# GM 3.6L variants FYI

Discussion in 'General Auto Conversion Discussion' started by maticulus, Aug 11, 2019.

### Help Support HomeBuiltAirplanes Forum by donating:

1. Aug 11, 2019

### maticulus

#### Member

Joined:
Aug 11, 2019
Messages:
6
1
Good day. Since I joined this forum specifically to offer some clarification in a thread it turns out I can not post in after the fact, I'll kill to proverbial birds with one stone by combining my intro along with the info I intended to share, here.

I am an Army veteran/heli mechanic from long ago, that loves the subject of flight and aircraft, although I have not transitioned over to the point of becoming a licensed pilot (yet). I am also an avid motor head. I stumbled across this forum while doing research for a future performance build on a GM 3.6L V6, where I found the thread discussing the LFX motor.

In that thread I read some valid points and respectfully, as much that was far off the mark. I'm providing information that I believe those interested in the idea of this platform will hopefully appreciate, those who are not enthused might become such and those who think it's upside down might lighten up a little, or not.

The now 3.6L is an impressive motor over the iterations, DI=direct injected;
2.8L, 3.0L and
3.6L LY7 242-275 hp,
LLT DI 281-304 hp,
LFX DI 301-321 hp,
LF3 & LF4 twin turbo versions, 420 & 464 hp respectively and the latest, a clean sheet design 3.6L LGX 305-335 hp with active fuel management.

With the exception of the LGX, the motors share several interchangeable parts, except those associated with the smaller displacement motors and all listed above are VVT on intake and exhaust. The Turbocharged 2.8L was a Saab application that I understand carried single camshaft VVT.

The 3.6L has had its struggles with timing chain elongation issues early on, mostly associated with an extended oil change interval (owners mistaking the oil life left monitor, for oil level & running it too low), an alleged defective batch of chains and the effects of cheap motor oil. That problem is a non issue now.

Although some engines used oil (not just the 3.6L), I understand it was mostly the result of the PCV system design causing the motor to ingest it as opposed to burn it. Add an extended oil change interval to an oil consuming engine and put an individual behind the wheel that doesn't bother to check the oil level after 5000 miles, because the oil monitor says there's 50% life left and yes, there's a good chance it will run low and ruin the motor as several did resulting in a recall and recalibration of the oil life monitor.

My 08 3.6L LLT in a CTS does not use oil that I have detected between 5k mile intervals. I replaced the timing chains as preventive maintenance at ~176,000 miles and it has about 195k on it now.

I read debate in the thread over the VVT usefulness in flight and can only say that anything that optimizes camshaft angles for efficiency should be considered in terms of fuel savings. Locking the camshafts straight up is easy, but efficiency varies more with a fixed camshaft over rpm changes vs. functioning VVT, which changes with engine rpm as well as other conditions.

There is a software program used by those who modify their cars called HPTuners. This software makes it possible to tune as well as adjust many important engine performance parameters, including camshaft angles. An aftermarket controller could be used also.

I stopped reading about halfway through that thread, one thing I try to avoid on the net is arguing vigorously with strangers over a difference of opinions, in the absence of documented facts, I'm married and get plenty of that up close and personal already.

Here's the thing, a lot of what I read were rules that applied to old technology being repeated, with little to no consideration given to modern technological advances. One statement that really shocked me was the suggestion that direct injection had little advantage over port injection under a set of common circumstances.

That was eye "popping" for me. The LFX and LGX have a compression ratio of 11.5:1 and they are rated and SAE certified on 87 octane gas. The LY7 with 10.2:1 compression and port injection is also rated for 87 octane, however, under some circumstances it may encounter spark knock and may need to use 89 octane here in there. The owners manual points that out. There is no such caution for the DI motors in the owners manual. I wouldn't be surprised to find, given the reserve GM often includes in their drivetrain applications, that the DI motors can be bumped to 12:1 compression and still run 87 octane. Either way, that is an exceptional accomplishment by old standards.

There are DIY turbocharged 3.6L LFX motors generating in excess of 700 hp, with ~550 wheel hp set as the limit for the stock motor and those who are fortunate enough to own the Cadillacs with the twin turbo production 3.6L motors (ATS-V) are tuning and tweaking them into 10s quater mile cars, without gutting them and turning them into track cars.

