Adding boost to high compression engines

z31maniac said:

BrokenYugo said:

Along with all the technical stuff already mentioned, there's a high demand for more power in the FRS/BRZ. So kits are going to be put on the market along with whatever extra retarded/rich tune it may need to live, regardless of how good a candidate for boost the engine really is. 

Except the tunes aren't like that. 

I've said it here before- if you are getting power adders or recalibrations, you better know what you are getting.  Engines are calibrated far closer to the edge, so that we can get the most out of them.  Making power can be pretty straightforward, but if the tuner doesn't know what they are doing- it can easily be a big bang.

alfadriver said:

Knurled said:

alfadriver said:While LBT is normally best at 12.5:1 under ideal conditions, when knock is bad, you can actually add a lot more fuel, and make more power.  Which is to say- the loss of power due to being too rich is more than offset by the knock mitigation.

Tell that to the friggin' Audi guys who seem to home in on 12.5:1 as perfect and end up running timing so retarded that they get themselves into a "coffin corner" of high EGTs melting valves vs. detonation killing rods.  One build, the guy said he had a 3 degree window of tune between the two extremes...

It's hard to deal with people who hone in on one particular thing as the most of the most of the most important thing.  Harder when you have to convince them that it's far from the largest knob to tune with.  Even harder when that knob has some really amazing interactive properties opposite of the "conentional wisdom".  

I have recently learned that you can gain more power by boosting more with retarded spark.   Even with REALLY horrible combustion timing, the gain due to boost ends up being more than the loss in spark.  I honestly never dreamed that.  But it is true on many engines.  Still, when you see the pressure trace when that is going on, it's pretty shockingly bad looking.

That is why you see those monster motors on the dyno with the exhaust lit up like a lava lamp!

In reply to bentwrench :

Some.

And that's kind of nice on a very cold winter day...  

z31maniac said:

Knurled said:

alfadriver said:While LBT is normally best at 12.5:1 under ideal conditions, when knock is bad, you can actually add a lot more fuel, and make more power.  Which is to say- the loss of power due to being too rich is more than offset by the knock mitigation.

Tell that to the friggin' Audi guys who seem to home in on 12.5:1 as perfect and end up running timing so retarded that they get themselves into a "coffin corner" of high EGTs melting valves vs. detonation killing rods.  One build, the guy said he had a 3 degree window of tune between the two extremes...

Tell that to same FR-S/BRZ guys (like the OP) that are also 12.5:1

Air/fuel ratio, not compression.   When your tuning has to be so knife-edge that 5bdtc burns valves and 8btdc detonates, maybe you should try possibly adding some MORE FUEL to get the EGTs down.

Of course, any time I mentioned this, I got "but leaner is more power!!!"  Well that's great if you like taking your engine home in a bucket, just to prove a point, I guess.

z31maniac said:

Knurled said:

alfadriver said:While LBT is normally best at 12.5:1 under ideal conditions, when knock is bad, you can actually add a lot more fuel, and make more power.  Which is to say- the loss of power due to being too rich is more than offset by the knock mitigation.

Tell that to the friggin' Audi guys who seem to home in on 12.5:1 as perfect and end up running timing so retarded that they get themselves into a "coffin corner" of high EGTs melting valves vs. detonation killing rods.  One build, the guy said he had a 3 degree window of tune between the two extremes...

Tell that to same FR-S/BRZ guys (like the OP) that are also 12.5:1

The 12.5:1 I was referencing is the compression ratio.  I have no idea what my A/F ratio is and have no plans to fool with it.  I was just curious as to how people are getting away with boosting what seems to me to be a pretty highly tuned engine already.  Thanks to Vigo and others for the explanations.

alfadriver said:

z31maniac said:

BrokenYugo said:

Along with all the technical stuff already mentioned, there's a high demand for more power in the FRS/BRZ. So kits are going to be put on the market along with whatever extra retarded/rich tune it may need to live, regardless of how good a candidate for boost the engine really is. 

Except the tunes aren't like that. 

I've said it here before- if you are getting power adders or recalibrations, you better know what you are getting.  Engines are calibrated far closer to the edge, so that we can get the most out of them.  Making power can be pretty straightforward, but if the tuner doesn't know what they are doing- it can easily be a big bang.

