WUR Pressure Hot?

[Funracer]

Chasing an intermittent small idle issue. 1983 CIS with -90 WUR rebuilt last year.
Checked WUR pressures many times over the last few weeks 2.8 bar cold 3.7 warm both in spec. Decided to leave the FP gauge on today and drive the car. By the time the engine was good and hot (as opposed to just warm) I pulled over and checked the WUR pressure it had dropped to 3.4 bar, the low end of spec.



I have an AFR gauge installed and can see the AFR richen as the WUR pressure drops from 3.8 to 3.4 bar. This raises my idle from 950 to 1200 rpm.

Is this normal WUR behavior?
Thanks

[Walt Fricke]

When cold, the bimetalic spring inside the WUR case presses down on the top of a spring (actually, two concentric springs), which, through a rod, lessens the pressure on a diaphragm, giving more room for the fuel coming in to pass through, in turn lowering the control pressure in the fuel distributor, which allows more fuel to flow to each of the injectors for a given flow of air.

As the bimetalic spring warms up, it bends up so it exerts less pressure, which leans the mixture. At some point it bends enough so it doesn't even touch the cap on top of the pressure setting spring system. At that point, the system should be at the warm control pressure, and not change until the engine turns off. The resistance heater on that spring is always on when the engine is running. And, per the Bosch manuals, pure engine heat spreading through everything should be enough to move the bimetalic spring into its hot position and take it out of the equation.

I'm not familiar with the WURs which have extra spring heating stuff inside - maybe that also shuts off the heating element when the engine gets hot enough? But no matter - engine fully hot, bimetalic spring should do its job of getting out of the way.

And I bet you know all this.

Some early WURs also used air pressure to affect the main spring pressure on the flow control diaphragm. The WURs have an upper and a lower chamber. The chambers are separated by a small diaphragm. I think it is like what is in an aneroid barometer, but not sure that matters. The outer of the two pressure control concentric springs rests on the WUR cast body on its lower end. The smaller inner spring, however, rests on that diaphragm.

If you put differential air pressure into the two chambers, the diaphragm, and thus that spring base, moves. This was used on some models as a barometric adjustment - lower upper chamber pressure, more spring pressure up top where it counts, leaner mixture. Alternatively, lower pressure in the bottom chamber, richer mixture.

Your WUR doesn't make any use of this middle diaphragm (unless its trapped internal air pressing against atmospheric leans the mixture some as you drive up a mountain), as no vacuum lines are attached to your WUR.

The braid covered hose attached to the top of the WUR is not there to pull a vacuum. It is just a clean air vent to the atmosphere (or close to it), as it connects to the throttle body above the butterfly. The lower chamber is also vented to the atmosphere, but it doesn't matter if some dust gets in there.

So here is my WAG: Your upper chamber vent has somehow gotten blocked. When the whole body of the WUR gets fully warm from engine heat, the air inside can't escape, so it presses down on the lower diaphragm, with the same effect as the bimetalic spring system did.

A test of this theory would just involve pulling this upper hose off the WUR.

How's that for something I'd never thought about?

Here is a nice schematic of this system. It isn't for the 090, but I believe the system functions are the same.



Also a picture of the fancier resistance heater system I suppose is in the 090?
And a line up of different WUR setups for air pressure or not.

[Brendatrent]

Warm-up regulator ensures that the engine runs smoothly when looking for one cold start slowly getting warm. It is directly connected to the motor so that it can measure the motor temperature directly as a controlled variable. The Bimetal spring is heated in the process. KohlsFeedback.com

[Dave Kost]

Funracer,

I had the same phenomenon happen to me driving to the Pittsburgh Vintage Grand Prix this year. I have a 1982 911 CIS with Lambda . It always warm starts at 1100 rpm then settles in at ~900 rpm.

Well, driving to the PVGP, we hit a ton of traffic, I mean - drive 20 feet then stop, 20 feet then stop, for 40 minutes or so. This was Saturday mid morning temp pushing 90 degrees which very hot for Pittsburgh. I too noticed that the idle slowly crept up to -1050 or so.

Around 2:30 a thunder storm was coming so we left early as did hundreds of others. Again in stop and go traffic but 20- 30 degrees cooler. This time the idle stayed at ~900.