I would like to build, or own a plane some day, and several years ago before getting deeper into my education path, I acquired 4 part catalogs from major suppliers and purchased the book, "Fundamentals of Flight" to show my level of commitment, so I guess my visit here at this time is not too premature.

Whatever the case, I believe power to displacement potential makes this motor worth being considered and as for the theoretical speculation to the contrary that I read in that thread, I had to endure similar when I defied "ancient" performance rules stating that you must lower compression in order to run boost, along with other anticipation that I would fail, from individuals who had never attempted, or seen anyone attempt what I was planning to do.

I had actually thought the process through for feasibility after deciding to do it. I built an 11.5:1 compression, port injected GM 60 degree 3900 V6 and boosted it to 7 psi with the help of water/meth injection under boost. The motor was so powerful from the high compression that in a short while it destroyed the 6 speed transmission. It still managed 33 mpg hwy after I damaged it from being impatient while tuning it, by burning a few valves and dropping cylinder pressure in the affected cylinders. I have no doubts it would have done 35 or better had I tested it before the damage. I rebuilt it to 11:1 and ran boost pressure as high as 22 psi on 93 octane with water/methanol injection. I will never build a low compression motor for boost.

I applaud anyone thinking outside the "Box" to try something different in an effort to achieve an outcome that makes it all worth it, with respect to all safety concerns. Theory is just that, theory, but the proof is in the pudding. If we all adhere to the existing rules of the DIY "playbook", there will be no progress beyond those limitations.

Looking forward to surfing the forum and dreaming.

Winginitt likes this.
2. Aug 11, 2019

### plncraze

#### Well-Known MemberHBA Supporter

Joined:
May 12, 2006
Messages:
1,601
335
Welcome! As a former flying machine mechanic and amateur engine builder you have a good start.

3. Aug 11, 2019

### rv6ejguy

#### Well-Known Member

Joined:
Jun 26, 2012
Messages:
3,749
2,790
Location:
Welcome aboard!

DI and VVT is magic for cars in getting good power, emissions and economy. It's less useful for aviation applications where the engine spends most of its time near WOT and high power and at relatively high rpms (between torque and power peak usually). Use this tech if you want. Nothing is stopping you, but be aware that while quite a few folks have used the OEM ECUs with success in aircraft conversions, others have been bitten by various issues with uncommanded power reductions, engine stoppages, ECUs going into limp mode and phantom issues related to a wide variety of chassis sensors not present in the aircraft, which were in the car. You need to have all these things figured out and ground tested before you fly.

You'll likely also find some challenges getting a PSRU in place which is reliable and without torsional vibration issues.

I look forward to your build using this engine but don't think it will all go perfectly until you've actually done it. Thinking outside the box is fine. Getting several hundred successful flight hours on your auto conversion is the "pudding".

I don't know of a single DI auto engine which has made the transition to flight worthy hardware without a number of issues along the way. DI is much more complicated than port injection and it certainly hasn't proven to be more reliable in cars to date. In aircraft, reliability and durability are more important than power.

It would be a worthy accomplishment to achieve all these goals with this engine. I encourage you to work on this project.

wsimpso1 and pictsidhe like this.
4. Aug 12, 2019

### pictsidhe

#### Well-Known Member

Joined:
Jul 15, 2014
Messages:
6,873
1,856
Location:
North Carolina
VVT is great when you want to optimise over a wide rev range. But aircraft don't use awide range. They spend almost all their time within a +/- 10% band. The advantage that VVT can give across that band are very small. If you optimise the timing for the middle of the band, you'll be hard pushed to measure any improvement with VVT. Aircraft engines don't need low speed torque or efficiency.

wsimpso1 likes this.
5. Aug 12, 2019

### TFF

#### Well-Known Member

Joined:
Apr 28, 2010
Messages:
11,649
3,281
Location:
Memphis, TN
Making horsepower has never been a problem. All the technology can be applied; no one has the money to make it reliable. I’m sure you remember anytime a pilot came back with a big engine problem; they are pissed. They can’t get out and walk like a car if it blows. As much as I have fun messing with hot rods, I don’t hot rod airplane engines. If you are not interested in flying, but just developing an engine combo, we will all be happy if you have success. No one will hold their breath as it is one in ten thousand you get it that good. The airplane engine is optimized for a very narrow range. It all can just be built in without the monkey motion. Why don’t they do it, not worth the monetary effort. In a way it’s the avionics for the airplane that are what the hot rod engine is for the car. Airplane people tend to show off their radios not their engines.

wsimpso1 likes this.
6. Aug 13, 2019

### pfarber

#### Well-Known Member

Joined:
Feb 21, 2019
Messages:
292
15
Location:
Pennsylvania
The main reason you can run 11:1 on 87octane is the knock sensors and DI. DI puts the fuel in the chamber at the last possible moment at very high pressure.

