Sunday, March 15, 2009

Duramax LLY Overheating and Thermal Feedback Primer

I love what I do, and that is a blessing. Focusing on the troublesome mysteries of the technical world, I get to figure things out: things that have no apparent explanation... the elusive.
What follows is one such mystery, wrapped in a clever disguise and hiding in plain sight for years. It remained undetected for so long because there is not a single vehicle sensor, diagnostic, or gauge that is set up to alert us to what I finally found with patience, a casual observation, a $15 gauge and a paradigm shift.
As you read, keep in mind that the principles in this article can be applied to all turbocharged vehicles, not just the Duramax. You may well find inspiration to look at other unsolved mysteries by the time we are done. If you do, I would love to hear about it.
Duramax LLY Disappointment
It was 2004 when the Duramax LLY model replaced the LB7. Promises of more power, an advanced variable geometry turbo (VGT), among other announcements, were considered worth the wait. Unfortunately, it became clear right away that there was a problem. The engine did not seem to live up to the promises. Economy was reduced, performance was hindered and many towing customers could not use it for the advertised load capacity. The vehicles even overheated. Compared to the first generation LB7 Duramax, the LLY seemed to be dragging an anchor behind it.

When I was approached about these issues, it sounded exactly like what makes me get out of bed, so I started digging. I really had no idea what I was getting into, or that finding the root cause would take years of part-time pursuit. In fact, this became a cold case for me several times. While others continued to scour the cooling system for defects, something within me searched for a flaw that tied performance loss and reduced thermal capability together: a single process or deficiency that would explain both issues, perhaps even a domino effect. As it turns out, that is exactly what I would discover.

A Cooling System Riddle

The defective mechanism is not a part of the cooling system, yet it results in elevated oil, transmission and coolant temperatures. It is not related to power generation, yet it results in reduced dyno performance. This mechanism even feeds on its own detrimental performance, growing in destructive intensity; a process so elusive, it didn’t have a name yet. What I finally settled on was named for a cyclical, power-eroding process in the forced induction system: a death spiral I call Thermal Feedback.
Thermal Feedback is the gradual loss of performance under sustained workload, which results from the byproducts of the load conditions themselves.
This story begins with a look at the purpose of the intake. For maximum effectiveness, the intake needs to:
  • Source air (oxygen) for combustion
  • Keep that air clean
  • Keep that air cool and dense and
  • Not fight the turbo compressor with negative boost.

The Turbo Fight

“Not fight the turbo compressor?” As odd as it sounds, every intake does this, and it is the necessary cost of pulling air through filtration and the intake conduit. This restrictive fight always acts on the air in the direction opposite to the direction of flow. It can be said that this restriction is the true cost of air transport, or of any fluid.
As an example of this restrictive fight, consider the water flow in a garden hose. If you have an 80-PSI water supply to the house and you turn on the hose bib with no hose attached, you will get a strong, 80-PSI gusher. Without that pressure, there would be no flow. Now attach a 100-foot long, half-inch diameter garden house and that flow is much less. This is because friction is eating away at the hose outlet pressure. At the end of the hose you have less pressure, say 30 PSI (still 80 PSI back at the hose bib). Fifty PSI is lost in the fight (hose restriction). It can even be calculated at 0.5 PSI per foot of hose. Now, let’s upgrade to a three-quarter-inch hose. With the larger flow area, water velocity is reduced, so friction and drag is reduced. Flow rate increases, and the hose end pressure is higher now, say 55 PSI. Now we are only losing 0.25 PSI per foot. The intensity of the fight has been reduced and we get more water out of the end of the hose.
The same principles at work in this garden hose example apply to induction air plumbing. Engineers can design for more or less restriction – less is always better – but there are practical limits to limiting restriction caused by compact packaging constraints: the ever-present demand to fit a wraparound, snail-like power train into a seemingly shrinking engine compartment.

Thermo-Fluid Cliff Notes

  • The fight involved in moving air is affected by three factors:
  • The amount of air being moved,
  • The diameter of the conduit through which the air is being moved and
  • The number of turns in the conduit and the sharpness of those turns.
The turbo compressor, which is nothing more than a centrifugal pump for air, must work exactly that much harder to overcome these restrictions in order to deliver the desired flow rate or boost. If you have expectations of 20 PSI of boost and there is five PSI of frictional fight, or negative boost, then the compressor must work five PSI harder and output 25 PSI total boost. More compression is needed for the same end pressure requirement.
It is the job of the compressor section of the turbo to compress the air it receives from the induction system. One undesirable but unavoidable by-product of this compression it that the air heats up. In Compressor Section of a Turbo diagram, the discharge product is hot (red). Air may go in at 100ºF inlet ambient temperature (IAT), but because of compression, that same air will be coming out at 360ºF. When induction tract inefficiencies exist, the air is heated from its ambient temperature before it ever reaches the turbo: even more compression is required resulting in hotter, less dense air leaving the turbo on its way, eventually, to the combustion chamber.

