I have no idea of how this thing can do what it does. I understand WHAT it's supposed to do - limit the intake manifold vacuum during overrun conditions - I just can't see HOW it does it. I've never seen the inside of one.

The valve has three ports, "control" (the skinny one on the end), "manifold" (the fat one on the side), and "atmospheric" (the fat one on the other end). The control and manifold ports are both connected to the branch "tee" fitting on the intake manifold, so they see the same level of vacuum.

If I remove the decel valve and bench test it, when I pull a vacuum on the control port alone, at some vacuum level (which is adjustable), the valve opens and I can blow through the other ports. However, if I simulate what the valve sees in the engine - that is, the control port and the manifold port both at the same vacuum level, as I expected, the valve doesn't open, no matter what level of vacuum I pull. I expected this result, because there's no pressure differential across the ports to open the valve.

What am I missing here? Clearly, the valve works in a car. How?

i'm not much help here, but i simply removed mine and plugged the hoses ...

i like the off-throttle response much better without the decel valve ...
Andy

Yep, know all that - just want to know HOW this thing works, thanks.

Brad: What I thought I learned from Paul Anders.
My vacume line is tee'd from the distrabutor line and the valve gets it's air from the air box and feeds it under the butterfly.
:PERMAGRIN: Joe

Hi Brad...long time no talk....

A decel valve in MY old time memory....limits the speed revs going DOWN from WOT to idle or from freeway speed to onramp speed.....

It sounds like the control (skinny) end would need to be above the throttle plate to work, no? How else can this valve work if there is no way it can tell if the throttle snaps shut??

Ah, someone who is beginning to understand my confusion. Actually, no, if the control line was connected above the throttle plate, it'd never see any significant vacuum under any throttle condition. When the throttle is WOT, the pressure in the air box is only a few mtorr below atmospheric.

The way the valve tells the throttle is snapped shut is that the manifold vacuum level abruptly increases. The decel valve is also known as a "vacuum limiter" - it limits the maximum vacuum in the manifold. And, I know that it does what it's designed to do, but I'll be damned if I can figure out HOW, given how it's designed and connected.

Yup, that's what it does - it's HOW it does that, considering the way it's designed and connected, that I'm after.....

I figure someone like Krusty can explain it.....

OK.....the valve is open when WOT or at sustained speeds...I'm thinking that the valve works when the revs go UP and then slowly closes due to it's internal limits on decel...?

Maybe there is a restriction in the control end, such that when the manifold is under vacuum, the control port vacuum internal to the valve lags, and holds the valve open slightly until the pressure across the valve equalizes?

You may not be seeing the correct behavior on the bench because you are not testing the valve dynamically.

I'll take a wild ass guess and say it has something to do with the pressure waves in the plenum. Pulling a vacuum on the decel valve doesn't really simulate real world conditions.

what he said

I agree that pulling the vac on the bench isn't the same situation. But how does it differ and why does that make the valve open?

As you can tell, this has bugged me for a while. I spent a couple of hours looking at Google and trying to find out an answer. Even poked around at the USPTO database. I'm sure the explanation is simple, I just don't see it.

Try Google using....add on smog devices from the 70s.....or smog devices that didn't work and are NLA....or suckass smog devices....

There's a guy I argue with at symposiums from the Ca Air Resources Board that was instrumental in some of this stuff....he's older than dirt and really believed in this crud back in the day. He was behind the repeal of the rolling 30 year smog exemption in CA. I'm surprised that no one has ever done a drive by on his office....I'll look him up and PM ya....

You mean like a metering port? That's an idea. I haven't seen any evidence of it, however. I'll check it out.

I tried using about a million Google terms, including "explanations for dumb-shits" but all I got back was my photo.

Most everything I found told me what the valve did, which I know - not how.....

Sounds like what Mark and Jeff are saying makes sense. Could it be as simple as the larger port will see waves faster than the smaller port, thus metering the valve back and forth over the atmosphere port?

We just need to find a junk one and cut it open.

FI Hose Diagram

I never had a decel valve.

But according to the above diagram the valve is purely a vacuum system. I thought it threw a switch to the FI brain.

The diagram shows the control (I think) attached to the reference (atmos) of the fuel pressure regulator.

Yeah, there I go, always assuming D-Jetronic without saying so, sorry. I'm talking about the way it works in the D-Jetronic setup, not the L-Jet setup. You're right, it's totally different in the L-Jet, I have even less of an idea of what it's doing when I look at that diagram.

Agree, that's what I want to do, I need to find a donor valve. I still don't think that the way it works is as complicated as being determined by pressure waves - I think it's probably really simple and I'm just missing it.

