Tuning timing for a headlift prone engine

Hi guys,

I have a turbo EK Honda Civic that I am working on, turbo'd and street tuned myself...I have recently discovered I have damaged my head gasket (pushing coolant) and will be replacing it. I am unsure if the cause though was just aged gasket or if it is overly aggressive timing as D16 engines are known for lifting heads with timing, due to small bore and high cylinder pressures.

I am running E85 so did not consider knock being an issue when adding timing. The thing is that, from what I've seen on other mate's D16 builds, a D16 will lift the head on a dyno before it even reaches MBT or starts knocking (if not using e85), so there is no way of knowing that it will happen until you've already added too much timing and done the damage.

The question is then, how do tuners know when head lift will occur if they are only watching out for knocking when adding timing? All of the mentioned builds, myself included, have used ARP head studs as well to no luck, it is a very common issue.

As for my build I will be repairing the head gasket and then removing quite a bit of timing across the board to eliminate it as a future cause of damage, but then adding some boost to hopefully compensate for any power loss. Equal power, less cylinder pressure - is this a logical move? It appears from experience that D16s prefer more boost and don't like much timing at all.

Thanks everyone.

Moses,

I hope at least some bits and pieces of the following are helpful. I don't know how much power you're running, but stock gasket with off the shelf ARP studs works well on that engine at 400 wheel HP on E85, yet I've seen loads of people lift the head and blow the gasket at half that.

Andre has touched on the subject, but it's very common to over advance timing on E85 without being aware of the damage caused until parts fail, and I would say there's a good chance timing is over advanced so you're likely on the right track.

Keep in mind adding air mass via boost will increase cylinder pressure not reduce it, but ignition timing can cause far greater pressure spikes if over advanced. Altering ignition timing also alters the shape of the pressure trace, the timing and speed of pressure rise, in addition to peak pressure.

Running more boost can cause similar issues if intercooler and turbo efficiency or exhaust pressure won't safely support it, so as always, make sure the package as a whole can accomplish the task reliably or increasing boost may introduce new issues causing similar failure.

Also, if you're on stock ignition system and a modded stock ECU, timing control on those is not very good and you have to leave a buffer. If you do a proper ECU and COP conversion, the level of control goes up dramatically, but that's not absolutely required, and tuning with an eye for safety is sufficient.

Also keep in mind how wheelspin impacts engine operation. I've often seen people tune the older Hondas and they're fine on the dyno then blow on the street because during wheelspin lambda shifts greatly as areas perhaps not fully tuned are encountered.

In terms of monitoring, I use crankcase pressure which will show the issue before knock occurs when I can.

Without any added sensors you can find where ignition timing doesn't gain as much per degree advanced, back off a little from there, and likely stay out of trouble.

Best of luck and keep us posted.

Hi Mike,

I really appreciate your response.

I think I understand what you are saying about how ignition timing alters the pressure spike. I also understand that increasing boost will technically increase the base cylinder pressure. However what I am struggling to understand then is, if more boost increases cylinder pressure, how a lot of D16 engines are happy with 400whp if it is via 30psi and not much timing vs 400hp with 20psi and heaps of timing and eventual (if not immediate) head lift - if the cylinder pressure is still the same (assuming cylinder pressure is what is used to achieve power), then the shape of the pressure trace or the amount of "spike" must be incredibly different to stop head lift altogether in the higher boost application.

So between the option of having 250whp with 12psi and too much timing causing headlift, I think I will aim to achieve 250 with 15psi+ and very little timing and try to increase the efficiency of my IC and cold side.

I will hopefully be running COP and sequential injection next month once the head gasket is repaired, and am using a full standalone ECU similar to a MegaSquirt so there is a lot of changes that can be made.

Good idea with the crankcase pressure sensor, but I would have no reference as to what is too much pressure as I think due to the cheapness of the engines no-one has bothered monitoring a D16s cylinder pressure while it's lifting its head.

Thanks.

Moses,

Coolant pressure is a good way to see head lift.

Crankcase pressure helps you spot over advanced timing, ideally before the head lifts.