As for VVT it is not needed, and if you tune the ECU you can lock it out. But my thoughts are to leave it there... if I can save a half gallon taxing or while in a less then cruise power mode (like descent) then its free money. ANother thing I am thinking of is a low power climb. What if the VVT give you enough power at a lower RPM and you can climb at say 60%-80% power and not 100%? VVT being able to move that power curve a little bit might lead to bigger fuel savings. With a 600-800 mile leg an 'economy climb' might be worth a few gallons of fuel.

Car motors do have 'extras' that make them more car-freindly but almost all of that can be stripped off if you like. The bottom line is that they are reliable sources of power much cheaper than the certified motors. The ONLY real downside is the weight.

7. Aug 13, 2019

### maticulus

#### Member

Joined:
Aug 11, 2019
Messages:
6
1
I didn't get updates on the responses, but for the most part agree with all of the concerns posted, or have them in mind, especially the issues with automotive power train control modules, that act like my Mom, when I was a kid; "I don't care about what you want, I'm telling you what you're getting". Indeed, making sure appropriate code changes to eliminate the computers ability to "under parent" are an absolute must so that a drive-by-wire system is in place that behaves like drive-by-cable.

As a "grounded" enthusiast with strong roots in automotive mechanics and performance, I believe I have a very good hold on pretty much all of the concerns proposed regarding the use of automotive engines. What I don't have is an equal knowledge base regarding similar aspects relative to flight, but I'm in the right place to fix that now.

pfarber, knock sensors have been around for a long time and although they do serve as a guard against damaging spark knock, they are only a safety and not a requirement in the choice of compression ratio choice. DI does exactly what you said, injects the fuel at the last moment (among other things) making it possible to run much higher octane than previously possible on a given grade of fuel, but it is not the only reason, there's also more optimum fuel atomization from that high pressure, combustion chamber design advances, smaller quench/squish areas and perhaps some help from the oil squirters that spray the bottom of the pistons to keep heat under control.

That's why it's so wonderful and although it may not garner as much appreciation in a flight application at a theoretical glance, heads bow down to its use on the road. Keep in mind that's the level of experience I'm speaking from at this time, so my aspirations are for as much efficiency as possible and choice of engine can change as needed.

My car has pretty much the same fuel economy rating as my Mom's. Her car has the port injected 3.6L and is about 500 lbs lighter than my car with the DI 3.6L. 252hp, 251 lb-ft, 10.2:1 comp vs. 304hp, 273 lb-ft 11.3:1 comp. They both have 6 speed transmissions with final drive ratios very close and about 1/8" difference in tire height, and my car is also bigger, yet my hwy mileage is about 5 mpg better than hers on the same grade of fuel over the same route and with about 20k more miles on the odometer. Not saying it's all due to DI, but no doubt in my mind mostly. Efficiency, efficiency, efficiency.

I appreciate the input/ideas.

8. Aug 13, 2019

### rv6ejguy

#### Well-Known Member

Joined:
Jun 26, 2012
Messages:
3,749
2,790
Location:
I think you mean CR rather than octane...

Heads may bow until there are maintenance problems with the HP pumps, injectors or carboned up intake valves which are still happening on a number of brands almost a decade after SI DI was introduced. There have been no net cost savings for lots of folks who keep their cars and need to have the heads taken off to have the carbon removed from the backs of the valves.

I think you'll find the fuel efficiency gains are minimal in the aircraft environment over port injection since we don't run at part throttle much.

9. Aug 14, 2019

### wsimpso1

#### Super ModeratorStaff Member

Joined:
Oct 19, 2003
Messages:
5,917
3,146
Location:
Saline Michigan
Well, let's consider a few things about airplanes.