Clearing Away The Smoke

One day last summer, I was looking for suitable intake areas to locate a water mist nozzle for pre-turbo water injection, an effective way to knock down this compression heat. As I removed the entire intake, I came to the end, pulling off the compressor inlet, or mouthpiece. It is the middleman that connects the intake tube to the turbo compressor. After removing it, I looked inside and my initial thoughts were a thunderstorm of disbelief, shock, and, strangely enough, pleasure. The internal shape and dimension were a tragic crucifixion of air flow efficiency, something I would applaud only if it were emerging from my son’s play dough factory.
Mouth agape, I quickly measured it and then I had pencil on paper to determine what the internal air velocity would be under typical max power. In those few moments that changed my perspective on this issue, I had enough to conclude that I may have finally found that most elusive of defects.

Air Flow Models

Two of the factors mentioned above that contribute to air flow restriction are the diameter of the conduit and the sharpness of those turns. Now we will look at two air flow models to discover how these factors negatively affect airflow and how improved design allows the air to flow more efficiently.

Duramax LLY Mouthpiece Air Flow Model

In the first model, representative of the stock LLY mouthpiece, air is flowing around zero-radius miter corners. The conduit narrows and then enlarges again. There are several flaws at work in this model that conspire to increase pressure drop which, because of the fight (remember the garden hose), means less and, in this case, hotter air flowing to the turbo. First, in the top airflow model on the next page, note the immediate flow separation at the inside of the turn on the first corner (Label A). The main flow stream, subjected to huge centrifugal forces (would you believe 9000 Gs?) then runs toward the outside of the turn. If you look at the blue lines of flow, you will see that as the main flow stream progresses, the lines get closer together toward the outside wall (B). This is a pseudo-compression and a result of the conduit’s restrictive nature. In this regime, a large low pressure area – like a hurricane – is created on the inside of the turn (C), the lighter colors depict these forces. The low pressure is pulling on the main flow: equilibrium is struck only at great cost to flow.
The air that centrifugally flows to the outside of the turn travels further than the air on the inside of the turn. Because of this, the air on the outside ends up travelling faster relative to the air closer to the inside of the curve. Friction between those air currents results from their speed variation. As these adjacent air streams slide across each other, viscous layers of friction are formed with more energy lost as it is converted to heat. On a molecular level, it is no different – and no less real – than rubbing your hands together. The only way to reduce the friction would be to give the turn a larger radius sweep. Finally, the formation of stagnation pools in the outside corners (D), adds to momentum losses.

Improved Air Flow Model

In the second flow model, the throat size is increased. Inside flow detachment still exists (A) but the flow progresses with less compression. The low pressure area (C) is much smaller and the darker colors signify less pressure degradation. The lines of the main flow stream (B) are further apart, so there is less layered viscous friction, meaning less energy wasted as heat. Downstream, re-attachment to the inside wall occurs quickly. As a result, the downstream flow pressure is increased, going from green (less pressure) in the restricted regime to blue (greater pressure) with the throat size increase. 

The difference between these two airflow models can amount to several PSI. This shows the importance that plumbing size alone can make in undersized applications. A second factor is the turn radius. We see that flow separation occurs in both models and this is due to radius issues. Air cannot stay attached to the mitered corners of these models with any significant velocity: the unrounded inside corner is very inefficient and offending. The higher the expected velocity, the greater the radius required to avoid detachment. As a general rule, the elbow radius should be at least twice the conduit diameter. Manufacturers routinely violate this rule in deference to packaging constraints, as in the case of our tightly constrained induction plumbing.
Let’s face it, within practical limits, there is nothing we can do about this space limitation except to can make the most of the space that we do have. If the best we can do is delay the air detachment, this will decrease the low pressure zone’s size and reduce its negative pull on the rest of the airflow. This is where optimization kicks in. Provide the largest inside radius possible along with an increase in velocity-reducing plumbing size. Reworking these two concepts alone can effect an 80% reduction of the offending pressure drop.

Fantastic Voyage

Following combustion air from the airbox, it travels through a four-inch tube and finally to the mouthpiece. In the stock LLY mouthpiece, flow area gradually narrows to 2.4 inches and airflow, at full throttle, accelerates to 250 MPH. At this point the flow is forced to turn 100 degrees on a sharp bend as represented in the first model. One pound of air every second is forced to make an immediate right turn at 250 MPH. It is a staggering energy conversion that results in enormous restriction, reduced pressure, and added heat from the friction. If an aircraft wing were subject to this same force, it would snap before the pilot could even black out from the G-force acceleration. My curiosity kicked in and I calculated the centrifugal force experienced by air at this turn:
A pilot flying at 250 MPH at the air races, turning on a 100 foot radius experiences a 9-G force (nine times the standard gravitational pull), enough to black out if he’s not real careful. The 250 MPH air stream in this tight radius mouthpiece experiences a thrilling, E-ticket ride of nine thousand Gs!

