964 pinging

[scotsman60]

My '90 C2 is pinging. In fact it's been pinging for a while but I can drive around the problem. This week we had 90+ temps in Souther Californai and the pinging is getting to be annoying enough for me to do something about.

Here's what I've got and what I've done so far,

1) I'm running 91 octane gas - it's the best I can get here.

2) I've run a bottle of Techron through the system and it didn't make any noticeable difference. (I changed the oil afterwards)

3) New plugs, new leads, new distributor caps, new rotor arms and new coils - No difference.

4) I'm getting a spark on all plugs - distributor drive belt checks out OK


I've got some questions,

1) If the knock sensor detects pinging and adjusts the timing at any time should I get an engine warning light? (I haven't got any lights)

2) I'm thinking of hooking my DVM up to the knock sensor wire. Anybody got any ideas about what kind of signal I mugh expect to see and how it should change when pinging occurs? (I'm going to assume to start with that I'll see constant voltage and then a change in voltage as pinging is detected)

3) It's possible that the knock detectors are either faulty or incorrectly attached. How do I access them? If is remove the fan and the alternator can I get access to them? If I'm in there I'll probably replace them anyway - any special things to know about this procedure?


Any and all ideas and suggestions gladly accepted......






Doug

[Steve@Rennsport]

Hi Doug:

When a 911 pings, its damned serious business and the car should NOT be driven until the cause is found and resolved. The consequences are far from being just "annoying". You will not get any warning lights; the ECU does its own thing and cranks back the timing until it ceases.

I would first mix 10 gallons of unleaded race gas (Unocal 100) in the tank and see if that make any difference.

Is the car chipped?

How many miles on the engine?

[scotsman60]

Steve,


Thankx for your reply. I'm aware of the potential seriousness of the pinging and I didn't mean to appear flippant, however from discussion on the Rennlist board it appears that pinging in 964's and 993's is fairly common in California in hot weather.

The car has a standard chip and has 74,000 miles.

Today the temperature has dropped to around 60 (from 90) and I have no pinging whatsoever. I have located a gas station (76) that sells 100 octane gas and I plan to mix five gallons into the next full tank - probably tonight. Of course, I'll need to wait for hotter weather to know if it really makes a difference...

If the problem goes away by increasing the octane I will try and figure out what mixture I need and may also try octane boosters - do you have any experience with octane boosters? Can you recommend one? Any issues with octane boosters?


I'm still suspicious of my knock sensors. As far as I can tell the DME can retard the timing by up to 9 degrees in response to knock sensor signals. If it's still pinging at this point I should get an engine warning light but I'm not.



Doug

[Steve@Rennsport]

Hi Doug:

I don't need to tell anyone that the C.A.R.B. really didn't do anyone any favors when they downgraded the pump premium from 92 to 91 and thats simply not good enough for many cars in hot weather. Many 911 owners now know to use some unleaded race gas in hot weather as SOP.

The air-cooled 911 is one of the most sensitive and susceptable powerplants in the world as cylinder head temps and therefore detonation thresholds change with ambient temperatures. In other words, 91 octane fuel is simply not up to what Porsche specified and you must simply mix some race gas with pump premium in hot weather. Failure to so will result in a whole bunch of broken rings. The knock-sensors do not have a real wide range to handle that fuel.

Octane boosters have a few serious drawbacks.

First, you cannot achieve a homogenous mix in the tank so each and every tank is different and inconsistent. Second, octane boosters make a mess inside the engine and leave deposits on plugs, valves, and piston crowns. Lastly, they really do not offer a very big gain in octane for the dollars spent. Race gas is FAR cheaper and much more effective.

[scotsman60]

OK,

I mixed 5 gallons of 100 octane ($4.50/gallon) with 13 gallons of 91 when I filled up last night.

The pinging is almost completely gone and the motor appears to run cooler!!!

I'm feeling better, but not completely satisfied for two reasons:-

1) I got stuck in stop and go traffic going through Laguna Beach and the motor temp went up. The needle went to just above horizontal - about two thirds of the way between the first and second tick. At this temp the engine pinged under acceleration from low revs. I can live with this fairly easily because it's easy to avoid - keep my foot off the gas (OK - not so easy....)

2) 100 octane gas is not that easy to find and it's expensive.

I'm going to research octane boosters - this may be more economical and easier to live with than trying to find 100 octane all the time. I'm not convinced that mixing is an issue - if I put the booster into an empty tank and then top up with a full tank of gas I reckon that the mixing will be pretty good. If they all make a mess insude the engine then that would be a concern.....

Any suggestions for keeping the temps down in stop and go traffic? I have the under-tray removed already. Can I use a different thermostat (or adjust the existing one) so that the oil cooler kicks in earlier? Will this have a significant effect on the cylinder head temps?

[scotsman60]

Steve,


When did premium go from 92 to 91?

I know that it's been 91 for a while now but I'm trying to figure if the switch happenned while I owned my 911 or before.....




Doug

[Steve@Rennsport]

Doug:

1) When the oil temps went up, so did the cylinder head temps and so down went the threshold of detonation: voila,....pinging.

2) Caveat Emptor on octane boosters. You were warned/informed,..use that stuff at your own risk.