Yes, you have to know who to use and who not to use. 

But basic research will easily show, especially on this particular platform, who to use and who not to use. There are only 4 tuners I would consider using, and only 2 of those I would use for a Forced Induction application.

In reply to z31maniac : would you mind sharing your opinions with us?  

Shiv - Open Flash Table gets a lot of hate, but that's only by the fanbois who insist it's terrible, it's OTS tunes (they all are unless you do a dyno or remote tuning session, regardless of tuner) and it's a great tool if you want to learn yourself and it also datalogs, no secondary equipment needed.

Moto-East

For FI:

Delicious Tuning or HRI Tuning

Interesting reading about how the Audi and FRS tuners are so locked onto 12.5:1.  10psi and a good tuner in an S2000 is getting you up in the 400whp vicinity, but usually 13:1 tapering down towards 11.5:1 AFR or lower by redline. Again, all depending on the tuner.  

Any input on tuners using Lambda vs AFR for tuning?

The most comprehensive packages also require modifying the short block to lower compression on the twins, I believe.

So what exactly are we worried about igniting early..... the air/ air mixture? Frisbees are direct injection are they not? Wouldn't that mean that there's no fuel to pre-ignite during the (high) compression stroke? Maybe I'm completely wrong, but isn't that the whole reason for direct injection? That there's no fuel in the combustion chamber to ignite early. 

They have both direct and port injection and i don't know when each system operates. Either way, i am fairly sure that even direct injection gas engines have all their fuel present before the spark occurs, so it's still possible to detonate. 

sesto elemento said:

So what exactly are we worried about igniting early..... the air/ air mixture? Frisbees are direct injection are they not? Wouldn't that mean that there's no fuel to pre-ignite during the (high) compression stroke? Maybe I'm completely wrong, but isn't that the whole reason for direct injection? That there's no fuel in the combustion chamber to ignite early. 

Other than the cold start, all DI systems inject their fuel during the intake stroke.  That's where you gain a little power by cooling and shrinking the charge on it's way in.  So all of the fuel is there for a long time.

Acutally, DI has a unique problem- at low speeds and very high load, you can get some really nasty pre-ignition, which then results in some very destructive knock (some call it Mega Knock).  Very fast holes in pistons bad.

DI engines are still less senstive to knock overall, thanks to the charge cooling during the intake stroke.

alfadriver said:

Acutally, DI has a unique problem- at low speeds and very high load, you can get some really nasty pre-ignition, which then results in some very destructive knock (some call it Mega Knock).  Very fast holes in pistons bad.

I found that whatever engine controller runs a BMW 135i has a PID called "Super Detonation" and can be viewed per individual cylinder.  

I also found that when the Valvetronic motor goes tango uniform and takes out the drivers in the PCM, it's roughly $9000 to repair, which ended my desire to own a 135i/335i.

Further, I found that accidentally getting 3rd gear instead of 1st in a Mazdaspeed3 while parking it will make the engine detonate like someone was firing a .40 caliber under the hood.   This mostly ended my desire to own a Mazdaspeed6.  (Then I found not one but two within a few hours' drive with cloth seats and non-sunroof, and I wanted one again)

Knurled said:

alfadriver said:

Acutally, DI has a unique problem- at low speeds and very high load, you can get some really nasty pre-ignition, which then results in some very destructive knock (some call it Mega Knock).  Very fast holes in pistons bad.

 

I found that whatever engine controller runs a BMW 135i has a PID called "Super Detonation" and can be viewed per individual cylinder.  

I also found that when the Valvetronic motor goes tango uniform and takes out the drivers in the PCM, it's roughly $9000 to repair, which ended my desire to own a 135i/335i.

 

Further, I found that accidentally getting 3rd gear instead of 1st in a Mazdaspeed3 while parking it will make the engine detonate like someone was firing a .40 caliber under the hood.   This mostly ended my desire to own a Mazdaspeed6.  (Then I found not one but two within a few hours' drive with cloth seats and non-sunroof, and I wanted one again)

Done that in my MS3. Thankfully it didn't go boom. 