One pound of air every second getting a 9,000-G pull consumes gobs of energy. Where does all that energy come from? Oddly enough, it comes from the exhaust and it results in higher exhaust gas temperature (EGT). This will be very important later.
Using a $15 Magnehelic pressure differential gauge, I tapped some sensors into the stock Zytel Nylon mouthpiece. With stock level airflow, total intake resistance was 4.8 PSI, a staggering 136-inch water column. Of that amount, the mouthpiece alone is responsible for 84% or 4.1 PSI. The pie chart on the opposite page shows the relative deficiency of each intake component.
Note how much is a result of element resistance: the smallest slice. It is the same element that gets replaced with so-called better flowing elements on promises of better performance. The truth is that even if you had the perfect element (or no element at all), you could only improve airflow a total of three percent.
This was a great opportunity to test an aftermarket element. I removed the AC Delco filter and replaced it with a drop in from another manufacturer. The three percent total intake resistance improved to two percent: calculations suggest that this one percent improvement will result in an estimated 0.15% increase in airflow. Despite creative marketing attempts to the contrary, this is what you should get, assuming the element is new. In my real world (no flow bench) physical testing, I found no measurable improvement to flow, never mind what happens when the filter gets dirty. Off it came.
As the pie chart shows, the mouthpiece creates five times more resistance to flow than all of the remaining parts of the intake, combined. The end result is that the turbo needs to make more boost in order to compensate for these losses. This means the compressor must work to create an additional four PSI, spinning considerably faster: that is bad.
The Intake Restriction chart to the right illustrates an important point: the more airflow there is, the more the potential for improvement there is. When we fix airflow flaws like this, we are removing a fixed percentage of the problem. Clearly, if we only ever use 15 pounds per minute, then we would have little to discuss because losses are well under 0.5 PSI. If, however, we want to examine race day air flow of 65 pounds per minute, there is over six PSI that can be improved upon. If I can remove 80% of that restriction, it will be a huge improvement.

Getting In Shape

Every PSI is equivalent to 28 inches of water column (IWC). At 54.5 pounds per minute there is 4.8 PSI (136 IWC) of restriction: lots of negative boost, or fight. On a mild temperature day, stock tuned, mass air flow (MAF) rate is about 45 to 50 pounds per minute, depending on elevation, or roughly 700 cubit feet per minute (CFM).
I realize you may not attribute a weight, or mass, to air. To give you an idea, the air in an eight by ten-foot room weighs 50 pounds. We don’t realize it because our bodies are tuned to resist the air that is constantly pushing down on us. Another example: every day our lungs inhale and exhale 30 pounds of air. In the Duramax 6.6L, one pound of air runs through the mouthpiece each second at maximum load. It takes us two days to inhale as much air as the Duramax workhorse does in one minute, foot on the floor. At stock intake rates, that entire eight-by-ten roomful of air is inhaled in one minute.

Induction System Restrictions

When the induction system is restricted, we can observe a relationship that can be predicted in loss equations: the restriction is a function of the square of velocity in the conduit and bends. That is, if airflow is doubled, the loss/restriction doesn’t just double, it increases fourfold. On the chart below, compare the 20 pounds per minute (A) and 40 pounds per minute (B) restriction. At 40 pounds per minute, the 60 IWC loss is fourfold the 20 pounds per minute loss of 15 IWC. If we try to plot 75 pounds per minute of airflow, we run off the chart because the losses are so great, around 250 IWC, or about eight PSI. Attempting to spin the compressor fast enough to overcome the negative boost puts it into choke and supersonic airspeeds: overspeed mechanical failure is the foregone conclusion. It is our turbocharger’s equivalent of a heart attack.

