Monday, December 15, 2008

Duramax LLY Overheating: The GM Solution and Beyond

My first experience towing with my 2005 LLY Duramax was a thrill – and a disappointment. In stock form, other than some gauges and a power program that added about 100 crankshaft horsepower and roughly 180 foot-pounds of torque, there was no disappointment with the power – it was awesome! I could yank around a 27-foot fifth-wheel travel trailer and almost forget that it was there. I had all the power I needed beneath my right foot and the ability to accelerate on any hill I encountered.
The disappointment was my cooling fan: it engaged far too often for my satisfaction. It wasn’t even hot outside. The ambient temperature stood at a meager 65°F: I wasn’t towing particularly hard either, running at about 65 MPH on mostly flat ground and only a few smaller hills. A stock Duramax would pull just as well on this route and generate the same amount of heat – for the most part – I really was not into all that much extra power.

After towing for a few hours, I could very easily predict the behavior of the cooling fan. With every short hill I pulled, the fan clutch would engage for about three to five minutes near the top or after I crested the hill. To improve my lot, I got into the habit of running every hill: accelerating before the base of the hill and backing out slightly as I climbed. It did not help much.
As a technician, I remembered that a few LLY owners had complained about their cooling fans running too much. Not having experienced the aggravation for myself yet, I had told them the fan was doing its job. Now that it was happening to me, I realized that the cooling fan was running way more than I would have expected. While traveling, I mused over possible explanations for the undesirable behavior of my cooling fan. I noted on my digital gauges that the engine temperature was running between 198 and 220°F, a significant jump from the typical 190 to 194°F when running empty. My cooling fan was engaging at the right temperature, but I couldn’t figure out why, at 65 MPH, there was not enough airflow over the radiator to keep the engine temperature low enough to keep that fan quiet.
My expectations made things even worse. I used to tow with a 1980 GMC one-ton dually with a warmed-over 454, making nearly 400 horsepower. She was fully-loaded with air conditioning. The fan clutch on that old gasser would never engage at speeds above 50 MPH, even in the hottest ambient temperatures and towing the heaviest loads. Above that speed, there was always enough airflow over the radiator to keep the engine quite cool. We had little mercy on that truck and it always handled whatever we threw at it. Now, here I was with the most advanced diesel pickup engine on the face of the planet, fully designed for towing, and the cooling system seemed barely adequate and certainly annoying.
Before resting the entire blame on the cooling system, I needed more information. Was my radiator, intercooler, or A/C condenser restricted with debris? I hoped that something other than the cooling system itself was at fault. Before our next trip, I gave it a close inspection. My hope for a scapegoat was dashed: no debris or restrictions. We headed out for a camping trip in southern British Columbia. There I was, growing more nervous with each mile that passed. High altitudes test the limits of any tow vehicle and long hills are much more frequent in the Rocky Mountains. If I was struggling in tamer terrain, what would these conditions do to my truck? On our route, we encountered a long pull we call Killer Hill. It boasts some eight percent grades with an average climb of five percent up to a peak of 6,810 feet, running a little over three-and-a-half miles. Ambient temperature that day was about 80°F. As we climbed the hill, my worst fears were realized: I watched as every alarm on my gauges fired off. Exhaust gas temperature (EGT) was peaking over 1450°F, engine coolant temperature (ECT) was over 235°F. What made this more annoying was that I ran the bottom of the hill at about 75 MPH and backed out of it to make the climb. I ended up crawling up Killer Hill in low gear at about 35 MPH – defeated; and, even at that speed, the cooling fan roared all the way! My 1980 GMC would have flown over the top of that hill, with my right foot on the floorboard at 70 MPH, towing the load we were pulling, no problem. We had climbed that hill with her pulling heavier loads in hot ambient temperatures with nary a peep out of the fan. It seemed bitterly ironic to me that a significantly modified 454 from 1980, using a stock cooling system, would perform so much better than a 2005 Duramax diesel making about the same power. 25 years had not seemed to do much for cooling system technology. Disappointing? I could hear that old GMC laughing at me, taunting me over my conversion to diesels. And my LLY DMax made no reply. Something had to be done.
At this point, I was also trying to figure out why I had not heard even more complaints about overheating at the dealership. There were a couple of owners who had bad experiences, but they seemed to be isolated, more extreme cases. More research, however, revealed that overheating under these circumstances was, in fact, a significant problem – even more so in hotter climates. We have a more moderate climate, and the hills in our area are generally quite mild, at least in comparison to runs like Killer Hill. Also, I think that the owners with whom I interact are more likely to think that an overheat during a hard pull in hotter ambient temperatures is simply to be expected. But I had experienced better: that 1980 GMC one-ton. It never overheated once, under any condition.