My theory is that these engines grow as the temperatures rises. Maybe that induces some air leaks. The throttle is mechanical too. A 356 guy told me that their engines grow 1/8 inch when heat soaked. Same scenario with the 356 - Idle increases when heat soaked. Asked what is the solution- he said just live with it.

Also, after 40 years or so, the bimetal strip and spring may/can deform and wear, so the used WUR doesn't respond as a new one would. Again just a theory.

Also remember these lambda CIS cars were an after thought to meet USA emissions so Porshe could sell their iron over here.

[boyt911sc]

Originally Posted by Brendatrent

Warm-up regulator ensures that the engine runs smoothly when looking for one cold start slowly getting warm. It is directly connected to the motor so that it can measure the motor temperature directly as a controlled variable. The Bimetal spring is heated in the process.

Most of the informations above are NOT CORRECT. We are talking about Bosch WUR #0-438-140-XYZ. Do you have some experience calibrating this type of warm-up regulators? The WUR you were describing is not to be found in any CIS Porsche motors. Please educate us. Thanks.

Tony

[T77911S]

could be the FV/lambda changing the pressure once warm.
you could check the duty cycle of the FV when its 3.7 vs 3.4



dave
i was once told that the 911 engine does not "like" long idle times. when setting mixture with a gas analyzer. i was told you have to get it set fairly quick due to long idling. dont know if this is true.

how much time are you talking about when checking and it drops to 3.4.

sounds like you may be too lean if the idle comes up when the mixture richens.

[Dave Kost]

T77911S thanks for your comments.

I set the mixture by the Porsche spec book for lambda 911's. That was about 3 years ago, and nothing has changed since.

Car always starts at a flick of the key, idles at 1100 cold, 900 when warmed up. Never lean back fires or pops - unless I pull off the FV relay under the seat. I don't race the car, so I'm not worried about lean running under heavy load and high RPM because I never go there. I personally don't think the lambda cars were meant to be raced or to sit in hot traffic for extended periods.

Like I said the Lambda 911's were designed to meet emissions in the US. If you have modified exhausts, cams etc, there will be trade offs. I have SSI's and 964 cams and I run pure gas.

Remember these are 40 year old cars and I don't expect them to be perfect all the time , just most of the time.

I was just curious why Funracers fuel pressure changes?

Other that staying out of hot extended traffic - I have no worries.

[T77911S]

if you my run closed loop you have to set per the spec.
if you run open loop all the time you can run it where you want, well where it wants.
i have found they prefer to set more to the rich side. just on the edge of wanting to surge or RPM dip when letting off the gas. i always set my (non lambda) just lean of those two.
me, i would start with best/highest idle hot. (if open loop)

[theiceman]

lower end of spec is still in spec... dont go chasing things that might not be the issue .. if its in spec its in spec.. thats the way i see it anyway .. is your throttle body closing all the way and is smooth in operation ? do you have cruise control that might be binding things up > how old is your o2 sensor ?

point is dont get target fixation in one area , look at the system , not just one part.

[ahh911]

Walt,
The 090 has no bottom diaphragm, just a spring to the bottom half of the case. Easy to check if the vent is blocked anyhow, but I find it hard to believe it could be blocked, it's a fairly large diameter.
If it was blocked, which it isn't, (and by some chance the pressure inside builds, which really isn't possible, I should just delete this whole line) the control pressure would go up with temperature not down after a long heat soak.

I have cracked open the return line from the wur and there was a slight difference in control pressure...

If I set my warm idle close to the distributor advance rpm with the engine warm and oil up to temperature (Primary thermostat open, the one to the trombone still closed) then when even hotter (the second thermostat opens) the rpm will increase 100 or 200 rpm, then when very hot it will return back to set idle. I set my rpm when hot to just under the point where the distributor starts to advance. That solves the idle increase issue, 800rpm I think. (these characteristics are for my stock 1981 SC, usa, with O2 operating, I have no idea what will happen with it disconnected)

The OP still is asking why his control pressure changes between hot and very hot? 3.7 to 3.4.
It has nothing to do with vacuum leaks. Right, it's an 83, usa version?