What’s Four PSI Negative Boost?
The change from the LB7 to the LLY brought with it a compressor mouthpiece that was engineered to fit in a tighter space; tighter because space formerly occupied by induction intake components had been dedicated – in the LLY – to a new EGR tube. The result was a four PSI burden of negative boost. Four PSI of restriction just didn’t seem like anything that should be so fatal.
Here are some factoids I came up with:
  • Enough wind to create a four PSI load on your body would impart a total force of 8,000 pounds, considered enough to create skin-peeling injuries. 
  • Four PSI is the weight of the entire atmosphere up to two miles above the Earth's surface. 
  • A one gallon gas can experiencing a differential of four PSI is subjected to a crushing 1,200 pounds of total force. 
  • The strongest tornado ever recorded contained less than a four PSI wind. Its wind load would impart a force of 120,000 pounds to the side of your house – equivalent to the max thrust of an eight-engine B-52 Bomber: more than enough to mulch it. 
  • A typical passenger aircraft wing loading is one PSI, so four PSI is four times more pressure than is required under the wings to keep a 350,000-pound jumbo jet aloft. 
  • And it is just conjecture but I think that four PSI could have held back the walls of the Red Sea as the Israelis passed through. Moses rocks!
Here is another way to look at it. If the element got dirty enough to become a four PSI restriction, it would see a force equivalent to three grown men standing on it. The element is considered worthless when enough dirt accumulates to create a 0.4 PSI restriction, so a four PSI intake restriction would be equivalent to stacking 10 clogged filter elements on top of each other and running the air for your turbo through them.
Now that we see how the stock LLY induction mouthpiece causes a four PSI restriction, I want to look at two alternatives that seek to mitigate the restriction. One of them addresses the induction side of the turbo and the second works downstream.

Duramax LLY Overheating Scenario

Rex Racer, who lives near sea level is given a choice of two modifications to make to his inefficient induction system. Each involves a three PSI reduction in restriction. The vehicle in question is our manifold absolute pressure MAP, sensor governed, variable geometry, turbocharged LLY Duramax. In each case, intake plenum pressure is governed to the same 34 PSI or 20 PSI boost. Air flow was experimentally determined, and ranges from 45 to 49 pounds per minute.
Modification A (cost $500) involves redesigned pre-compressor intake plumbing, resulting in a three PSI reduction of restriction upstream of the compressor (the intake side). It involves a well-designed intake with new pre-compressor plumbing.
Modification B (cost $2,000) involves reducing the losses downstream of the compressor. The improvement also yields a three PSI reduction. It involves an optimized charge air cool, new boost tubes and improved intake plenum riser.
Which option should Rex go with? Which option has the best performance improvement? What I am getting at with this scenario is which type of restriction is more critical to avoid, restriction near the compressor inlet, or restriction near the compressor outlet? Or, is there any difference?
In each case, the boost at the intake plenum is identical. That is how the variable geometry turbo works: as hard as necessary to satisfy the boost sensor located in the intake plenum. So, if the boost is the same, the only thing we can accomplish is to make that 20 PSI of air easier to produce; make the compressor easier to turn, and less parasitic to the engine. We also know that the lower the air temperature, the better it is for air density and power production. Reducing the heating of the air serves several positives. In fact, and as we will see later, there are five major benefits when reducing compression heat this way.
Also, if turbo RPM can be reduced while delivering the same boost, it requires less power to drive it. That results in horsepower given back to the wheels, also examined later.
To figure out which modification, if any, is better, we must examine thermodynamic compressor performance for heat impact. A compressor map is the place to start evaluating. It is used by designers to mate the turbo to a specific application. For example, if I needed an air supply of 80 pounds per minute, the compressor represented in the map would not work. The compressor map shows a 68 pounds per minute maximum design air flow, the right edge of the map curves.
This map also represents the thermodynamic performance characteristics of the compressor for various conditions. It is most useful for predicting thermal efficiency, or how hot it will make the compressed air. Efficiency is depicted by concentric “islands”. The limits of intended operation are bounded on all sides by curves. You don’t want to find yourself unintentionally operating near a boundary.
We can plot the full throttle operating point (label 1) on this chart when we know the conditions. The vertical axis is pressure ratio (PR). PR is simply the Compressor Output Pressure, divided by the Compressor Input Pressure or PR = COP/CIP.
At sea level, ambient pressure is 14.7 PSI. If there were no resistance losses at all, a perfect system, then PR=COP/CIP=34/14.7=2.31 (label 2).
With reasonably low resistance, it would be around 2.6, but this is not the case. In our problem there is three PSI on both sides of the compressor that are individually considered. So initially, PR=(34+3)/(14.7-3)=37/11.7=3.2.
Note that three PSI restriction is added on the outlet or work side and subtracted on the inlet side. If you don’t see why, it will be explained soon.
Plot this pressure ratio with the airflow measured at 45 pounds per minute in the map shown above. We might sense something is wrong already. Turbochargers are usually sized so that the operating point (OP) is usually located somewhere near the middle (shaded areas) in the more efficient islands, buffered from the problematic boundaries. This OP is dangerously close to the upper RPM limit of operation.
Restriction: It’s Less on the CIP Side
We need to compare OP of the stock mouthpiece with operating points for the modifications I listed above. But first, it is important to understand the impact of removing restriction, and how that changes CIP or COP.
Too often, the assumption is made that turbo CIP is the same as atmospheric pressure. This can be far from the truth, as in the case of the stock LLY mouthpiece.
On the low pressure (CIP) side, Mod A removes restriction, increasing the pressure at the turbo inlet (CIP). Mod B’s impact on COP is similar. Each is like going to a wider garden hose. Basically, if 20 PSI is demanded at the plenum, and if there is three PSI of restriction between the compressor and the plenum, then 23 PSI must exist at COP, since three PSI will be lost in between. If the restriction is removed, then COP is beneficially reduced, to 20 PSI. Summarizing, in each case there is three PSI less compression. At first glance, you might draw the conclusion that they are equivalent results. They are not.
For Mod A, CIP is increased by three PSI, PR=COP/CIP or PR=(34+3)/14.7=2.52. For Mod B, COP is decreased by three PSI: PR=34/(14.7-3)=2.91. Plot these two solutions on the compressor map (Mod A is label A and Mod B is label B.)