And for both you just run the engine through the engine operating range on a known safe tune as your baseline. The increase when something is wrong is generally pretty pronounced.

There are many factors related to whether detonation or pre-ignition occurs and ignition timing is only one of them, albeit a very important one.

Beating up a small turbo to make power vs. running a larger one at lower boost is another common path that can lead to head lift from detonation or pre-ignition due to the much higher temperatures generated.

Having too hot a spark plug is a really simple one. Hopefully you're using NGK 8 range.

Over advanced timing can occur before MBT on E85 and will cause a pressure spike far beyond what you'll achieve without detonation.

A couple more psi boost and a bit less timing advance will likely net you a much safer tune as long as the items I mentioned in the prior post like turbo, intercooler, exhaust system are going to operate efficiently still after the boost increase.

The problem with "head lift" is the PEAK cylinder pressure, and this isn't directly related to power/torque, but they are related.

If you consider the pressure in the cylinder during the compression, ignition, and power stroke, you will realise that the pressure initially increases relatively slowly from the compression, then when ignition occurs the pressure will increase more rapidly approaching, and passing through TDC, to a peak shortly after TDC.

The classic illustration* of this -

If the ignition is delayed, the pressure increase from the combustion is delayed and the peak will occur AND it will also occur when the cylinder volume is also larger, so further reducing the pressure peak. This means you can potentially increase boost and get more net power/torque while reducing the pressure peak.

If you look at the graph* below (if it worked), you can see it rather well illustrated-

*Source - https://www.researchgate.net/figure/Theoretical-cylinder-pressure-curve-for-spark-ignition-SI-engines_fig1_304562783

You may wish to check it out, it should be helpful for you to grasp the idea.

Hi Gord,

Thanks a lot for the response, I do understand what you're saying, and both graphs are there. So ideally what I need to do is retard the timing so peak pressure occurs when there is more volume in the cylinder for it to fill, which reduces peak pressure overall.

Thanks a lot Mike and Gord for your responses, I have learnt a lot from this interaction. I may initially just remove a fair amount of timing (I have read some tuners have halved timing from their MBT/head lift point and not lost much power on these engines) and not touch boost, if it still seems to be making good power then I will leave the boost alone as it is still a stock bottom end with pencil rods.

Moses

Pretty much, you're losing a bit off the maximum pressure, which is what drives the piston, con'rod and then turns the crank. But because it's so close to TDC the force is almost directly down and applying little torque but, as the crank turns, the pressure applies a bit more torque because of the con' rod angle.

If you're still using the stock rods, minimising the peak pressure is definitely a good thing for safety, as it's the pressure on the piston producing the force that bends it.

You may find it helpful to also check up on the relationships between pressure, volume, and temperature and think about how the change in cylinder volume during the compression and combustion strokes interact with them.

Moses how much timing are you running at peak tq and peak RPM?

Hi Mike,

I am unable to say as I do not have access to a dyno currently. The car was booked to go on one last month but was cancelled on part of the shop, haven't rebooked it yet as I need to fix head gasket first. I know it's not a reliable indicator, but my "gut feeling" when driving, I'd say peak torque was in the 4500-5500rpm range, in which timing was ~24 degrees, and then tapered off to max 18 degrees at 7200rpm redline at about 12psi. I am unsure if this helps but from what I know now this is definitely far too much advance for my engine.

Thanks.

Moses,

Usually I'd say something like...I'd need far more info to say if that's too much or too little timing, because there are so many factors. In this case I can tell you 24 degrees at peak torque on that engine, at that boost level, is WAY too high, and that will kick the gasket out.

Also, as a general rule timing at peak tq will generally be lower than at redline. HPA explains why in their courses including ignition timing tuning, which would likely help you in your tuning efforts in general. Chip tuning a Honda doesn't exactly fall into the standalone ECU category, but the Practical Standalone ECU tuning course would likely be your best bet vs. the practical reflash tuning course, unless you have other projects that would be reflash based.

Hi Mike,

Sorry for the late response, spent today getting the head off and installing some sway bars...I see what you mean. I am using a full standalone ECU so I will check out the course.

Thanks for the answers,

Moses