much as you're thinking because D-Jet is vacuum based. So this 'controlled air leak' will lower the vacuum from really-high-overrun-vacuum to somewhat-normal-idle-vacuum.

I wonder if disabling the deceleration valve would cause high overrun vacuum to hit the MPS and cause the MPS to fail prematurely... (IMG:style_emoticons/default/idea.gif)

That makes sense...I think the functionality of the decel valve only applies to a limited range, dependent on how the spring pressure is set. So if the MPS measures manifold pressure, then high vacuum (idle or overrun situation, throttle closed) would cause the MPS to signal a leaner condition, while low to zero vacuum (WOT)would signal a richer condition...? if that's the case then maybe the decel valve was used as an additional way to lean out the exhaust emissions during these high vacuum situations even though the MPS and the "ecu" already does most of that....? (IMG:style_emoticons/default/wacko.gif)