Exploring Solutions

I was already aware of a GM service bulletin (06-06-04-036D) addressing overheating in extremely hot ambient temperatures. The bulletin applied to all 2004 and 2005 LLY Duramax engines. GM’s solution retrofits the new-style LBZ air intake to the LLY engine. I had a difficult time getting my mind around how a change in intake systems would solve an overheating issue. As a result, I did not pay much attention to the bulletin. Instead, I spent my time examining the airflow over the cooling stack: the A/C condenser, intercooler and radiator. The configuration seemed quite restrictive to me, especially the intercooler. I wondered how the radiator could ever receive adequate airflow to do its job. I also put some thought into how close the fan was to the engine. The airflow would have to take quite a turn at that point after it was pulled through the radiator. I was starting to believe that the whole problem was caused by poorly designed airflow over the cooling stack. I had begun to map out an air-to-liquid intercooler – intending to remove the factory air-to-air unit and substitute a liquid heat exchanger – that would allow better airflow in the cooling stack. Of course this would have required a serious engineering effort unto itself: a large capacity air-to-liquid unit that would handle the DMax massive airflow and heat rejection requirements, an additional liquid radiator, electric coolant pumps, thermostatic controls, all the fabricating and plumbing and so on. An expensive, complex, and highly experimental solution which may have only helped improve airflow over the cooling stack and little else.
Being a GM technician, their proposed solution lingered in my mind. As skeptical as I was about it resolving the overheating problem, it was obvious that the engineers had spent considerable time designing the new intake system. The LBZ air intake offers a much better design than the older LLY. This is true for a couple of reasons. First, the air filter proved itself to have higher capacity, lasting much longer on our dusty oil field roads than its LLY predecessor. Secondly, and most importantly, the LBZ design is a true cold-air intake system. Any hotrodder worth his salt knows that a cold-air intake is an absolute must for optimal performance; but this truth applies not only to hotrodders. For this reason alone, I was considering building or purchasing an aftermarket cold-air intake system. To understand the significance of a cold-air intake, it is crucial to understand the affect that cold air has on the turbo. Colder air means higher density air; this denser air “helps” the turbocharger – and the entire engine – by allowing it not to work as hard at its task of compressing the air before sending it on to the engine.. The turbocharger works more efficiently (meaning less power taken away from the engine) and with less heat.
I started digging into some of the Internet forums and found that the overheating issue is quite a subject for debate. In fact, I began to wonder if research on the Internet would yield any useful information. Fortunately, I did run across a Michael Patton article (these were pre-maxxTORQUE days) called Thermal Feedback Loops in Turbocharger Applications that piqued my interest. And yes, it was as fun to read as it sounds. It sought to explain how a proper cold-air intake system could potentially solve the Duramax overheating issue. I was intrigued. GM engineers certainly seemed to agree according to the service bulletin that they provided. I downloaded the article. As it turns out, the problem was not all that difficult to understand. In fact, it should have been obvious to me just by looking at the air intake system. To be fair, extraordinary effort was required (on Michael Patton’s part) to figure out the source of the problem in the first place.
I want to consider an example from the world of sound to illustrate what happens with the LLY under-hood “warm” air intake. Almost everyone has heard feedback from a sound system and “squeal” when someone turns the gain on a microphone too high. The reason for the feedback is simple: when the microphone hears sound, it feeds that sound into the amplification system. If it happens to hear sound coming from the speakers at a high enough level, it sends that sound back into the system only to be amplified more. That sound then is reproduced by the speakers at a higher volume, allowing the microphone to hear it at that louder level only to send it into the amplification system again. This happens in a very fast cycle until you hear the characteristic squealing noise or “feedback”. And then it takes very little time for the noise to drive everyone crazy. That is the essence of feedback: when something generated by a system gets re-introduced into that system and is made more intense by the operation of the system. In the case of a sound system – as with the LLY overheating issue – this effect causes a very undesirable result.