Cruise power - In the case of a given fixed pitch prop, torque expressed at the prop is primarily a function prop speed squared, and secondarily a function of airspeed. Power is torque times prop speed. So, if you have 100% torque at 100% rpm, that is 100% power. Want to run say 75% power - which is pretty common for cruise - that is about 91% of max rpm... (0.909^3 = 0.75). Go to a controllable prop, and you can get other combinations, like 100% torque and 75% rpm, and 75% Torque and 100% rpm, but you get the idea. Backing off to 50% power is 79.4% rpm and most of us do not run much lower. So, in cruise you will be at big fractions of max torque and rpm even at what seems to be modest power settings.

Climb power - Best rate of climb speed is the speed where we have the least power required to fly level and thus have the most power left over to climb with. This is the speed where we turn fuel into altitude most efficiently. While many of us will climb for a couple minutes at full power, most will back off some, but let's say you have 100% power available to some altitude. Well, if your airplane takes 25% of its total power to fly level at best climb speed, and you want to climb at 50% power, you will be converting half of your fuel burn to altitude. Go to a more typical cruise climb of 75%, and the same speed will result in twice as much altitude gain per minute... Going lower power in climb really drags out the time spent at climb speed. Since we are typically going only 60% of our cruise speed while at best climb, this can really extend the time spent on a trip.

Once you are flying, you will find that you prefer bigger climb rates and higher cruise speeds...

Now let's think about a typical two hour trip. Engine start and taxi takes 5 minutes, then you spend 15 minutes in take-off and climb, 90 minutes in cruise, 15 in descent and landing. If your bird makes 0.40 lb/hp/hour at 300 hp (you are detuned enough that you do not need mixture enrichment to keep the exhaust valves in the engine), that is 120 lb/hr or 20 gal/hr at 100%. Let's say we climb and cruise at 75%, and descend at 37%. At taxi, it burns 2 gal/hr, in cruise it burns 15/ hr, and in descent it will burn maybe 7.5/hr. So, you get 5/60*2 + 15/60*15 + 90/60*15 + 15/60*7.5 + 5/60*2 = 28.5 gallons. Your average fuel burn is 13.1 gal/hr. You burned 2% of your fuel during taxi, 13% for takeoff and climb, 7% in descent. With those small fractions of your fuel burn where power is back, you want to run VVT and DI to make it what, 20% better for the taxi and descent conditions. With valve timing and fueling set up for cruise already, the VVT and DI will not change climb and cruise burns, but will reduce your trip fuel burn about half a gallon. And the tradeoff is a finite life fuel pump, carboning up the valves, and a lot more work to get it all working and tuned and not destroying the engine too fast... It just does not sound like it is worth the trip.

Now if instead you cruise at min power required speeds, yeah, VVT and DI might save you some more fuel, but compared to the reduction from just going down in speed, well, it will still be pretty small.

Seriously, VVT and DI is far less attractive in airplanes than it is in cars where we rarely run above 30% of max power...

Billski

mcrae0104, pictsidhe and rv6ejguy like this.
10. Aug 14, 2019

#### Well-Known Member

Joined:
Jul 5, 2014
Messages:
384
234
Location:
Australia
Without turbo-normalising, there may be some interesting reprogramming to get the DI and VVT optimised for significantly lower MAP at full throttle : for one thing, cylinder pressures will be much lower.

11. Aug 14, 2019

### pictsidhe

#### Well-Known Member

Joined:
Jul 15, 2014
Messages:
6,873
1,856
Location:
North Carolina
The other way around for VVT. A NA engine will need the same cam timing whatever the air density and pressure, it will need tweaking with a turbo where the engine sees different relative intake and exhaust pressures. DI should already be mapped for a range of cylinder pressures. At altitude, it will 'see' part throttle sea level conditions.

12. Aug 14, 2019

### rv6ejguy

#### Well-Known Member

Joined:
Jun 26, 2012
Messages:
3,749
2,790
Location:
Mapping a DI VVT engine which wasn't turbocharged originally, to work optimally over a wide range of altitudes and not using closed loop control (running on 100LL) will be challenging for an individual without considerable time and resources- like an auto OEM has. Way more variables now present.

Throttle angle usually has little bearing on mapping. MAP and rpm or MAF would be the main determining factors.

