Boost Issue?

[piperporsche180]

Hello guys, here is my problem.

I will enter the freeway a little aggressive and my 1987 951 with minor bolt ons will haul @$$ through a few gears then I will slow down at highway speed.

Here is the problem, sometimes the car will go through first and second and slam me into my seat then it will feel like its naturally aspirated, it will feel slow. I will pull to the side of the highway, shut off my car and re start it then the car goes back to being fast. Do you guys know what the problem could be? Thanks.


Randy

[wild man]

What is the boost gauge reading when you lose power?

[piperporsche180]

It is reading 2 bar, there is no boost leak.

[wild man]

Wow - 15 psi and it feels naturally aspirated?? You've definitely got something strange going on. Good luck with this one. I don't think I can offer you any help - without a whole lot more information, anyways.

[piperporsche180]

Im thinking its a by-pass valve issue, Im upgrading to the Type RS by-pass valve. Im hoping that corrects the issue.

[nize]

what do the blink codes show?

[piperporsche180]

Im looking for a KLR diagnostic reader, is that the codes I will have to look at?

The dummy lights on the dash are all off.

[nize]

check this thread regarding the blink code tester;
Porsche 951 (944 turbo) reference list

[guru944]

Maybe Overboost(fuel cut limiter). 2 bar thats 29.4psi of boost, is too high on a street car. Get a real boost guage suggest Autometer or VDO.

1. Suggest reducing boost to 1 bar(14.7psi) or even lower (10psi or 7) and see if car responds the same.

2. Consider a manual or electronic boost controller.

3. Check fuel pressure and Use a Air/fuel meter to monitor part & full throttle fuel levels. You could also be setting off the knock sensor/KLR (check that as well).

4. Check maf meter for low signal glitches or intermittent operation.

hope this post was helpful...

later,

J

[piperporsche180]

The boost is not set at 2 bar, I believe it’s at 14.7lbs. The 951's boost gauge is deceptive. I believe 2 bar on the gauge is the equivalent to 14.7lbs. I’m thinking it could be the knock sensor. I’ll have to find a KLR testing tool for my laptop.

Thanks guys. I will keep you informed on the issue!

[333pg333]

Almost sounds like you're going into limp mode?

[wild man]

1. Fuel is NOT cut, if knocks detected, only boost is cut (down to 7psi, if stock, I think). I'm not sure about overboost (>17psi) though, as my autothority MAF conversion eliminates it because it runs at 20psi (but gives 320hp [310 on a non-S], without ANY other mods). Added: I think that it DOES cut fuel if overboost, but that is NOT stated as a symptom (I'm pretty sure you'd know it, if it was).

2. The stock vac/boost gauge displays absolute pressure, that's why 2 bar is really 1 bar of pressure (14.7psi @ sea level.

3. The car does NOT have MAF, unless it was converted to, with chip(s) that support it, but then NOT likely to have overboost protection.

4. If stock, I'm thinking boost WOULD be cut (to 7psi), if in limp mode. If that isn't correct, then limp mode would seem to make sense. The first 2 steps that it takes if detecting knocks, is to cut timing by 5 degrees (if I recall correctly) on each step. But that still wouldn't cut power by the amount stated in the symptoms.

[nize]

Originally Posted by wild man

1. Fuel is NOT cut, if knocks detected, only boost is cut (down to 7psi, if stock, I think). I'm not sure about overboost (>17psi) though, as my autothority MAF conversion eliminates it because it runs at 20psi (but gives 320hp [310 on a non-S], without ANY other mods). Added: I think that it DOES cut fuel if overboost, but that is NOT stated as a symptom (I'm pretty sure you'd know it, if it was).

2. The stock vac/boost gauge displays absolute pressure, that's why 2 bar is really 1 bar of pressure (14.7psi @ sea level.

3. The car does NOT have MAF, unless it was converted to, with chip(s) that support it, but then NOT likely to have overboost protection.

4. If stock, I'm thinking boost WOULD be cut (to 7psi), if in limp mode. If that isn't correct, then limp mode would seem to make sense. The first 2 steps that it takes if detecting knocks, is to cut timing by 5 degrees (if I recall correctly) on each step. But that still wouldn't cut power by the amount stated in the symptoms.

the above are true only if the car is stock. we already know this car is not stock, and if it's boosting to 2bar on the factory gauge we know that boost has been increased (stock boost is 1.8bar), or the gauge is malfunctioning.

1.8bar on the stock gauge = .8bar of boost = 11.6psi

if boost has been increased, most likely there is a boost controller. if there is a boost controller, it is definitely bypassing (ie; disabling) the factory cycle valve. if it's disabling the cycle valve, the klr can not pull boost for limp mode.

before we go down the path of incorrect diagnosis, speculation, and time-wasting, we really need;
1) a detailed list of specific mods on this car.
2) what the factory blink codes show.

without both 1 and 2, we're really all just wasting our time.

[wild man]

Yea, I overlooked putting 2 and 2 together, regarding the fact that he has a full bar of boost, which doesn't come stock. But if I may disagree about WHY it isn't stock, I think that it would more likely be that it is due to installation of of aftermarket chip set, with constricted banjo fitting (otherwise you wouldn't get claimed hp out of chips), than an MBC with a deleted cycling valve. But - I could be wrong about that, or, he could be one that does have those mods. Why don't we ask him (hint to him, if he sees this)?

[nize]

even with aftermarket restricted banjo fitting, the stock cripple/safety mode would be bypassed. pretty much any mod that increases stock boost will bypass the stock cripple/safety mode. it has to, otherwise you can't raise boost because the factory cycle valve will not allow it.

the only way i am aware of where you can retain the stock cripple/safety mode while increasing boost is via an aftermarket complete engine management system swap, where the aftermarket system has a cripple/safety mode ability. all this has already been discussed in various stand alone ems threads.