LLY Underhood Airflow

For the same kind of thing happens with the LLY, only with heat, not sound. As you drive happily down the road, the air coming into the airbox is generally close to ambient temperatures, due to the air pressure on the front of the vehicle. However, the stock LLY airbox is fully open to under-hood air. When the engine temperature starts to rise from towing, that warmer air from the engine does influence the air that is pulled in by the airbox. The vicious cycle begins when the cooling fan kicks in. The hot air coming off the cooling stack gets blown over by the fan to the area of the airbox. This hot air is then sucked into the air intake and ingested by the turbocharger and, ultimately, the engine. That in itself is not a good thing – remember the hotrodder principle: cold air is best for power – but it is only the beginning of the thermal feedback loop.
We have to take into account two other components in the Duramax intake system: the turbocharger and the intercooler. First, the turbocharger needs to compress the incoming air to the desired pressure in the manifold, around 20 PSI at peak fueling in this case. At this point, two phenomena cause the temperature of the compressed air to rise sharply over its inlet temperature:
  1. The action of compressing air
  2. The heat generated by the turbo compressor itself.
No compressor is ever 100% efficient. Extra heat energy is added due to the inefficiency of the turbo compressor – normal operation for any turbocharger or supercharger. Enter a second component, the intercooler. It improves performance dramatically by cooling the heated, compressed air that comes from the turbocharger. This system works reasonably well for its intended purpose, creating a dense, relatively cool, air charge for the engine. But what happens to air passing over the intercooler to cool the air inside coming from the turbo? It is heated: not only by passing over the intercooler but also by passing through the radiator.

These two graphics illustrate the airflow of the LLY engine. The first (above), occurs when the cooling fan is “off”. Notice that the area of the air induction receives relatively cool air. The second (below), occurs when the cooling fan is “on”. It is clear that the operation of the cooling fan sends heated air into the air intake area – the beginning of a disabling thermal feedback loop.

This creates warmer underhood air temperatures. Some of this warmer air finds its way into the air intake. This becomes a particular problem when the cooling fan engages: then there is tremendous temperature increase in the air that is pulled into the intake. Of course, that heated air gets fed
right back into the turbocharger, only to be heated much more as it takes yet more energy to compress it to the desired pressure. The superheated air ends up being cooled by the intercooler, this time causing the ambient air that flows over the intercooler to get much hotter. That now extra-heated air ends up flowing over the radiator and back in the intake again and our thermal feedback loop is in full, red tide bloom.

Considering an Aftermarket Solution

Now, consider how poorly the radiator works when the air running through it is much hotter than the outside air. This feedback loop leads to the runaway overheating condition that I had experienced on Killer Hill. When this revelation finally got through to me, I went right over to a Duramax and had a close look. I could clearly see how this problem could happen just by looking at
the intake and the intercooler system. The solution became quite simple in my mind: get a cold-air intake. It seemed clear to me that someone at GM had dropped the ball when they installed an underhood warm-air intake in these trucks.
I considered using an off-the-shelf aftermarket cold-air intake; but there is a need to be choosy here as well. One concern with many aftermarket intakes is their cold-air effectiveness. Most do have a baffle which somewhat shields the air filter from underhood air, which is desirable. But for the LLY we wanted complete – not partial – isolation to stop the runaway thermal feedback problem. That left me with two options: Bank’s Ram-Air system or Volant’s cold air intake. Both are completely sealed from the engine bay and have the option of providing a cold air ram scoop. However, there was another concern that made me hesitate: dust.