I do get some weird pebbles in a tin can noise at high-ish boost in 6th gear going for a pass. It's a little temperamental. 

Yeah, the DI has significant cooling effect and allows you to run more boost that yu would otherwise.  I see 25 psi+ on my Ecotec.

DI also has it's own issues - carbon deposits on intake valves, and with a turbo engine you can't run any sort of cleaner through the throttle body, or it can kill the turbo bearings.

Even non DI engines can have significant boost with modern engine comtrols - an S54 BMW has 11.5 comprssion ratio and can still be turboed 

https://youtu.be/silQi_8x3gw

https://www.bimmerforums.com/forum/showthread.php?2123962-E30-S54-TURBO-1060-WHP-1130Nm-(833lbs)-1167WHP-E21

Interesting about the preignition, I must've been misinformed. It would seem to me that you could benefit greatly by not having fuel present until the last possible moment to prevent preignition but maybe the pressures needed to overcome the pressure differential would be difficult to achieve. 

I read that the carbon issues are why toyota designed the d/i on frisbees to also have traditional style injection, to wash the carbon deposits away. 

In reply to sesto elemento :

From a cooling standpoint, there's quite a bit of that when you cool the charge during the intake stroke.  And you get a lot of time to evaporate the fuel.  In spite of injecting fuel at 700-3500 psi, it's still needs some time to evaporate.

And one other problem with super late injection- one needs to not hit surfaces much to avoid particulates.   Other than the cold start, where you can run incredibly late ignition timing because of the late injection, compression injection hits the piston enough to cause a particulate problem.

DI is really cool, but it, too, has it's compromises.

wspohn said:

DI also has it's own issues - carbon deposits on intake valves, and with a turbo engine you can't run any sort of cleaner through the throttle body, or it can kill the turbo bearings.

The throttle body is downstream of the turbo, running cleaner through the throttle body will not go anywhere near the turbo.

I wish I could have pulled the intake off first before doing it, but I did a decarbonizing on a CX-7 in preparation for replacing the injectors.  (My first DI injector pull.  I saw a nightmare video of trying to pull the injector from a 5000 mile Focus, where the front of the car was 2 feet off the ground suspended by the injector due to carbon locking) 

Not only did the injectors come out with simple finger pressure, but the intake ports were pretty dang clean.  Not as clean as a port injected engine that sees a lot of high fuel flow, to be sure, but not clogged with great gouts of carbon.

'Tis part of why I want a Mazdaspeed6.  I really want to do a before-and-after shot of the intake ports and valves, but that's just not something you do on customer cars.

sesto elemento said:

Interesting about the preignition, I must've been misinformed. It would seem to me that you could benefit greatly by not having fuel present until the last possible moment to prevent preignition but maybe the pressures needed to overcome the pressure differential would be difficult to achieve.  

What you are describing is a diesel engine, except you're still using a spark to ignite.  The flame front speeds of gasoline prevent this from being really feasible.  It works on diesels because of the slow burning fuel (also why diesels tend to have larger torque numbers as the flame front continues to exert force on the piston during a larger arc of the crank throw.  Gasoline engines are pretty much done pushing by 20-25 degrees ATDC, diesels more like 50-80 degrees).

Getting the injection timing correct (we're talking milliseconds) for gasoline to operate like a diesel is nearly impossible.  Even if you could get a base map established, the processing speed of the ECU couldn't keep up with adjusting it.  In the time it would take a knock sensor to tell the ECU, the ECU to process it, and the mechanicals to respond you would have already melted 2 pistons.  Diesels are pretty easy.  Throw in some fuel and it burns slowly.  Gasoline goes boom really fast. (also one of the reasons diesels are limited in their RPMs. Its hard to get all that fuel to burn in the shorter times you give it with higher RPMs)

Even with a diesel, the reason they sound like they do is because they are basically operating every combustion event as if it were a controlled ping.

In reply to curtis73 :

It might interest you to know that Mazda has a compression ignition gasoline engine developed and going into production cars for the 2018 or 2019MY.

The only problem with saying something can't be done is when someone else is already doing it

I thought most of the Diesel's efficiency advantage was a combination of always running lean, and the flame kernel being insulated from the water jackets by the excess, unburnable endgas air.  Less heat gets lots to the cooling system and certainly the EGTs are extremely low at low loads because it's running so incredibly lean.  I've never heard of them making more torque per displacement than spark ignition engines, though.