[jkb58]

Basically 3 safety functions of the stock DME/KLR:

1) overboost protection – which is a hard fuel cutout/dramatic stumble when the DME determines there is too much airflow through the AFM. This is a function of the DME and the setpoint is mapped on the factory DME chip. Usually trips when running higher boost (manual boost control modifications) with the stock chips. Typically occurs at higher rpm in higher gears, above 5k rpm in 4th or 5th.

2) “limp home” mode – a KLR diagnostic protection that cuts boost down to the “basic charge pressure” of approx 1.2 bar abs by opening the CV (and retards timing 3 deg) if it detects a fault with any one of various controls and parameters. The fault code can be found (using the factory plug or a test light made from radio shack parts…).

NOTE: the factory gage is not deceiving, it shows bar absolute pressure. 1.01 bar = 1 atm = 14.7 psi, so 2.0 bar abs on the factory gage is about 14.3 psi gage at sea level. Stock boost is 1.75 bar abs (+/-10%). So around 10.8 psig (+/- 1 psi). Anywhere around 10.5 to 11.5 is pretty normal for a stock car.

3) Knock control – KLR function of reducing knock if detected. It retards timing on the knocking cylinder by 3 degs. If knock continues it will retard up to 6 degrees, will also start retarding other cylinders and will also start reducing boost (by the CV) small amounts (I recall something like 200 mbar increments) until knock is no longer detected. The car will lose maybe 5-10% horsepower if the KLR starts pulling timing. Its not going to be a huge decrease in power.


Modifications to the stock parts and aftermarket replacement parts will often alter the DME/KLR’s ability to implement these various safety functions.

As requested above, the original poster needs to offer more information on the car setup – is it bone stock, or are there some modifications (chips, manual boost changes, etc).

[jkb58]

Originally Posted by nize

even with aftermarket restricted banjo fitting, the stock cripple/safety mode would be bypassed. pretty much any mod that increases stock boost will bypass the stock cripple/safety mode. it has to, otherwise you can't raise boost because the factory cycle valve will not allow it.

If the cycling valve is still installed and functioning, the KLR will have some ability to reduce boost (open the WG) if limp mode is triggered. Shimming the WG or putting a carb jet in the banjo will increase the peak boost, and delays the cycling valve response time, but does not eliminate the KLR/CV boost control. So, in limp mode the KLR will still kill the boost, it might be a little higher than the 1.2 bar w/ a bone stock system, but it will still be significantly reduced.

If completely bypassing the CV, and running a manual boost controller, dual port WG (not the stock WG), etc. - then the KLR has no way of reducing boost in limp mode.

[wild man]

I concur that chips do seem to eliminate the over boost protection, even though most run at about 15psi, which is 2psi less than the amount need to invoke the OB protection. And honestly, I don't believe that the so-called cycling valve actually cycles as much as many are led to believe. I believe it is almost always open, allowing 11psi determined by the stock constrictor, built in to one of the valve's ports. When boost is cut, it closes up, cutting off the bleed-off, and allowing full pressure to the WG, dropping boost to 7-8psi.

200 mbar increments? I think there is only one "increment": 11psi - 3 psi = 8psi. That is low enough for regular fuel not to detonate. But if running 20psi, it will only go down to around 12psi (in 1 increment, I'm pretty sure), too high to protect it from regular fuel. I'm still looking at what the best way is to fix that issue. I'm not sure if that in it self triggers limp mode, but it does make sense that boost should always be cut when in limp mode.

[nize]

Originally Posted by jkb58

If the cycling valve is still installed and functioning, the KLR will have some ability to reduce boost (open the WG) if limp mode is triggered. Shimming the WG or putting a carb jet in the banjo will increase the peak boost, and delays the cycling valve response time, but does not eliminate the KLR/CV boost control. So, in limp mode the KLR will still kill the boost, it might be a little higher than the 1.2 bar w/ a bone stock system, but it will still be significantly reduced.

If completely bypassing the CV, and running a manual boost controller, dual port WG (not the stock WG), etc. - then the KLR has no way of reducing boost in limp mode.

you are correct. i should clarify;
even with aftermarket restricted banjo fitting, the stock 1.2bar cripple/safety mode would be bypassed. pretty much any mod that increases stock boost will bypass the stock 1.2bar cripple/safety mode. it has to, otherwise you can't raise boost because the factory cycle valve will not allow it.

[wild man]

While I do hate to get arguments going, I'm going to have to beg to differ, regarding the statement that boost-cutting capability of the valve is "bypassed" if an aftermarket restrictor is installed. From what what I have experienced, that simply isn't the case. First of all, the STOCK setup does indeed set boost by way of a restrictor, or constrictor (take your pick). The stock restrictor is located in the center hose barb (nipple) of the valve. Installing an aftermarket restricted banjo fitting simply puts a smaller restriction at the opposite end OF THE SAME HOSE. And the smaller restriction will then simply override the larger stock one.

With a restricted banjo, boost cutting still works, but it is crippled to some extent. If stock can drop boost to 1.2bar, then a 50mil (20psi) banjo will only allow the valve to drop it down to about 1.7bar, but it still drops it from 2.3 (we are talking in absolute terms here).

At this point, the only thing that I am able to come up with that might lower it down closer to the stock 1.2bar safety level is to remove the restriction from the banjo end of the hose, and place it WHERE THE STOCK ONE IS LOCATED, at the center port of the solenoid valve.