Dust is a mortal enemy of the piston rings and of the leading edge of the turbocharger compressor wheel. Some aftermarket intake systems can allow, in my opinion, too much dust through. Most if not all cotton/gauze/foam-oiled type air filters are not fine enough. With the aftermarket filter only catching the larger particles, most smaller particles pass through easily. In contrast, this is not a problem with the stock paper filter – it has more filter surface area and a finer filtering ability than the cotton/gauze/foam-oil varieties.
Air filters, in themselves, could provide a lengthy discussion for another article, but let me sum up with a couple of pictures. For me, the evidence becomes quite obvious by visually inspecting the air intake. In the stock intake duct (pictured, above left), there is virtually no dust after many miles (90,000) of service. The only dust present is a very light, thin film that barely rubs off with a swipe of a finger.
However, the unnamed aftermarket intake (pictured, above right) leaves a heavy film of dust on the sidewalls of the intake duct. The dust is also evident when inspecting a turbo compressor wheel: visible wear and sandblasting on the leading edge of the vanes. This may not be as evident with trucks used primarily on the highway, but it is a significant concern with trucks used on dusty, dirty roads.
That turned me to thinking about fabricating my own cold-air intake with the stock air filter. I used a lower airbox from a new-style 2007 GMC Sierra gas truck. It fits the Duramax air filter, and it will fit the classic style trucks. They have larger openings and a large foam seal that sits tight to the fender well. Letting in air from the fender well on a Duramax requires removing or cutting the seal over the holes from the factory. I opened both of these to allow cold air into the system, but there was one problem with this: the openings in the fender are not nearly large enough to feed the air-hungry Duramax. My air filter restriction gauge typically showed 50 to 80 percent restriction with a brand-new air filter. So, the easy, almost-free method was not feasible.

The GM Solution

The LBZ intake was looking more and more like an attractive option. I returned to the GM bulletin and put together the parts I needed to upgrade to the GM LBZ air intake. In the process, I discovered a few other parts that GM wanted installed to ensure the effectiveness of their intake system: there were three extra baffles and a duct to install on a Chevy truck in addition to the intake system itself. On a GMC truck, even more was required: seven seals to install around the headlights and hood as well as a foam baffle. That led me to believe that there was more to this cold-air intake than meets the eye. GM engineers were obviously trying to manage airflow into the air inlet of the filter assembly. It seemed logical to me, based on the efforts of the GM engineers, that this system would prove much more effective than the LLY air intake.
Another component of the LBZ intake caught my eye as well; one that is not provided for in the service bulletin. The air duct that clamps onto the front of the turbocharger itself is completely different between the LBZ and the LLY. I had an opportunity to compare both ducts and discovered that the LBZ duct looks like it will allow far more air than the LLY duct. In fact, the LLY duct looks terribly restrictive. At the elbow – where it bends into the turbo inlet – there is a sharp, 90º corner on the inside radius. I’ve ported enough engines to know that a sharp, 90º inside corner is nothing but trouble. At the air velocities this air intake sees, there should be tremendous turbulence at that 90º corner and through most of the bend. The result of this design is that the turbo must work even harder to draw air for boost. This means less efficiency and more heat. It made me wonder why GM did not provide the LBZ turbo inlet duct in their service bulletin. But I had seen enough to know that I wanted to try the new part on my truck. (The redesigned LBZ intake (left) is much larger and more efficiently designed than its LLY predecessor (right). The restricting design of the LLY intake contributes strongly to the vehicle’s overheating issue.)

