Timing at Steady State Low RPM in a DI Turbo Engine

We have all heard the horror stories, and that LSPI haunts these Turbocharged DI engines (In my case, a 2L Ecoboost). In these, most of the issues I hear are from passing on the highway, 6th gear, low RPM resulting in LSPI.

I was skeptical of the steady state tuning process before Andre eased my mind about it in the courses, showing that even heat wasnt a concern at lower load/RPMs at WOT steady state. However, now I would like to touch up lower RPM timing and am unsure if there is some sort of limit where I should stop and follow trends back down the RPMs that differ on a DI car verses a port injected car.

When doing WOT pulls I am following the advice given by the previous tuner I was working with prior to starting down this adventure of doing it myself, which was start your WOT pull at 2,500 RPM, no lower to avoid risk of LSPI.

I have not attempted it, but the turbo in this car is so small I wouldn't be surprised if I could get into low boost pressure at 2,000 RPM for example, I am assuming this should be avoided - but curious to hear some thoughts on low RPM steady state tuning on these DI engines.

Thanks

I thought I posted a detailed reply to this... maybe it never posted. Long story short, keep the boost down under 2500 rpm. Try to do it in the tune, not just relying on you keeping your foot off the gas.

Ah - good point, if I were to swap to a 3D boost target table (verses the OE strategy I am using now, of basically a ceiling it will stay below, not actual targets) I could limit/eliminate boost in low RPM. The ECU will close the throttle however to control the boost, but I suppose I could still get all the way to atmospheric pressure using that technique.

As usual thanks for the input Raymond. I have my base timing table where I want it and I started tuning the E30 map last night and results so far are fantastic. I think you are familiar with the COBB Ecoboost tuning, but I am just using a 3D compensation timing table on top of the base timing/knock response timing adders to add more timing for the E30, gained 30ft lb and 25HP on the first test run (using virtual dyno and a flat section of road), no knock! Added two more degrees across the high load sections last night and we will see how that goes!

I'd be interested in the detailed reply though ;)

Maybe I can get more detailed tomorrow. 16-17 bar BMEP is roughly the safe limit for LSPI based on my experience. That's about 200 lb ft at the crank on a 2.0 Engine up to 2500rpm, as rule of thumb. Be careful with ethanol spark adder too. It increases cylinder pressure despite no knock, especially at high rpm..

Continuing my thoughts on this - if you do a quick Google search of "Ecoboost 2.0 torque curve" you can see a grainy photo of a display from some auto show or conference that shows the steady state torque curve of one iteration of the Ecoboost 2.0 . Steady state WOT torque curves can only be generated on an engine dyno because you need a lab to control the air and coolant temperatures.

I created a red triangle in the upper left corner to show the LSPI risk zone. The "upper left corner" is the LSPI risk zone on any torque curve of a boosted DI engine. Some are less prone to LSPI than others; for example the GM 2.0 LTG engine is pretty good for LSPI, while the early Hyundai 2.0T DI engine is terrible. There are a lot of contributing factors to LSPI, which is why it's so hard to eliminate and why so much money has been spent by the industry to understand it. Those factors include the oil chemistry, design of the piston bowl, injector, and combustion chamber as well as things like valve timing (cam duration and VVT tune) and injection timing.

But the easiest way to be safe is to limit engine load under 2500rpm. Easiest way to do that is to limit boost by wastegate control and by not increasing overlap there as it tends to increase boost as well. Using your foot to control load is fine during tuning but ultimately risky. You should be able to give the car to your mom to drive and not worry that she's going to cause LSPI to blow the engine up. Minimize risks involving human error. Don't assume whoever is driving the car (yourself included) isn't going to do something dumb.

My other point about cylinder pressure is this: even if your engine isn't knocking, when you advance the spark you get more efficient (better brake specific fuel consumption) but you increase cylinder pressure. Ecoboost engines already have cylinder pressure limits baked into the stock tune. For example, the original 3.5 Ecoboost was limited to 80 bar steady state. See attachment.

Thanks Raymond, I will post up a graph of my timing curve and vdyno - so far I am seeing rather large jumps in TQ every time I add timing, I am peaking at 25 degrees and stopped there as the MBT timing table (baked into the OEM tables) indicates that this is MBT for the 6k RPM and above range at the loads I am seeing.