Duramax LLY Stock (left) and Improved (right) turbo mouthpiece.

And The Winner Is…

Mod A! It has the best combination of highest compressor efficiency and lower pressure ratio results in the lower heat production and lower compressor RPM. Remember, each mod is producing the same boost level, yet, because of the efficiency difference, the airflow amount will differ.
The turbo is easier to drive, since less work is required to drive the lower RPM, 15,000 RPM less on the chart! This means lower exhaust back pressure and lower EGT, all with about five percent more compressor efficiency (less heat). Oddly, the above benefits result from the elimination of a mere three PSI restriction which is accomplished with the more efficient mouthpiece. Even if the price tags were reversed, Mod A is still, by a large margin, preferred and will yield the best performance results all around. The more efficient configuration puts less energy into heating the air and more into compressing. The combustion air is cooler and more dense when restriction is removed. That is the key to making more power.
Note also the new charted operating point. Mod A resides in a heavily buffered location and is no longer in danger of running off the map if conditions, such as altitude, change. An altitude increase yields a lower CIP, which as the denominator of the PR equation, means a higher pressure ratio.
Beyond this, there is a bonus that is not immediately apparent. I mentioned it already but it is worth repeating. Air flow increases eight to 10 percent with Mod A. This is very important, that is 10% more air with the same boost and the same turbo that is just acting like a bigger turbo, with less heat and turbo lag; best of all worlds without the $2,000 replacement cost. In other words, restriction elimination on the inlet side decreases the heat-producing negative consequences while increasing airflow and power – and not just marginally but – dramatically!
As a side note, this is similar to what twin turbochargers do to obtain very high efficiency. The COP of the first compressor increases the CIP of the second, larger compressor.

Performance Tuning-Boost Controllers

Boost enhancement has become a very popular (cheap) performance add-on. Should you decide to augment the boost of the LLY with a common six-PSI boost enhancer you can plot the new operating point (label 3) on our map with the following:
PR = (34+3+6)/ (14.1-3) =3.87 (assumes 1100 feet).
This is a compressor overspeed every time you punch it. They do not tell you that, mainly because the inventors never considered what you are seeing here and proving that your 50 pounds of molten metal exploded because of irresponsible innovation is an uphill battle. In reality, lifespan reduces rapidly in proportion to the distance the operating point is placed from the map boundaries. Boost enhancement has created numerous problems and ruined turbochargers prematurely and this is why. The turbo manufacturer will not give any life expectancy for this kind of use. I caution, if you tow with an LLY truck above 6,000 feet with one of these devices, you will soon be replacing the turbocharger.
And that is one definition of irony: justifying the use of boost increase to reduce turbine damaging EGT, only to realize that the compressor then failed due to the resulting induced overspeed and mechanical stress.
One of the biggest pitches for these devices is reduced EGT. It is well known that high EGT will shorten turbine blade life. But some people have become consumed with this operating parameter as the all-telling longevity indicator, and this is a big, big mistake. One way to reduce EGT is to provide surplus air dilution, that is, more air than can be ignited with the provided fuel amount.
As we now know, air dilution in this manner comes with disadvantages affecting longevity, dangerously increased turbo RPM. Blades exceed Mach speed and enter choke region operation. Centrifugal blade creep mechanically stresses the metal blade, threatening wall contact. Easily, 50% of the turbo life can be eroded in a few seconds of thrill seeking. In extreme tuning cases, like stacking multiple tuners, turbo lifespan can be reduced to time served. It is clearly the most destructive way to reduce EGT. Do not use these devices for towing or other extended duty cycle work. Still skeptical?