13. Aug 14, 2019

### maticulus

#### Member

Joined:
Aug 11, 2019
Messages:
6
1
Some good points are mentioned above, rv6ejguy, I'm not sure which comment you're referring to, but my point overall is that DI makes it possible to run much higher compression ratios than carb and port injection for a given octane rating.

wsimpso1 thanks for the detail, I appreciate that kind of scientific expression. There is so much variability with auto motors vs. aviation, the likely best way to put things into practical perspective would be to compare them by power and torque curves/dyno graphs and bsfc.

rv6ejguy boosting naturally aspirated engines is a very old accomplishment and one of if not the main influence behind aftermarket engine management software/hardware that makes it possible to modify OE programming. There are limitations, but it's usually with how much the motor can take before a connecting rod crumbles. Unless all piston crafts are required to run 100LL, unleaded is also much, much cheaper if that ended up being the only fuel savings.

I believe turbocharging is a must to fully take advantage of an automotive engine of such circumstances. There's a good bit that I've not mentioned having kept in mind since it should go without saying.

As for this valve coking, high pressure fuel pump failing issue with DI motors. I have to say that is quite a general and broad paint brush application. Much of it is early production related and foreign make/model troubles, particularly BMW and Mini Cooper off the top of my head. It is not an issue discussed as problematic with GM applications that I have observed, or experienced.

I get the feeling that aviation specific motors do not require any maintenance given the concerns expressed so far with automotive motors, which come new with a strong warranty of dependability for a given period of time when properly maintained.

I had a lot of reservations about computer engine management early on, but it was mostly because of my lack of control over the computer. Now that control is back where it belongs, I appreciate the potential it offers.

I am not familiar at the moment with the series of aviation engines often employed for more comparative insight. I have attached a GM dyno chart for the latest GM 3.6L. Take a look and give me some insight on how useful, or useless that kind of performance range is for practical flight application.

While trying to find an example of a, aircraft V6 weight for comparison (3.6L= ~340 lbs) I stumbled across this: https://www.experimentalaircraft.info/homebuilt-aircraft/viking-aircraft-engines.php

#### Attached Files:

• ###### 2016_LGX_3.6L CTS DYNO.jpg
File size:
38.6 KB
Views:
15
Last edited: Aug 14, 2019
14. Aug 14, 2019

### pfarber

Joined:
Feb 21, 2019
Messages:
292
15
Location:
Pennsylvania
DI and VVT work fine. While not optimized for aircraft use the fact that a 4.6l motor can make 280hp all day long for less than $10k installed (the$10k installed number is my planning number for putting an LV3 in my BD-4B) means I'll rebuild it every 500 hours if I have to to remediate a valve issue. More power, less costs. Oh darn I have to spend $750 to rebuild a motor every 2-4 years. What certified engine owner wouldn't beg to pay that as their rebuild cost??? Although there are aftermarket FI and ECU's for motors, they are still 20 years behind. We've gone from carbs->TBI->PI->DI and every iteration has increased performance. At this point PI is the mature technology, but DI is really not having many issues. I agree that there is nothing wrong with any of them, but most newer motors are switching to DI, so you can only assume that the millions the auto companies are pouring into DI tech will do nothing but increase its presence. 15. Aug 14, 2019 ### rv6ejguy ### rv6ejguy #### Well-Known Member Joined: Jun 26, 2012 Messages: 3,749 Likes Received: 2,790 Location: Calgary, Alberta, Canada From your post "making it possible to run much higher octane than previously possible on a given grade of fuel". I think you meant to say CR. DI engines are generally about 1 to 1.5 points higher CR than the same model engine with port injection. Valve coking affects all DI engines, including GM. Google it. Lexus has both DI and PI on many of their engines to avoid the issue and some other OEMs have copied that idea in the last couple years. You might see in my signature that I've been flying a turbocharged auto conversion for 16 years. My background is building turbocharged race engines (40 years) and my business for the last 25 years is manufacturing programmable EFI and ignition systems for automotive and aviation. We sold over 10,000 controllers in that time. I know at least a bit about these two topics. Few forged connecting rods crumble with high boost, generally piston ring lands are the weakest link in OEM auto engines. I'm not a fan of sintered rods as used on many GM products. Fine for stock engines, not so much for engines making a lot more hp. These may indeed crumble. They were not designed for high boost applications. If you fly cross country, you'll have to use some 100LL fuel as unleaded mogas is available only at a handful of airports. 100LL isn't friendly to O2 sensors. Closed loop using leaded fuel isn't a reliable solution in my experience. Aircraft engines may or may not need much maintenance depending on the frequency of use, brand and treatment. A dressed 3.6L GM V6 is likely to weigh in at over 380 pounds with PSRU and accessories. Not bad for the hp. You'd be between a Lycoming 360 and 540 in weight. The Viking engines have had numerous reliability issues over the last decade of development. The jury is still out if the latest offerings are reliable and durable. Last edited: Aug 15, 2019 sotaro and wsimpso1 like this. 16. Aug 14, 2019 ### pfarber ### pfarber #### Well-Known Member Joined: Feb 21, 2019 Messages: 292 Likes Received: 15 Location: Pennsylvania My understanding is that turbos in AC are to normalize density altitude, not really to boost HP. Denso is rumored (and it came up on google) to have lead tolerant O2 sensors. From the race guys, it seems a generic O2 sensor 'lasts about 15k-20k miles'. If true, an O2 sensor should last a year or two minimum, depending on your annual flight time. Take 15k / 65mph = 230 hours of life as a starting point. Longer if you mix 100LL with Mogas. How much does it cost to IRR a magneto when it hits 500 hours???? What the price of an oil change on a certified engine? Throwing out O2 sensors (maybe$150) every 200 hours doesn't seem to expensive.