The other issue I have with the decel valve is: what's the right setting for it and what's the adjustment procedure? I know mine's been jiggered with (that's because I'm the jiggerer), and I'd like to set it back to the original setting. But the things have been NLA forever, so I don't have an OEM unit to compare mine to.

No wonder everyone rips it out. It's easier to do that.

LOL!

I just went to the best site for D-Jet information and read that he couldn't figure out how the decel valve worked either. Duh!!!!

I thought the control was taken from below the butterfly thus seeing the high vacuum under closed throttle deacceleration and it merely adds extra air to prevent backfiring.

Brad, have you tried hooking up the larger lines the other way around?

--DD

Nice thinking Dave. I bet that does it. Now I'm curious if there is a right way and a wrong way to install these.

It's hooked up the "correct" way, according to the Porsche Workshop Manual and Dave's own diagram.

I think I've got a clue. I have a branch "tee" in my parts bin. I tried and experiement with it, where I sucked through the port that goes on the manifold vacuum side. The diameter difference between the port that goes to the "control" port of the decel valve, and the port that goes to the "manifold" side of the decel valve is enough so that there is a very large difference in the resistance to flow, acting as a metering port. It's just like was suggested earlier, that the difference causes the vacuum level on the "control" side to respond much more slowly to changes.

That guy can be a dumbass, trust me. I know first-hand.

The only thing I can figure on this decel valve is it's acting like an AAR. Raising the amount of air flow to increase RPMs. When pressure is given on the throttle the valve inside the decel moves allowing more air to pass. When you let up on the throttle the valve or piston within the decel is pulled back to its original resting place (slowly) allowing the engine to decelerate slowly by restricting the extra amount of air. And, like said above, one will need to be taken apart to get to the skinny of it all!

(was this all said already)???

....b

well how is explained by why...

why: the valve was needed to control emissions runoff from the FI adjusting from part-throttle very quickly to idle. the valve slowed the abrupt transition and prevented backfire, aka unburned fuel exploding later down the line.

how: the port meters the air coming through the sensor only when high manifold vac is present. I bet there is some spring loaded plate, much like a variable fuel pressure regulator. I have a old D-jet decel valve I can send someone for dissection.

I think this makes sense, because unless the valve was designed to work on transient vacuum changes, there would be a risk that the valve would be slightly open all the time at idle. It would be difficult to guarantee the valve would open under overrun conditions but not at idle, because I think the vacuum levels are similar in both cases.

What do you think the adjustment does?

Valve Picture

Give it manifold vacuum, normally they (in general terms) draw air from the air cleaner and inject it into the intake manifold. Nothing complicated, the valve reads the change in manifold pressure (differential pressure) and injects extra air to lean out the mixture in a time its likely to otherwise become richer. Triggers much like a turbo blow off valve.

pssst.... the guy whose site you linked to is the one who asked the question.


i think he knows what the valve looks like.

Thanks for the heads up, now you know how it works.

oh, another thing - adding extra air to the manifold does not lean out the mixture, instead it effectively raises the idle speed temporarily by allowing air to bypass the throttle valve. unlike a carb, adding more air after the throttle valve also causes the injection system to add more fuel to maintain the correct fuel mixture (thanks to the MPS).

Cool, send it on, I'll characterize it on the bench, then I'll use my dremel to dissect it, and I'll post the whole thing up to my web page.

Thanks to everyone for putting up with my D-Jet obsession....

Well you're the main guy in D_jet, thank you for your hard work dissecting the rest of this 30 year old system.

Pm me your addy, I'll put it in a manilla envelope.

Actually, not. No offense here, but you don't understand what I'm getting at. Just knowing WHAT something does and WHY it does it, is not the same as knowing HOW it does it. I know exactly WHAT the decel valve does and WHY it's supposed to do it - I knew that ages ago. But I found that when I tried to understand HOW the valve worked, I realized that I didn't know, and that's why I started this discussion.

I know it may seem tedious, but asking these questions about 5 years ago regarding how the D-Jet system worked is what led to the web pages I have complied, with the help of many other people on this list, Pelican Parts, and Rennlist. Understanding HOW often leads to insights as to why the system responds the way it does to unaticipated conditions.

For example, I'm beginning to suspect that if you don't hook up the decel valve to the branch "tee" connector, and make other vacuum connections to the same sources (e.g. different hoses and/or ports), while it looks like it's correctly connected, it actually doesn't work the same way. The branch "tee" restriction seems to be an important part of the way it functions. Maybe this is one of the reasons why people have certain idle and driveability issues.

Lastly, if it's got an adjuster screw on it, after 35+ years, believe me, someone's taken a crack at it. What's the correct adjustment? Why? How do you calibrate it? Again, stuff I want to figure out and document.