Another LLY Scenario

What is the boost augmented pressure ratio if you are in Denver, given stock intake? You can rerun the above numbers using atmospheric pressure of 11.5 PSI (Denver) to replace 14.1 PSI and replace the 34 PSI of MAP with the augmented 40 PSI. Because of the added airflow, the restrictions are up to six PSI on the intake (cold) and 4.5 PSI on the outlet (hot) side.
Comparing LB7, LLY, LBZ
Comparing the three generations of Duramax compressor mouthpieces, we can see how the LLY made the engine especially susceptible to issues. Remember the 9,000 Gs that exists for stock flow in the LLY piece? The LBZ piece comes in at 2,000 Gs with its lower velocity and larger turn radius. The LB7, with its narrow ducts but large radius sweep, comes in at under 1500 Gs.
The LBZ did away with 60% of the problem, returning the beloved Duramax to the people. In the last issue, Joel Paynton did an exemplary job of demonstrating the changes in vehicle behavior by replacing the LLY mouthpiece with the LBZ version. The vehicle becomes thermally stable, stronger, more responsive, more economical. All of these improvements can not be addressed with cooling system add-ons and hyped-up band aids.
The restrictive LLY mouthpiece possesses at least five major performance symptoms; damaging domino effects stemming from its own shortcomings. Five different band aids would be necessary to address each of these symptoms, yet there is still one that cannot be fixed with a makeover: parasitic power loss. It is so much cheaper, and more effective to fix the source.

Geek Speak

If you are not a geek, you can skip this part, it is not important enough to matter, as long as you believe the rest of this article. By way of theoretical proof, this whole article can be summarized thermodynamically. The turbine (right) and compressor (left) are connected by a shaft, a fact that simplifies this problem enormously.
The turbine uses exhaust gases expanding across it to spin the whole assembly. This process, governed thermodynamically, cools and slows the exhaust, extracting work energy in the process. The compressor does the opposite: it uses the energy provided by the turbine, to compress (and heat) combustion charge air. What happens in the compressor section (boost) determines how much work the turbine must deliver, across the shaft, by the turbine. Simple enough.
If the compressor makes more heat energy (for any reason), then it requires additional energy from the exhaust, via the turbine.
Remember the statement I made earlier about the energy needed for 9000 Gs coming from the exhaust? Well here we are, full circle. That four PSI intake restriction that makes the compressor less efficient requires the turbine to provide more shaft power. In turn, requiring more energy from the exhaust tract which drives it. Simply, the more the compressor has to work because of inefficiencies, the more exhaust energy must be converted to work in the turbine.
The energy equation sometimes gets pretty complicated, but not here. The shaft connection makes this very easy to solve. Restating, and mercifully skipping the long derivation, the work energy delivered to the shaft by the turbine, W(t), must exactly equal the total heat energy being produced in the compressor.
  • W(t) = Q(c), or
  • W(t) = M*Cp*(T2-T1), or
  • W(t) = M*Cp*(COT-CIT)
  • M is the mass air flow (MAF) of the charge air
  • Cp is the heat capacity of air
  • COT=Compressor Outlet air Temp, CIT=Compressor Inlet air Temp.
This is the equation that unlocks the compression reality. All thermal feedback issues or restrictions or whatever, on the compression side of this operation (right side), must be serviced by the turbine (left side).
Now we can calculate exactly how much turbine effort, W(t), is required. Since COT decreases with restriction removal and compressor efficiency increase, Q(c) is reduced and therefore, turbine work W(t) is reduced. In other words, less HP goes into the turbine to drive the turbo. How much less? Well believe it or not, that skinny little shaft is seeing 80 to 110 HP at stock air flow rates. That was a big double take for me. Probing further, elevation makes it worse.
In Denver the shafts sees 120 to 130 HP unless you add a boost enhancing device and then it will see over 220 HP! Do you think this may have something to do with upper elevation turbo failures? If those numbers seem unfathomable, recall that the turbo spins at 110,000 to190,000 RPM in these examples, so the torque on the shaft is relatively low. 220 HP on a one-inch shaft is impressive still – and it all comes from the exhaust. At somewhere around 300 HP, with the overspeed this represents, the turbocharger quickly self-destructs with a sheared shaft and materially fatigued bearing. I am very impressed that a 50-pound air pump can withstand these stresses for any length of time; hats off to Garrett.