Testing: “One, Two, Three, Four”

In order to measure how efficiently the new intake would provide cold air while the cooling fan was engaged – the Achilles Heal scenario for the stock LLY intake – all I really needed to measure was the intake air temperature. That is a simple task with the scan tool equipment provided by GM. We were looking at more than just the intake air temperature, however, so we needed a way to determine if the LBZ upgrade actually mitigated or, better yet, prevented the runaway overheating. By comparing the engine temperature rise between test runs, we could paint a reasonably accurate picture of how much less heat the intercooler was sending on to the radiator right behind it. Another parameter that would be good to watch was the turbocharger vane position. With the variable-vane turbocharger on the LLY, it would be easy to monitor how efficiently it was working. The variable-vane control system closes the vanes to provide more boost and opens the vanes to provide less boost. This boost control system is advantageous because it is more efficient in using the exhaust gases to drive the turbo. It also can be set up to reduce the spool-up time. If the turbocharger is working more efficiently, it takes less vane angle to generate the same amount of boost, generating less heat in the process. This means less heat for the intercooler which also means less heated air sent to the radiator. The turbocharger vane position parameter can also be monitored by the scan tool. Intake air temperature, engine coolant temperature and turbocharger vane position data would give a pretty accurate picture of just how well the new LBZ air intake worked compared to the LLY version.
I set up the truck with the same tow tuning and fifth-wheel trailer that I used during my initial overheating experience on Killer Hill. Each run would start at roughly the same temperature and be carried out at the same speed, 65 MPH. To gather the data, I used EFI-Live Flashscan V2, which allowed me to log multiple engine parameters at the same time and store them on my laptop.
Four test runs were required to provide good test data – one for each modification:
The stock LLY intake. The truck was returned to stock cooling condition: the fender holes were covered back up. Torture for my truck and for me, but necessary to provide a baseline for the testing.
Simulating an aftermarket cold-air intake. The LBZ airbox would be installed without any additional baffles. The fender holes would be opened back up. The LBZ airbox gets close to the fender and draws air from the fender and behind the headlight. But it is not perfectly sealed without the extra baffles. Without the extra sealing, it would simulate most aftermarket cold-air intakes I have experienced.
Testing the intake installed to GM bulletin specification. Following GM’s instructions to the letter by installing the extra baffles and hardware, this test would be able to clearly show the effectiveness of GM’s solution.
Adding the LBZ turbocharger inlet duct. This part was not included in the GM service bulletin, but it looked like it would make a difference in turbocharger efficiency. One final run with this part installed would be able to determine if it nets an improvement GM’s solution as posted in the service bulletin.
With the fifth-wheel trailer hitched and the truck loaded up with my laptop, some tools, and all the new parts, I set off in search of a good test hill, one that was long enough to get the fan to engage and still have at least 60 seconds of pulling in order to demonstrate the overheating effect. As I remembered, the cooling fan would normally come on after I pulled the hill, while on level ground. Sure enough, every time the fan came on, the intake temperature shot up 20ºF rather quickly. That proved conclusively that the stock air intake was getting flooded with warm air coming off the radiator whenever the cooling fan clutch engaged. The engine temperature remained between 206ºF and 212ºF when this happened: nothing to get alarmed about (maximum engine temperature specification is 250ºF). Interestingly, the GM diagnostic test for the cooling fan clutch lists an approximate cooling fan engagement range of 185 to 205ºF engine coolant temperature. This number has a wide range because the cooling fan clutch itself engages based on the air temperature that is coming off the cooling stack, which depends on the A/C condenser, the intercooler, and the radiator. There is also a hysteresis – simply put, an unknown – in the engagement of the clutch. For instance, if the engine temperature is climbing quickly enough, clutch engagement will lag behind while the temperature rises quite high. These variables render engine temperature a poor indicator of exactly when the cooling fan will engage. Cooling fan clutches have been replaced due to perceived inconsistent operation when, in reality, they were working fine. After some driving, it became apparent that none of the hills around my hometown would provide a definitive test. There was only one hill within 60 miles that would even turn the fan on while the truck was still climbing, and that only allowed about 30 seconds of fan operation; not enough to drive the engine into an overheat. There are two factors that contribute to the runaway overheat condition: temperature and altitude. Local temperature was only 62ºF and the altitude was around 2,500 feet; not a good testing environment for this particular problem. I could not do much about the air temperature so I needed a considerably higher altitude. I only knew one place within range of a single day’s trip that fit that description. I had to return to Killer Hill.