I think the big advantages with DI is there is no fuel injected until the exhaust is closed.

When fuel is in the intake charge some fuel can be pulled over during overlap and that fuel is wasted into the environment.

With the VVT most DI motors enjoy this is even more critical as they may be doing some strange things with valve timing.

I don't think there is any charge cooling on DI motors, especially since the fuel is injected later and does not evaporate in the runner.

The big emissions thing is to burn the fuel at the lowest temp possible to limit the formation of oxides.

With port fuel injection you can also only inject the fuel once the exhaust is closed, at low throttle openings when that sort of thing actually matters.  Once duty cycle gets over about 20-25% then you have no choice, but at that point you're outside normal idle/cruise paramaters anyway, on all but the most scrody underpowered engines.

When I was setting up our first Buick Grand National/MS3-Pro conversion, I went as far as to get another BGN with a stock computer and scope the injector timing versus cam sensor pulse timing, so I could duplicate what GM did timing-wise.  I did this, because I went around the full 720 degrees in 30 degree increments and didn't notice a significant difference at idle and low load. 

Running in batch fire mode before I got the cam sensor reading properly however, DID make a huge difference - it only ran on four cylinders!

That should be a clue and a half as to what really happens during the overlap period.  Until your intake manifold pressure gets near atmospheric and your in-cylinder pressures get high enough to get a good solid forceful slug of exhaust out the pipe to pull a vacuum at the end of the exhaust stroke, during the overlap period the low intake manifold pressure and relatively high exhaust pressure will conspire to push exhaust up the intake rather than pull intake out the exhaust.  And even when intake blowing out the exhaust does happen, if ever, it only tends to happen in a rather narrow window.

Building, tuning, and driving peripheral and bridge ported rotaries for the past ten years REALLY drives that home!  You only *think* piston engines have any kind of overlap period...  the engine in my car right now has something like 160-170 degrees of overlap period.  They run like absolute dogE36 M3 until intake manifold vacuum is nonexistent, and then they stop sucking exhaust gases back in during the overlap period, at which point VOOM.

bentwrench said:

I think the big advantages with DI is there is no fuel injected until the exhaust is closed.

When fuel is in the intake charge some fuel can be pulled over during overlap and that fuel is wasted into the environment.

With the VVT most DI motors enjoy this is even more critical as they may be doing some strange things with valve timing.

I don't think there is any charge cooling on DI motors, especially since the fuel is injected later and does not evaporate in the runner.

The big emissions thing is to burn the fuel at the lowest temp possible to limit the formation of oxides.

You can actually inject fuel closer to Exhaust Valve Closing (EVC) than you think in most cases.  Normally, during overlap, the intake manifold is at lower pressure than the exhaust manifold- so the flow isn't going out the exhaust, it's going backward up the intake.  There are times where that flow does go the other way, and for emissions, it's not the fuel that leaves, it's the excesss O2 that's the real problem.  The fuel can be dealt with quickly in the catalyst.

BUT, that cross flow issue for DI isn't the real issue with later than TDC injection timing- it's just that hitting the piston too much is a bad thing in many aspects- evaporation, particulates, mixing, etc-   It all falls apart if you inject too early.

There is cooling when the injection happens on the intake stroke.  It's more effective in the cylinder than in the runner- as it causes a minor drop in pressure, drawing more charge in.

And while you have a point about the NOx formation, it's a hard balance that buring it at a high as possible temp is better for themal efficiency.  And NOx can be dealt with in the aftertreatment.  Although, that high pressure and temp also puts the engine really close to it's knock limits, which is bad.

In reply to alfadriver :

How many injection events are there in a gasoline DI engine in a single stroke? It sounds like it's just a single injection that can be timed a little. In my head, I often equate gas DI with diesel, but a diesel engine can have several injection events in a single stroke. I'd imagine this is made easier since it doesn't have to time injection with a spark in addition to piston location.  With diesel being able to inject fuel at different times, and in different amounts, it allows a combustion process to be tailored for an application a little better in regards to power/emissions/etc.