Free Lunch (Snack)

The magic of the turbocharger is suppose to be that most of the turbine drive energy is not coming from the engine’s power production, but rather from waste thermal energy in the exhaust. There are varying opinions on just how much turbine drive is parasitic and how much is free. I have seen claims range from five to 50 percent parasitic. I could not determine this exactly for this specific application, but I did determine that it is somewhat variable. For simplicity, I finally settled on 25% for evaluation purposes.
Since easily one-third or more of the diesel fuel ends up expended as waste heat in the exhaust, we tap this energy to spin the shaft connected compressor. A reality, however is that there are some mechanical losses due to back pressure created by the additional exhaust restriction of the turbine. There is a parameter that can be monitored that gives insight into how parasitic the turbine is being. It is called drive pressure, measured at the turbine inlet. Drive Pressure Ratio is the ratio of this drive pressure to the boost developed by the compressor. A ratio of around 1.0-2.0 is usually a healthy system. That means 20 PSI of boost would be seen with typically 20 PSI of back pressure, for example.
In reality, testing has revealed actual values of 2.0-2.5. Our 20 PSI of boost is accompanied by 40 PSI of drive pressure. This suggests that there is too much restriction somewhere. If we make a compressor side change that removes two PSI of restriction, that should remove three times (six PSI) of drive pressure; and that is what happens, it drops to 34, a 15% reduction. This, in effect, makes the turbo less parasitic, and makes boost production more of the free lunch we have been taught to expect. With this change, the drive ratio is 34/20 or 1.7, a vast improvement.
Let’s go back to Denver and 220 turbo shaft horsepower. If we use our 25% parasitic assumption, then we will see 55 HP lost (.25 x 220) at the wheels to additional back pressure in the above over-boost condition. By fixing the mouthpiece, that original 220 shaft HP is reduced to the range of 160 to 170 HP, a 30% reduction (while simultaneously experiencing a 10% increase in compressor mass air flow). Now the parasitic loss to back pressure is .25 x 160 or 40 HP. That difference of an additional 15 HP (55-40) can be measured at the wheels.
It doesn’t end there. Coming out of the charged air cooler (CAC), there is a 90ºF air temperature reduction with this improvement, from 330 to 240ºF. Power depends on air density and air density is a function of both pressure and temperature. The general rule states that you reap a one percent power gain for every 10ºF of air temp improvement. For this 90ºF, that means nine percent, or another 24 HP gain. Added to the 15 HP parasitic savings, the total is 39 HP. This 10% power gain clearly qualifies as thrust you can feel. Assuming this added power can be used for faster speed, cooling airflow will be result. Also exhaust gas temperature has gone down, a very nice incidental. Some have reported over 200ºF reductions.
You can find an Excel spreadsheet calculator for these statements at: (link redirects to Michael's web site).
To sum up what happens with excess restriction, more of the engine's power gets converted to waste heat in the induction system, power that is deprived from the wheels. That waste heat is manifested as increased EGT and CAC heat saturation. There are several performance debilitating affects of restriction, and removing it is essential.

LLY to LB7: Heat Relief

OK, now let’s ask an obvious question: “Why doesn’t my old waste-gated LB7 suffer these same issues?”
Glad you asked; and it is true. For the 2001 to 2004 LB7, as exhaust meets the restriction of the turbine, it proceeds through the turbine or is routed around the turbine in bypass, via the waste gate. In turbocharged vehicles, the turbine is the exhaust's biggest restriction. As such, when the turbine is pushing back (back pressure) hard enough, the gate opens and the exhaust bypasses the turbine resulting in limited shaft HP. In other words, the turbine will only be pushed so hard, limited by the waste gate. It is immune to failure and a little like trying to do pushups in wet cement: push too hard and you sink right in (or, in this case, exit via the waste gate). When the exhaust pushes too hard, that added pressure forces exhaust to go around the turbine, preventing added turbine and shaft stress. It’s perfect.
Now back to the shaft connection, and the compressor. That spin force is turbine (and shaft) horsepower. Since the turbine torque on the shaft is waste gate-limited, then the compressors job is also limited. From the math above, we also know that the turbine power exactly equals compression heat. Can you see it now?
The heat of compression is limited by the turbine's waste gate Conversely, the waste gate is set to limit the amount of heat/compression that the shaft-connected compressor is pre-certified to produce. The waste gate is perfect induction overheat control; perfect thermal feedback containment. The absence of the waste gate must be addressed if we are to completely address proper induction design.
On the 2004 and later LLY, the VG turbo in the LLY has no waste gate… no hardware safeguard against thermal feedback. The LB7 shaft may be limited to 100 HP, regardless of how much thermal feedback exists, and so 100-horsepowers' worth of compressor heat. With no waste gate, the LLY can run well beyond 200 HP, so induction heat can double with thermal feedback influence – and does.
Now if anyone tells you that a waste gate limits boost, you can politely disagree and say that it does not. It limits boost-related heat production. As described earlier, thermal feedback cannot be eliminated completely without a work-limiting device such as a waste gate. What can be done? Nothing is as simple and reliable, but tuning software can control boost so that the VGT attempts to maintain a constant pressure ratio with changing ambient pressure conditions. This has challenges, and will be the topic of a future article that will introduce a new tuning tool, EFILive.