Facing the Killer

There is something tangible about the character of Killer Hill. People remember it, even though they may not be aware of my nickname for it. Maybe it is the solitude of the remote location. There are no full time human inhabitants, just a seasonal lodge about five miles southeast of the top. Cell phones do not work, period. Even the wildlife seems scarce in this area of the Canadian Rockies. Vast glaciers occupy the gaps in the mountain range; their unpredictable crevasses have claimed the lives of several adventurers. Imposing towers of granite surround the narrow valleys and there are crystal blue glacier-fed lakes and snow-covered mountain meadows. I can not imagine anyone not being awe-inspired, if not fear-inspired, by the grandeur of these mountains and valleys. Killer Hill itself climbs to Bow Summit, located on the famous Jasper-Banff highway (also known as the Icefields Parkway) that spans two national parks. Though remote, the highway is a well-traveled tourist route during the Summer. In the Winter, the area is deserted due to extreme weather conditions. In the Spring, the tourist traffic picks up again, so I would not be completely on my own if I needed help. Still, it made me a bit nervous to torture test my truck on a highway hill in a wilderness mountain range 200 miles from home.

The trip out to Killer Hill made it clear that altitude does make a difference. As the truck travelled higher into the rarefied air, the cooling fan operation became longer and more frequent. When I arrived at the hill I made one preliminary run at lower speeds to get a sample data log with my laptop and determine if I needed to make any adjustments to my test strategy. This also allowed me to select turn-around points and to normalize the engine temperature. The long run back down the hill would serve to cool the engine and get it ready for each subsequent test run. All of the runs would start at an engine temperature of 185ºF. At 65 MPH, the intake air temperature for each run would begin at 52ºF. The hill starts at 5,800 feet and ends at 6,800 feet; just the conditions that I needed for my tests. The hill runs 3.7 miles long and there would be no problem getting the fan to engage, even at the cool ambient temperature of 47ºF.

Test Run #1: Stock LLY Air Intake

For the initial part of the first run, the engine temperature rise was not as sharp as my first experience on this hill. This was likely due to the much colder ambient temperature, about 47ºF compared to about 80ºF. I seemed to be climbing the hill with no significant issues. When the fan came on, however, things headed south rather quickly. Intake air temperature shot up to 103ºF – double the initial 52ºF. The engine temperature continued climbing until I could smell coolant at 241ºF. The engine had yet to reach its overheating specification of 250ºF but I was not going to chance it. I ended up backing off part of the way up the hill and continued slowly for the rest of the pull. At the top, I checked for the cause of the coolant smell – it appeared to be overflow from the surge tank, nothing to get alarmed about. Still, I was happy I had not taken the risk and pushed my truck harder. This test confirmed that the stock intake is a poor performer.

Test Run #2: Simulating an Aftermarket Cold-Air Intake

I removed the stock airbox and installed the LBZ airbox without installing the rest of the baffles specified in GM’s service bulletin. I also uncovered the holes in the fender to provide more air. I reprogrammed the ECM for the different MAF sensor scaling for the new airbox and moved the engine oil dipstick. Back at the bottom of the hill, I punched it again. This run did perform slightly better than the first. The intake air temperature was just slightly cooler than the previous run, the fan turn-on point was higher up the hill. But when the fan did engage, the same intake air temperature climb – thermal feedback – came into play. The ramp up was not as dramatic as it had been with the stock intake, but it was there. Intake air temperature reached 84ºF during the climb. Eventually, just within sight of the top of the hill, I was forced to back down. Engine temperature was back at 241ºF. This simulation of an aftermarket cold-air intake represented an improvement over the stock intake, but only a marginal one. For me, it reinforced the need to be very choosy about aftermarket cold-air intakes, if you should decide to go that route. Only those products that are true cold-air intakes – that completely seal off the filter from exposure to underhood warm air – would help relieve the thermal feedback that occurs in a stock LLY.