Phil

[Funracer]

Good discussion guys thanks for all the quality input. Got to leave town tomorrow morning for a long weekend. I will look into in more detail next week.
Regards

[Walt Fricke]

Phil

The bottom diaphragm to which I refer is the one shown in the diagram as constituting the base of the inner spring - whose excursion is ultimately limited by part of the case, as shown. I'm pretty sure that the architecture of the 090 is the same in this regard, even if some values may be different. And I'm thinking of pressure in the upper chamber, which is above this diaphragm. Pressure there will depress the diaphragm, reduce the pressure on the fuel control diaphragm up top, and lower the CP. Far from certain that the subject car's system has any blockage there, but pressure above atmospheric (which is the pressure on the underside of this diaphragm) will lower the CP.

Agreed that air leaks aren't going to change CPs, though they can change AFRs and affect idle. A bit hard to imagine one which only appears when the engine is especially hot, though.

The Lambda/O2 sensor system is a bit complicated here, because if working well it can affect idle (not in the way described as the problem). The system reads from two temperature sensors, one which kicks in at 15*C, the other at 35*C. Per Jim Williams' research, on this late US model it provides some temporary idle enrichment when you go from cruise to idle. This one supposes this is to prevent the engine dying before it stabilizes. But after a few seconds this effect goes away. There are different possibilities for all this idle business in the Lambda system one supposes.

However, none of that affects control pressure. The control pressure you measure is, I believe, entirely controlled by the WUR and its springs. The FV acts on the differential pressure in the fuel distributor, not up top above the moving piston which controls fuel flow based on what the WUR sends as pressure. The differential pressure system, if I understand it right, affects fuel flow to the injectors. Before the FV system, this pressure differential between the lower and upper chambers in the FD is set at 0.1 bar. Varying this with the FV (which has a fuel supply independent of the return flow line from the WUR) is how the Lambda system is able to go lean when the O2 sensor says too rich, and rich when showing lean. All this could account, if not working right, for the AFR difference seen on the AFR gauge while driving. Not sure how it would also affect the idle CP on the pressure gauge as noted. Still seems a mystery, though far from as bad as some CIS problems when it comes to using the car.

[ahh911]

Walt,
Below is a picture of the bottom half case for my 090. The spring sits directly on where I've placed the teflon, which was only for a test, underneath it is aluminium of the case itself. I don't have any other rubber diaphragm except for the one on top where the fuel travels above.

Also, Walt, the control pressure is maintained by the wur as you've mentioned, but not exclusively as it must be proportional to system pressure and it is also affected by the return line pressure, if it's blocked then it too can affect the control pressure. I don't really see any other possibilities except if the source path to above the FD plunger to system pressure is flawed.

Phil

[Walt Fricke]

Phil - thanks for the picture. I assume that if the teflon were removed, one would see that entire area is just cast aluminum or whatever the casting is? It wouldn't look like what the earlier WURs (this is an 033) show in this view:



The round part with the red grease is the top of a threaded brass cylinder, which was installed from below. It is the base for the large spring, and doesn't move (though in theory you could screw it in or out some if you tried hard, and some WUR designs - cars other than Porsche used CIS - had a hole for a large hex key to adjust it from below). The silver part in the middle of that area is the base for the smaller inner spring. Below that out of sight is a diaphragm, as shown in the diagram previously posted on which the movable inner spring mount rests.

Does your (the 090) have an inner spring? Because that is only of use, really, if it allows something more than just a different, single outer spring.

I have been assuming that Bosch did what it usually does - make as few changes to tooling and so on as is absolutely necessary. But I can see that the Lambda system could do for ambient pressure differences what earlier bits did.

Pinching the WUR outflow line raises CP, does it not? Unlikely to happen absent messing around with that area of the engine? Hard to see how it could reduce CP.

I've pondered the effect of system pressure on all of this. Funracer could perhaps check his system pressures (which ought not to change no matter what temperatures there are?) cold, warm, and then if he can induce the puzzling full hot condition. My take is that as long as the fuel pump can produce a raw pressure well above system pressure, the pressure reducing system would keep system pressure extremely stable. Weird things can happen to that pressure regulating stuff, though - O rings in there, for instance. Hard to see how heat related.

If the pressure from the fuel pump fell below the regulated pressure, then yes, I suppose the CP would follow suit. The CP flow is from the lower chamber's system pressure spaces through a tiny pinhole in the main diaphragm to the top of the cylinder in which the fuel metering piston runs, and Tees over to the WUR (not much flow there), which regulates it as CP.