I certainly am concerned about cyl pressure as if you reference that OEM table I am blowing way past it, however in the notes it does indicate that OEM table was based on engine dyno results on 91oct fuel. I increased it 2 degrees up top on my actual 91oct calibration because I was seeing it limit my spark and I was still able to add timing without knock, but eventually ran into the knock threshold there (17degrees up top).

Comparing my current timing curve to the MBT table listed in the OEM calibration (which also indicates it was produced using an engine dyno and high octane fuel, didnt specify just how high of oct however) I am right on it at 6KRPM and above, from 4K-5K I am a few degrees under it, and down low I am 8-12 degrees below it.

Currently I am leaving high RPMs alone and seeing how the engine TQ reacts to timing added lower (3K-4.5K). I hope I am not wrong in feeling safe with the pressures I am creating as long as I am staying under that OE defined MBT curve... Latest results are attached, the two runs shown are a result of adding 2 degrees from about 3k-5k (I will attach my timing table when I get back home).

268HP @ 4734RPM

368ft-lbs @ 3398RPM

I think I am approaching peak TQ I am going to see, just based on looking at dyno charts from other tuners on the OEM turbo right around 400ft-lbs seems to be a limit - I have seen 300HP produced out of the stock turbo, but probably not going to happen at my elevation.

EDIT: Also, very much appreciate the stress on ensuring the car will be safe no matter who would drive it, going to concentrate on putting in safe guards that dont rely on your foot... I do already have some inverted timing adders at 2k RPM and below - need to get in there and look at boost limits down low as well to ensure load stays low no matter what down there.

Please post WOT pulls showing stock timing, your 92 octane timing, and your E85 timing. The MBT spark shifts with E85 as the combustion speed can change. I'm not sure how accurate that map is.

Also, is that ambient condition accurate-about 92kpa ambient pressure? That's going to push your turbo harder. A free flowing intake helps offset that. Usually stock tunes have about a 10 percent margin in them to prevent overspeeding the turbo at altitude. When you open up the intake you increase that margin, but when you crank up the boost at high rpm (low and mid rpm isn't a big deal) you push it harder.

When you lower the ambient pressure you stress it more. Remember the y axis of a compressor map is pressure ratio of outlet over inlet pressure. If outlet pressure is 200kpa absolute and inlet pressure is 100 kpa (near sea level), PR is 2.0, but if inlet pressure is 92, your PR is almost 2.2 . So you creep towards the edge of the map.

Will do, I need to get back to my other computer to get those but to answer you other question yes, that is the ambient pressure here yesterday when that was taken. I am at 2,800 ft.

I have the stock intake just with a more open filter - well and a larger, better flowing FMIC.

Also for reference I am on an E30 mix or 91oct, no 92 here :( Good point on the PR - I am certainly riding the choke line up top based on the airflow I am seeing in my datalogs. I may be wrong but my main concern is how far I can safely take peak TQ up, and if I could risk damage to the rotating assembly before I would see knock. I was going to stop adding timing when I stop seeing large jumps in TQ, but that doesnt really tell me if I am pushing the bottom end to hard.

Because the hardest thing I use this car for is autox I am more concerned about a nice wide TQ through the midrange than I am with peak HP, so leaving the top end timing where it stands. I have logs from my purchased tunes and the timing up top in my tune matches there, but I am quite a bit higher through the mid-range, I will post that log up as well.

Raymond,

I thought I had some logs of stock/stage 0 but not finding them, I will have to either get some or find them. I am attaching an excel book, with sheets that are logs of 3rd gear pulls of:

My Tune, 91 oct and E30

Purchased Tune (Stratified Automotive E Tune), 91oct and E30

Also including the base timing map that is active during these runs, the ignition offset table is only active on the E30 slot, the other table is the borderline PD table, that is active during both of my tune's pulls as the base table. You probably already know - but the offset table is multiplied by the octane adjust ratio, which I have default to -1, not seeing anywhere near enough knock to move it off -1 during these pulls. I attempted to make the timing offset table nose over into removing timing at RPMs and loads that I am not achieving as a safety precaution. Also didn't add any timing in cruise load areas... I would think thats normal?

I am stopping at this timing for E30 until I get more comfortable with going further - there is for sure some more room for advance in the mid range, I did run another version with a bit more timing and did see some small TQ gains, so backed off and staying here for now. :)

If you would like/would be easier we could skype or something sometime.