3) California dropped MTBE from all pump gas in late 2002. Thats when it changed from 92 to the current 91.

If you want to truly help prevent pinging, do something worthwhile to reduce your operating temps. The very best and ONLY effective was is to add an additional front oil cooler. Its not cheap, but you get what you pay for. The thermostat is not adjustable.

[TOMADVAUTO]

Usually when knock sensors fail the ECM will substitute the failed sensor with a predetermined value or go into "limp in mode". In the case of a knock sensor the ECM would bring the timing back to a safe point to prevent pinging. Usually 964 ping is caused by a broken distributor belt. You checked the belt. Is it possible that it jumped? Remove both distributor caps, rotate the engine while watching both rotors. Make sure the rotors line up with the index marks on the distributor housing.
If that checks ok, disconnect the ignition final stages one at a time and drive the car and see if the ping is gone.

[scotsman60]

Steve,


After some research I'm coming round to the idea that mixing race gas is the preferred way of increasing octane levels - thanks for the warning/info.

I've discovered that I have a problem with my oil cooler fan - it's not coming on when the temps go up. I checked this out a few months ago and all was well but as of yesterday the fan wasn't running when I forced the temp up to around 248. I'll investigate/repair this weekend.

I also plan to modify the fan operation so that it kicks in at lower temps in a effort to control the motor temps during stop and go driving. I think that the stock temp settings are too high for CA gas. If this doesn't work then adding a secondary oil cooler as you suggested would be the next step.

Now that the pinging is gone I can hear another noise. It's similar to pinging and occurs under the same circumstances but it's very definitely different. I suspect arcing of the ignition wires. If I disconnect one of the distributors (Take the high tension lead off of the coil and disconnect the coil supply) then the noise goes away. This happens with either coil. As soon as I have both coils/distributors active the noise comes back. I'm having trouble locating the problem because it only occurs under load - my wife is unwilling to accelerate hard with me perched on the back of the car........ so I may hook up a video camera to my laptop and see if I can see anything arcing in the engine bay that way.



More later,




Doug




When I get the oil cooler fan operating again

[Peter C2]

Doug,

Misting the wires with water will help in seeing the arcs.

[Steve@Rennsport]

Hi Doug:

Peter's suggestion is very sound and this is what we do here. Wet wires produce a corona and arcing in a dark garage or at night with the engine running (of course).

Keep your oil temps in the 190 degree range really helps prevent detonation, but you still might need a mix of race gas in hot weather.

[ischmitz]

Let's get some things clear: As far as I know 1990 964 cars did not have the check engine light hooked up at all. So you won't get a light. If you connect a hammer to the diagnostics port you can monitor pinging. Now if you don't get a warning on the hammer and can hear pinging at the same time you can be sure your knock sensors are faulty.

The knock sensors are essentially very unsensitive microphones. They are piezo cristals that generate a voltage under deformation. If they are broken or not connected there is simly no voltage. Hence the DME would never notice a broken knock sensor that never sends a signal even if you have heavy detonation.

If you can hear the pinging you can be sure you are on your way of permanently damaging your engine: broken rings, hammered rod bearings just to name a few. And it ain't cheap to fix those puppies, ask me how I know.....

My recomendation is to have a qualified Porsche shop to verify that you knock-sensors are working and that your DME is able to retard timing.

One comon reason for pinging is that older 964 engine have carbon build-up on the head domes and pistons. Those act like a catalyst and lead to premature detonation under high compression. High ambient temperatures and poor gas quality compound this effect.

Ingo

[scotsman60]

Steve, Ingo,


I was doing a manual inspection of ignition system today in preparation for an investigation tonight in the dark and using a water mister.

To my dismay I found a broken head stud on the intake side of cylinder 3.

Do you think it's likely that the detonation contributed to the head stud failure?

Do you think it's likely that the head stud failure contributed to the detonation problem?

Anyways, I'll get a chance to check the rings and valves for detonation induced damage in the verfy near future......


Doug

[ischmitz]

Doug,

I would consider it highly unlikely that detonation caused a broken head stud. It's the constant heat cylcing that does that. And a broken head stud won't promote pinging either from what I know. The original head studs on a 964 are dilavar. When they corrode they develop stress risers. During my rebuild I upgraded to 993 TT studs that are also dilavar but said to be stronger. Some recommend regular 993 steel studs. There has been a lot of debates about what is better. Some don't like the fact that the 993TT studs are all threaded but I figure that what is good enough for a 993 twin turbo should work fine for a NA 964.

If you want to spend more money there is always ARP or Raceware as an alternative. For what is worth,
Ingo

[scotsman60]

Ingo,


Thanx for the reply.

My thinking was that maybe the detonation caused higher cylinder pressures resulting in loads on the studs greater than they were designed for - a long shot I know.....

Vertex has the Raceware studs for only $5 a pop more than the 993 Dilivar ones.....



Doug

[ChrisBennet]

Like Steve said, boosters only raise the octane a couple of points. In other words it will raise a tank of fuel's octane from 91 to 91.2.
-Chris

[Jeff Burger]

Maybe I missed it but have you checked the drive belt for the slave distributor?

[scotsman60]

Jeff,



The slave distributor is working fine. The motor starts and runs on either single ignition system.




Doug