Maybe the stainless thin flat diaphragm type springs, which depress to let fuel through, are more sensitive to heat than one might think?

[boyt911sc]

Walt,

The picture of the WUR-033 you posted is from one of my posts I used for illustration. Let me show some physical differences between WUR-033, WUR-045, and WUR-090:

WUR-033:
Vacuum assisted (top)
Double springs
Enrichment section (bottom)

WUR-045:
Vacuum assisted (side)
Double springs
Enrichment section (middle)

WUR-090:
Non-vac assisted
Single spring
Enrichment controlled externally by different devices (lambda system).





Recognizing the differences in their external and internal configurations would be helpful in understanding how they work.

Tony

[ahh911]

"Maybe the stainless thin flat diaphragm type springs, which depress to let fuel through, are more sensitive to heat than one might think?"



For the 090 wur there are only 2 connections, for control pressure to head back down once hot it's either the pressure coming in is dropping, there is a restriction in the outflow that is heat sensitive, i.e. the pressure regulator o-ring for the wur return is acting up and allowing more flow when hot, or the WUR itself is sensitive in some manner. (This assumes that the heater inside the wur doesn't break connection with 12V causing an open as the engine gets hot for some reason.)

To isolate the wur from the other possibilities, i'd do the following test:

O/P states that control pressure is normal when warm, but when further heated it drops back down from 3.7 to 3.4.
1. Record temperature of case WUR when this pressure reversal occurs while driving.
2. Fuse pump relay removed and fuel pump bypass switch installed.
3. Hook up wur to spare battery and charger, make sure connection is secure, I remove the two mounting bolts to the wur and insert a foam strip around the wur to insulate it from the rest of the car for these tests.
4. Monitor wur temperature. Periodically switch on the fuel pump, say every 5C and measure system and control pressure. System pressure should be stable as nothing else is being heated.
5. Don't go nuts and melt anything, but if temperature can't be reached, use a heat gun delicately to raise the temperature with some patience, it's very stable, so heat it up, then let it stabalize and take the temp.
6. Look for pressure reversal above set temperature plus several degrees no doubt.

A less thorough approach but maybe a more practical first attempt might be to just hit WUR with the heat gun after the drive and see what happens, while pointing a I/r temp reader on it till it reaches the temperature recorded when the pressure drop occurs. Not melting anything, but I don't like to leave the car idling while I'm fooling around with this stuff and other components are heating/cooling while you take the test.

Personally I'm not sure I'd be bothered as both warm values are in spec, above test for me would be for the experimental "fun" of it, and see if something comes out of it. It would be easier and mabye more meaningful if the wur master (Tony) would let us know if this drop in control pressure can on occasion be expected and for what reason on a 090 WUR alone.

[Walt Fricke]

Thanks to Tony, I now realize that the 090 omits the middle section and has a new lower system casting and is less complicated. So much for my WAG on this model, but the mystery remains. We may all learn something about differential diagnoses of the CIS from this yet from Tony.

As for Aah's guesses about the bimetalic spring heater, I probably should shut up because I don't know how the extra stuff associated with (I suppose) the WUR heater coil works. However, the circuit diagram for these models doesn't show a temperature controlled switch (same for the aux air valve). And, at least for the earlier WUR models for which Bosch has manuals, it is asserted that once the engine warms up, conducted/convected engine heat alone is enough to cause the bi-metalic spring to lose contact with the main spring's top cap. So it is hard to see how the heating element could be involved in this issue.

The aux air valve might somehow affect the idle, as it is basically a controlled air leak, which is closed off when warm. It operates pretty much as does the WUR's heat effect system, and full closure of its heated spring causes the valve to close against a physical stop. Maybe if its spring won't close this valve as it should, leaving an air leak, and the system's CO value is adjusted when warm but the valve isn't closed, the extra heat after some hard running might finally close it completely?

But doesn't the Lambda system when working correctly, at least above some temperature, control the idle AFR in case it doesn't close? I forget at what temperature you are supposed to set the CO level? But how would that connect to the warm vs really hot CP even if it explained the high idle?

If this were my car, I'd agree wih aah - in spec is in spec. But as I now race the car, having a lower CP when really really hot might be an advantage. These CIS engines produce more power with lower CPs/richer running - low 13s to high 12s for the AFR. Though the high idle might be irritating.