EDIT: Just realized as I was typing it out that my safety net on the offset table is a horrible idea, as if the octane adjust ratio ever went positive, from bad gas, then it would act the exact opposite of how I would want it to and add timing at those 'over the edge' conditions - doh! I will zero them out... added that attachment as well (E30 Offset New)

Go ahead and plot up RPM, timing, actual AFR, and boost and post some screenshots. I opened the files and they need a lot of processing to line up due to differences in time base and channel names. Make overlapping plots.

Hi Raymond,

Sure thing! Please see attached. For the combined charts I had to pick an RPM to reference but they are close enough to get the idea across - I also separated them out so that RPM information was directly referenced to the correct data. Tried to line them up as much as possible on the combined charts.

As you can see I am running less overall timing advance on 91oct than the Protuner Etune I have for my car, but our gas is pretty crap around here so I play on the safe side for 91Oct, could probably go a bit more... On E30 however low and midrange I am finding I can add much more timing and am seeing power gains - this is where I am wondering if I am missing something and this is safe but maybe the etuner chose to ere on the side of caution - or if they have a good reason to keep that timing profile.

Looking forward to your thoughts!!

Thanks

I don't have a plot of you running stage 0, but you've got a lot of extra timing in there vs what the stock tune does. This is ultimately your judgment call. It could run for years like that with all that extra timing, or one day you drive down the road and something lets go, and you wonder what the hell happened. Of course you could say that about any mod I suppose. IF you had cylinder pressure measurement (cost would be at least $60k for a Kistler box and 1 spark plug pressure transducer and the software needed, it's only for organizations with deep pockets) you could get a measurement to see how hard you are pushing it.

I'm going to attach two files, both based on steady state testing in an engine dyno environment. One is a 3D plot showing max cylinder pressure for a 2 liter turbo direct injected engine (not a Ford). The other shows the difference in cylinder pressure and spark at WOT on a port injected nonturbo V6, running 87 octane and E85. Look at those plots, think about them, and decide how much of that timing advance on E30 you want to keep. It's your engine and your wallet.

Thanks Raymond, very neat data. Is there a difference in how the pressure is applied via adding timing or adding fuel/air? What I mean is, there are quite a lot of folks running 350HP on these motors with stock internals, the protuner I was working with has a 400HP shop car that they beat on and it also has stock internals. Is it at all rational to assume that the pressures in those cylinders are stressing those motors harder than I am by adding octane and timing alone? Assuming no high levels of detonation in either scenario of course.

I did some more testing today running back to back passes just over the 2500-5000 RPM range as thats where there still seems to be power to be tapped. I copied the table I was using and set up two other slots one with +2 and one with -2 across that range and ran them.

The +2 started suffering from knock, nothing terrible but not something I would want to leave in that state as a finished map, good, proves I am close to the knock threshold of the fuel.

The -2 map showed large power losses up until about 4500RPM but then the power differences were marginal. I ended up just removing the timing above 4500RPM and will give that a shot tomorrow. I also just smoothed the map out a bit keeping it on the conservative side, overall lowering the map by a degree or so in some places.

I will get a snapshot of this and line it up with a stage 0 run as well as against the OTS protune's E30 timing and see where we are at.

It looks to me that above 4500 or so their timing map was on point, but there was certainly power left on the table below 4500 - checking their current offerings and they actually have a 'high TQ' E30 option, which guess what :) adds power in that lower RPM range. I will see if I can get my hands on a log from someone running that new version to see how far off my numbers are in comparison.

Before this vehicle I had a 400WHP STi, it was fun, but useless around town due to the lag. If I ever upgrade the turbo on this Focus it will still be small, 320-350WHP to keep that response. Anyways, that car blew a headgasket on me, and later spun a bearing (Disclaimer - I didnt tune it!! lol) and I rebuilt it myself - selling it after abut 10k miles was put on that motor again at 400WHP with no issues. The point of that story is yes, reliability is still high on my list - I enjoyed the experience of learning to rebuild an engine...but I am not wanting to do it again quite yet! :)

Thanks again for your time

An STI or Evo makes power with slow burning heads and a lot of spark retard and boost. Notice how rich they are (especially Evos) due to the hot retarded combustion. It's very fuel inefficient but doesn't cause as high of cylinder pressures.

More boost with less timing is harder on the turbo but generally has lower peak cylinder pressures. If you do a bigger turbo you reduce backpressure and pumping work which stresses the engine less. If you do a built block/bottom end you can handle the cylinder pressures better.