Spark plug issues

[NoDOHC]

For future reference, this is my daily driver. I do want reliability. I ran 13,000 miles with very advanced timing and had no problems with it, so I thought that I had a reliable tune.

Anyway, so what I am taking from this is that If I hear something that sounds like metal hitting metal, it is severe detonation and that the pre-ignition is inaudible. That would explain some of the symptoms that I saw on my spark plugs.

Based on Ted's advice, I retarded the WOT timing to 28 BTDC at 3,000 RPM and tapering down to 15 BTDC at 750 RPM and found that the engine felt more responsive. I drove it 800 miles only stopping for gas and the plugs looked as good as new at the end of this.

I also noticed that the engine is much smoother at low speeds (below 3,000 RPM).

My previous WOT ignition curve was completely flat to 1500 rpm (34 degrees BTDC) and tapering to 25 BTDC at 750 RPM.

I am thinking that my original theory was incorrect and that the spark plug erosion that I observed during hard driving was due to pre-ignition, not incorrect plug heat range. The worst detonation that I felt was about 2,700 RPM before (the 90% VE breakover point). This may have also been a contributing factor in the engine failure on the dyno.

Thanks for the advice Ted! I think this information is very good for posterity.

Originally Posted by RETed

"Hear" preignition?
Not the real early preignition events...
On a dyno, you'll see it on the graph.

Please expound on what you see on the graph that indicates preignition...
I will see if I can scan the graph of my 5,400 RPM blowup in and see if you can identify the event.

Originally Posted by RETed

In my experience, EGT's will start to climb when you start to edge into early preignition.
Once you overly advance the ignition timing, then you start to get serious preignition which misfires and drops power drastically.

Ok, so that explains the drop in power when going to 43 degrees.
What I hear you saying is that the 8.2:1 engine really made 210 WHp at 33 degrees BTDC ignition timing, the extra 6 WHp that I picked up going to 38 degrees was due to early pre-ignition.

I had EGT ports installed on my Defined Autoworks headers, so I will get EGT numbers when I run with those headers on the dyno (probably at least a month out).

[j9fd3s]

Originally Posted by NoDOHC

For future reference, this is my daily driver. I do want reliability. I ran 13,000 miles with very advanced timing and had no problems with it, so I thought that I had a reliable tune.

.

it depends on the drive cycle, with our race car, for 3 years, our drive cycle was WOT for 25hours at a time! which is almost the opposite of the drag racers, their season is less than 15minutes of run time. the honda guys expect to rebuild the engine at least once in that 15 minute season too.

so there is a HUGE difference between an engine with a 25 hour design vs one with a 7 minute lifespan. if the engine runs in like 7, i minute increments, they can run cardboard apex seals...

a street car yet again is different, as it needs to do stop and go and things, and the WOT cycles are short, at least here in CA.

i've been looking at the timing curves of various engines mazda has built over the years and i must say its surprising. the 85 and earlier timing curves are all in the FSM's, so i was interested to see what timing they ran in like a non emissions engine.

the answer? its like 15 degrees@wot. the later 70's engines with reactors run about 20, the catalyst engines run more like 27... i suspect the non reactor/non cat engine still runs 15 degrees because of emissions still, but more isn't always better, as you just found out.

the FSM's are all at www.foxed.ca the REALLY weird ones is the late 70's rx4/cosmo, its runs a normal mechanical advance, so there is a 15 degree timing split, and L&T advance together, but it ONLY has a vacuum advance on the trailing. so the vacuum advance ONLY changes the timing split.

the other weird one is the 74? rx3, its got a timing retard box, so whenever this box is on, leading timing is retarded 20 degrees. it MUST be to get the reactor warmed up fast, but when this is on, the trailing fired before the leading....

anyways, start small, and work your way up...

[RETed]

Originally Posted by NoDOHC

Please expound on what you see on the graph that indicates preignition...

There's multiple stages of preignition...

We do know that preignition = "uncontrolled combustion due to uncontrolled ignition events", right?
There's an old racer adage about "lean is mean" - you can lean it to the point that it'll produce power, but...
At this point, BMEP starts to spike - this is why the power starts to gain a little.
This is a pretty ragged and very narrow range though.
The dyno might or might not show this, depending on the dyno (and dyno set-up) - it could manifest itself with a more "ragged" looking graph.
(Be careful about blaming preignition on your ragged dyno graph - there's a LOT of other factors that can do the same thing, for example: insufficient spark due to bad plugs, wires, coils, under voltage, bad ground, etc. or bad fuel delivery due to fluctuations in fuel rail pressure, dirty fuel injectors, bad ground, dying fuel pump, dirty fuel filter(s), fuel temp too hot, etc.)
EGT's will start to spike also, but only experienced tuners will know that it's too hot.
You will almost never hear this with the human ear.

Keep pushing it past this point, and power will start to fall off on the dyno.
This due to the uncontrolled combustion - the engine is fighting itself to harmoniously push the crank (well, in our case, it's the eccentric shaft) in the right direction to put the power (technically, torque) down to the wheels.
EGT's will significantly increase at this point.
You may or may not hear this.

Keep pushing it even further...
You'll start to induce ignition misfire cause either, 1) mixture is too lean to ignite properly, or 2) ignition timing is too advanced and way outside the power producing range.
EGT's at this point will drop due to the fact that it's a misfire - you're not getting proper ignition and combustion, i.e. BMEP plummets.
Obviously, something it very wrong at this point, cause the engine just does not sound good at all.
Depending on the dyno model, you can get all kinds of weird lines.
My recent experience with Dynapack shows spikes - I'll try and grab some pics, as we're going to retune a customer car next Sunday.

Side note...
As dangerous preignition and detonation are (to the rotary engine), the absolute more critical point in the graph is at PEAK TORQUE.
When you make your runs on the dyno, keep an eye out on when torque peaks at what RPM range - you will see a general idea of the torque graph / peak over multiple runs.
Peak torque = peak BMEP, and having the engine detonate at this point is almost instant engine failure.
Keep this in mind when adjusting your maps in those ranges.


-Ted

[scotty305]

Originally Posted by NoDOHC

...
Anyway, so what I am taking from this is that If I hear something that sounds like metal hitting metal, it is severe detonation and that the pre-ignition is inaudible. That would explain some of the symptoms that I saw on my spark plugs.

Based on Ted's advice, I retarded the WOT timing to 28 BTDC at 3,000 RPM and tapering down to 15 BTDC at 750 RPM and found that the engine felt more responsive. I drove it 800 miles only stopping for gas and the plugs looked as good as new at the end of this.
...
I am thinking that my original theory was incorrect and that the spark plug erosion that I observed during hard driving was due to pre-ignition, not incorrect plug heat range.
...

I'd be very interested to see photos of the eroded spark plugs, if you've still got them.