Spark Plug Damage and Your Engine

[FRP]

Hi all, sorry for being so long between posts. I have number of things to post on over the holidays and will take the time to do so.

I wanted to take the time to comment on the recent posts / pics of spark plug damage. There have been numerous pics of severely damaged plugs coming out of turbo motors with various suppositions attached regarding the possible causes of the damage.

First; Whenever there is physical damage to a spark plug, in the form of shattered porcelain insulators, melted or missing side electrodes, melted center electrodes etc. - it is not the fault of the spark plug - it is the result of archaic and grievously destructive conditions inside the cylinder.
(The one possible exception is if some foreign object has been ingested through the inlet tract)
Spark plugs, and their condition, are the "canary in the coal mine" of a cylinder's internal operating condition.

Before we had fancy dynos with wide band O2 sensors and engines equipped with exhaust gas temperature sensors - tuners and engine builders routinely assessed jetting and spark timing settings by "reading" the spark plugs. This is because the conditions (colorations, carbon deposits, microscopic metal particles on the center electrode) of the spark pug tip, are such a good indicator of exactly what is happening inside the motor during operation.

When you remove a plug from your engine and find gross physical damage to the plug - you have severe tuning issues involving fuel mixture and or timing of the spark. In either case there has most likely been severe detonation (pre ignition) going on and very likely damage to the piston, rings, cylinder walls etc. even if the damage is not readily apparent.

The causes are generally a combination of excessive spark advance and an overly lean fuel mixture. It is true however that either excessive timing or lean mixtures, by themselves, can cause the same type of damage to the plug. It is very important to note that both of these are measured against the quality of the fuel being used at the time of the damage. A "tune up" that would be quite acceptable for race gas, with no damage to plugs or engine, could cause a complete melt down on pump gas, as could a 93 octane tune if say 89 octane was used and the car was run hard. (Fuel quality and consistency in this country is generally terrible and highly variable from state to state and pump to pump).

When you see a plug with electrodes that look like the tip of a used welding rod - you have BIG tune up problems with your ECU, or, you have severe fuel delivery issues that are causing the engine to go lean under boost when mixtures need to be at their richest! It is never the fault of the spark plug.

I run just basic triple electrode Bosch copper core plugs in the Bonneville Land Speed engine and they come out looking just like new with a light tan coloration on the porcelain insulator - nothing special just plain old 2 dollars plugs - in a 1000 HP 10,000 rpm motor. BUT this is on high quality racing fuel with more than adequate fuel delivery capacity and a conservative (for race fuel) fuel air ratio of 12.5 to 1 under load.

The last point I would like to make is that every driver won't always hear obvious overt knocking or pinging in you engine under every condition that can cause spark plug / engine damage - sometimes it can be quite insidious and that is why you need good monitoring instruments on your car. At a minimum you should have a quality wide band sensor and gage to insure you are staying in a safe A/F range relative to the quality and octane rating of the fuel in use. With any pump gas you don't want to run leaner than about 11.5 to 1 at full boost. (And even this level assumes a proper pump gas spark table in the ECU) With the wide band running you will know if something in the fuel delivery system changes that adversely affects the tune up - before you pull a melted plug from the motor!

[chaloux]

I'd love to see examples of different grades of plug damage. I am one of those guys that had perfect a/f and destroyed an engine because of detonation. I also never heard the knock, but that doesn't mean it wasn't there. Every time I pulled plugs, they looked "fine". I haven't seen thousands of sets of plugs so my assessment is amateur at best, and no one around me seemed concerned over my plug condition. I'll have a look at my plugs with a magnifying glass to see if I can see any pitting.

[loxxrider]

My 2 cents...

Most of the time you will not hear detonation under load unless it is very bad. The easiest way to hear det is at low load, high throttle position conditions (say, revving a low rotating mass, high displacement engine on ITBs from idle. My M5 does it all the time). That is harmless det though. The bad kind of det is that which occurs at high load, especially torque peak. Like I said, you aren't likely to hear that on most cars unless it is REALLY bad in which case the damage has already been done.

On reading plugs, I find that the best tuners out there still tune timing by reading plugs. I am lucky to have a few good friends who are very good tuners, so I've picked up some great knowledge over time. I recently had a conversation with my friend (the one who owns Fuel Injector Development) and we discussed street tuning timing with plugs.

It isn't just pitting you need to look for, it's little black specs on the ground straps, electrode, and especially the porcelain. Guess what those black specs are... carbon deposits from your pistons, combustion chamber, etc. which have been knocked loose by the shockwaves of the detonation! That is your very first sign of det, possibly even before you can hear it with knock ears. Touching on this det limit is harmless for tuning. In practice, you should increase timing across the board, do a pull, check for signs of det, and if you see the black specs, then you pull a few degrees out.

If you are seeing shiny specs, then you are WAY off and need to pull timing way back. You also need a lot of plugs to do this. Fresh set for every pull or two. You can also read the ground strap to help you get in the right area (discoloration should be right around the bend... too much timing and it'll be further down the bend toward the porcelain, too little and it'll be out by the electrode).

Hope that helps. This is a great discussion

[FRP]

Dear Chaloux

Wondering what your A/F numbers were and what octane gas was in the car when your damage occurred? Also, do you happen to know what the ignition timing at peak torque was?

[chaloux]

I may have a log of my tune at that time. It'll take a little but I could provide all that info. I would say around 15deg at peak tq. BUT my timing was likely a tooth out, i.e.. +6deg. It was either that or the wiring with vems was totally messed. Lambda at peak tq would have been around .85, whatever vems called for at that boost. Probably between .85 and .8. It happened with a blend of using 93 and 94 octane over time, engine blew on the way down to Carlisle. I have photos of the pistons in my build thread, the top of #3 was melted to the ring.

[FRP]

Chaloux

If your actual timing in the cylinders was really 21 degrees advance at peak torque it may have been higher at other spots in the tables. The likelihood is that your melt down occurred at part throttle mild to moderate boost at cruising speed. You would have to rely on some of the guys who do the tuning and set up the tables to cooborate this as this is not my area of expertise. I can tell you that my B-Ville race motor only runs 22 degrees timing at peak torque and it is set up for 118 octane leaded race fuel.

[85oceanic]

Just had time to sit down and read through all this and I gotta say it's some fantastic stuff! Thank you for the fantastic discussion topic FRP!

[chaloux]

I'm still in for pictures the descriptions here from both Chris and Jeff make sense but seeing is believing (understanding, really). I'll have to use the old Google googles.

[chaloux]

http://www.contactmagazine.com/Issue54/ ... asics.html

That's a nice thorough read for det and pre ignition. He really goes into different scenarios and what to look for in each case.