Sorry for the misunderstanding, I see your point. The valve itself uses a vacuum diaphragm inside with two vacuum chambers. The chambers are connected via a bleed. Normally the chambers are of equal vacuum/pressure but as the throttle is closed the first chamber receives/holds more vacuum which moves the diaphragm in that direction. Thats what triggers the gulp of extra air to the engine. Since there is a bleed the chambers quickly equalize and the gulp session ends. My guess is you can move the diaphragm which may change the time element slightly. I don't know, must be time to open one up.

Picture of a Gulp valve I think its much like the de-cell unit.

I see the mps and the leaning of the mixture is relative, an abrupt throttle closing will richen the mixture. Using the de-cell valve helps control that.

I bet its something similiar to a fuel pressure regulator. Two chambers, and one sping loaded valve that slowly equalizes pressure in the two chambers, something to that effect. I'll get that decel valve out to you on monday.

I know most djet engines I have seen don't have the factory setup. They are usually missing the rubber elbow and have a plastic tee instead. I haven't ever taken one apart either so that would be interesting to see the how it works.

Why does it need two vac lines pulling on it and one from the air cleaner? Is there a diaphram between the two vac lines?

The decel valve operation is very simple....

It does NOTHING until the throttle is slammed closed at high engine RPM. Then the high engine vacuum on the small line pulls the decel valve open. That creates a way for air to bypass the throttle plate, and slows down the engine deceleration. It is supposed to lower emissions and stop backfires.


Personally, I like the engine spinning down faster. Makes for better engine braking coming downhill or on an AX course.

I took the one on Betty's car off and replaced it with an idle up solenoid for the AC.


I think the fact that you have "readjusted" yours accounts for the fact that you can't get it to open on the bench.

Are you pulling the vacuum on the correct big line? If you pull vacuum on the air cleaner side of the valve, it will never open by pulling vacuum on the control line. The diaphragm is sucked up against the housing. Pull vacuum on the side and see if it opens.

Weird - I posted a response to Clay's post last night, but in didn't show up - I probably didn't click the "Add this" button after previewing...

Anyway, I'm pretty sure I figured out HOW it works last night, and it doesn't involve pressure waves or throttling orifices (well, perhaps throttling is important, see below). I'll know for sure when I open one up.

The key is that it takes a pressure differential to open the valve. But if both sides of the diaphragm are evacuated, then how do you get a differential. Simple, both sides of the diaphragm are NOT evacuated. I thought the guts of this thing looked like the guts of the fuel pressure regulator, where the "output" port has a relatively small seat area on the diaphragm. In the decel valve, the end that supplies air must have a seat that trumpets out inside of the valve, so that it covers nearly all of the diaphragm area. That way, most of the non-control side of the diaphragm is at atmospheric pressure.

When you pull a vaccum on the control side of a sufficient level to overcome the spring tension, the diaphragm moves and the seat opens up, allowing air to pass through the valve and limit the vacuum level in the intake manifold. Where I think the throttling is important is that the vacuum in the control chamber side bleeds down slowly, allowing the valve to stay open longer.

Clearly, some fraction of the non-control side of the diaphragm must be under vacuum, which explains why when I pulled vacuum on both the control port and the manifold port, my valve didn't open. I've probably got mine adjusted way too high and that's why it acts that way - or perhaps, something is wrong with my valve.

I have a 75 BMW with L-Jetronic. I seem to remember the decel valve allows extra air in during high vacuum conditions (i.e. deceleration) leaning the mixture and slowing the apparent throttle resonse on deceleration.

I do know friends behind me report a brief flame when I decelerate with this valve disconnected (unburned fuel without the extra air). I keep it connected except during testing.

Randy

OK, a bit more on this. I was poking around my parts box, and I found that I'd bought a new decel valve some time ago - same Bosch part design, but for a different application (it's a 0 280 160 133). Since it was nice and clean, I could look into the port on the side and see the operation of the valve.

Looks like my idea that the atmospheric seat trumpets out is bogus - the atmospheric valve seat is no bigger than the diameter of the port. I can also see a small helper spring between the port and the diaphragm.

The valve operates exactly as I would expect it to - when you pull a vacuum just on the control port, it opens at an onset level, set by the adjuster - in this case, about 15 psig. However, if you pull a vacuum on both the manifold port and the control port, the onset increases. If the vacuum on the manfold port is the same as the control port, for vacuum levels I was able to achieve with my hand pump (~ 25 inHg, about 3 inHg more than I've seen in overrun manifold pressure on my newly-rebuild 2.0L), the valve NEVER opens.

Still wondering how this thing works......

You still need my donor decel valve?

Yes, I want one to cut open. The one I have is new and a non-Porsche one, I'll put it up on ebay eventually.