Collateral Damage

This added heat hurts in many less obvious ways. Ironically, the heat overload has been known to melt MAP (boost) sensors, typically damaging them in a way that causes them to read low. You guessed it, that creates more boost, and more meltdown.
Also, if you own a medium duty truck with the black Zytel nylon CAC end tanks that are leaking and oil stained, they have melted and deformed; now you know why. Nylon and aluminum have dissimilar heat expansion properties.

Duramax LLY Induction Overhaul Kit

In addition to a machined LLY mouthpiece, this affordable six-piece Induction Overhaul Kit ends these mouthpiece-related performance restrictions and puts the LLY in a new class of performance and towing capability. The materials have been tested to 350°F. It does not require any tuning changes, does not require MAF re-scaling, does not require dipstick relocation and has a very clean, OEM appearance. It is far and away the smartest change you will ever make and is guaranteed with full refund. Certainly do so before playing with tuning or boost. Price: $295. Find more information about it at (link redirects to Michael's web site).


  • Fifteen to 45 percent reduction in turbo pressure ratio, three to six percent more compressor efficiency (seven to nine percent at higher elevation). The main results of this are:
  • Thirty to 45 horsepower increase on grade (typical) and as much as 60 HP with the fan off.
  • EGT lowered by 100 to 200ºF (typical) up to 250ºF on highly boosted applications.
  • Turbo longevity increased with lowered shaft RPM (~20,000 to 40,000 RPM less.)
  • Economy is increased with lowered parasitic activity, 1.0-2.5 MPG in field reports. This soon pays for itself.
  • Steep reduction (40 to 80%) of noisy fan activity on hot days or heavy loads.
  • Removes the cause of LLY load induced overheating: a heat-soaked CAC sitting right in front of the radiator. This also lowers transmission and engine oil temperatures.
  • Other benefits include: a reduction in spool up time, compressor stall/turbo bark is eliminated and lower under hood temperatures.

Wrap Up

Waste heat processes, made worse by plumbing inefficiencies, rob you of fuel economy, performance, longevity and load-carrying capability. There are no band-aid cures, you must remove the cause, or you will be left with all the other symptoms.
The LLY suffers from the introduction of undersized plumbing, the demise of the waste gate and the absence of a suitable work-limiting substitute. This limits the workload capability of all LLY vehicles and disables the cooling system radiator with added ambient heat generated by the charged air cooler. This situation is made worse by the lower ambient pressure found at higher elevations.
With boost augmented at altitude, there is six PSI of inlet restriction, and 3.5 PSI of outlet restriction. PR=(40+3.5)/(12-5)=6.2. It can’t be plotted because the page isn’t big enough! If you were unlucky enough to be reading this after you have already replaced your turbo, now you have an idea why it died. Use added boost only to define the power you want. DO NOT use boost devices exclusively to lower EGT. Instead, figure out why EGT is high, and then remove the cause. In the process, you will solve several other problems simultaneously. When you increase boost just to lower EGT, you create additional problems. Instead, first find and eradicate the source of performance eroding restriction, you will be impressed how this approach transforms a vehicles performance characteristic. It also increases MAF, Longevity AND Performance while reducing EGT with no harmful consequences;. If you try to use 32 PSI of boost at this altitude, the PR calculates to over 12.0 with the added air flow and restriction. Expect to pay.

Performance Garage Notes

It is widely accepted that 32 PSI boost represents peak capabilities for power production on the stock LLY. At this level, it has been shown that there is eight PSI of inlet restriction. If six PSI (75%) of this is removed, then you can use that six PSI to increase power. It also suggests that the turbo failure threshold is also six PSI higher. That is a bold increase in airflow capability, with lower charge air temperature.
Conversely, if you typically use 32 PSI now and make this modification, you should reduce boost to around 26 PSI to maintain the same MAF and power level. This will add to the heat reduction benefits.
The lower EGT in each case is considerably safer as well. The result of fixing it is more airflow, less heat, and significantly better performance. Ironically, this is what the intake companies want you to believe that they are doing for you. Ironically not one of these companies includes a better flowing mouthpiece. And ironically, four years after the debut of the LLY, not even one of these companies with their “Million dollar flow bench” builds an intake to address a real issue where real improvement can be had. Hey, I am just a guy with a $15 gauge and persistence. Skeptical? I hope so. It is disappointing to see how many people just believe everything that they are told.
You may have seen efforts to address the unconventionally high exhaust drive pressure ratios (2/1 and higher, up to 3/1) by trying to improve exhaust-side turbine flow, with debatable results. After all, the turbine is shaft-connected to an anchor, the compressor. All we did here was trim a good amount of weight off the anchor. The result? EGT and drive pressure dropped, a thermodynamic consequence of making the shaft-connected compressor more efficient.

1 comment:

  1. If you use this smoker your neighbor will impress and they will want to get best electric smoker like this. But this electric smoker is not available everywhere you can find this smoker only on our site.