You can also tune with 100LL and O2 sensors, then remove them and use that mapping as the open loop (If I recall mapping can be altered). Won't be as good, but definitely better than running open loop.

I am not a proponent of running turbo's. I'd prefer to derate the motor (I plan to RPM limit the LV3 for 200-220HP). If things go well I can always retune for higher.

17. Aug 14, 2019

### BoKu

#### Well-Known MemberHBA Supporter

Joined:
Aug 16, 2013
Messages:
2,327
2,926
Location:
Western US
This is all very interesting. But I have to say that in all likelihood it is probably moot in the absence of a rational and methodical plan for reliably delivering the power to a propeller at a reasonable RPM.

If I was of a mind to get into the automotive conversion business (I'm not), I'd start by worrying about the PSRU, and once that was reasonably well squared away, _then_ I'd go looking for an engine to hang it on. And I'd start by looking at the lightest ones, not the most powerful.

sotaro and wsimpso1 like this.
18. Aug 14, 2019

### rickofudall

#### Member

Joined:
Oct 10, 2009
Messages:
15
3
Location:
Udall,KS,USA
Has anyone looked at the Chevy LTG engine? I've been drooling over it since it came out. All aluminum, turbo'd, and DI. Chevy advertises "light weight" but doesn't tell anyone what that means. I'd put one in a Lotus 7 or Caterham in an instant but obviously aircraft application would require some mods and a gearbox. Still, it sure looks nice.
https://www.chevrolet.com/performance/crate-engines/ltg-four-cylinder/rear-wheel-drive

19. Aug 15, 2019

### pictsidhe

#### Well-Known Member

Joined:
Jul 15, 2014
Messages:
6,873
1,856
Location:
North Carolina
While modern 4 valve car engiens do have a decent specific output, there is a cost to that.

They are optimised for cars, so mainly for low throttle use. My decrepit 150hp truck was still accelerating when I chickened out at 105mph. How much throttle is a 300hp truck going to be using at 60? Yes, they can run at full power for quite some time. But it's at a high piston speed, so life and economy will suffer. Now, use it at a sane piston speed, and you might as well be using a much lighter 2 valve OHV engine. That is the reason that LS conversions are quite popular. They don't get run at max revs, either...

At 6800 rpm, a LGX has a piston speed of 3800fpm. That's just too high for a long and frugal life.

Maybe we need a sticky why-not for this topic? It seems to come up monthly.

20. Aug 15, 2019

### pictsidhe

#### Well-Known Member

Joined:
Jul 15, 2014
Messages:
6,873
1,856
Location:
North Carolina
I've been pondering a PSRU for an industrial V-twin for around a year. Anyone who thinks a reliable PSRU is easy has approximately a zero chance of building one.

nerobro, Hephaestus, wsimpso1 and 2 others like this.