Test Run #3: Testing the GM Service Bulletin Solution

For the third test run, I installed all of the extra baffles and one duct specified in the GM service bulletin for the Chevy. It became quite obvious why those baffles are necessary – they completely seal off the air inlet from the rest of the engine bay and they manage the airflow so that the engine can draw sufficient external air. Turning around at the bottom of the hill, I began my third run. The fan came on at nearly the same point as the previous run, only this time the intake air temperature only reached 73ºF (compared to 84ºF on the second run and 103ºF on the first run). Yes, all the hot air coming off the intercooler and the radiator was still affecting the intake air, but the result was significantly better than in the previous runs. In fact, this run was the first time I could hold 65 MPH over the top of the hill. The engine temperature reached 237ºF. It was interesting to note the rate of the engine temperature rise. It rose quickly to about 235ºF, and then plateaued with the fan engagement, eventually peaking at 237ºF. This phenomenon indicated the effectiveness of breaking the heat feedback loop. With a lower intake air temperature during fan engagement for each consecutive run, the rest of the cooling system performed better and it did not seem to take much reduction of intake temperature to make the difference that we needed. Between the first and third runs, there was a peak intake air temperature difference of 40ºF. And between the second and third runs, there was only a difference of 11ºF. That change, however, made enough of an improvement to reduce the heat rejection of the intercooler significantly, allowing the radiator, in turn, to work better.
Any remnants of skepticism about the ability of the upgrades called for in GM service bulletin 06-06-04-036D to improve cooling system performance were now fully set aside. There is no question that GM’s solution works to reduce hot intake air feedback. However, if I had a hotter day, a heavier load and a slower climb, it may have been possible, still, to drive the engine into overheat; it would just take longer than it would with the stock air intake. The result was not as good as I had expected it to be. Still, with the typical towing conditions that I experience, the runaway overheat was essentially solved. Now there was one test run left. Would upgrading to the LBZ inlet duct make a difference? Or had the GM solution already brought the LLY to its peak cooling capacity?

Test Run #4: Installing the LBZ Turbocharger Inlet Duct

I felt good at the end of the third run. My truck was still in good shape: no coolant scattered over the ground and the GM service bulletin solution had made a measurable difference. I changed the turbocharger air inlet duct at the top of the hill in good spirits. I actually thought about heading home at this point – my fuel level was getting low – But I had committed myself to this test. I was not prepared for how impressive a change one part could make.
After turning around at the bottom, I accelerated back up to 65 MPH again. The truck felt quite different from the previous runs – it came up to speed too easily. After the second corner, starting up the long straightway, I knew, without even looking at the data pouring into my laptop, that the LBZ inlet duct was making a huge difference. My truck geared up much earlier, running fifth gear where I had been running fourth on the three previous runs. It felt like I had thrown 2,000 pounds of weight off the trailer. Also, the climb in engine temperature was definitely slower. In fact, the fan turn-on point was much higher up the hill than it had been for any of the previous runs. When the fan did come on, the engine temperature read 230ºF, where it stayed for the rest of the run. This was the first time that I had seen the fan keep the engine temperature under control, meaning that the temperature did not continue to climb for the rest of the run.
This test proved a very strong indicator that the LLY turbocharger inlet duct is unduly restrictive. The fact that the engine temperature was controlled on this last run – with the LBZ part – demonstrated that the intercooler was shedding less heat into the radiator. This tells us that the turbocharger was working more efficiently, creating less heat as it compressed the air to the desired boost level. But why did my truck appear to make more power?
To answer that question, I had to go back and look at the data logs recorded by my laptop. Remember that the turbocharger vane angle tells us how efficiently the turbocharger is working. Higher vane angle corresponds to more closed vanes, which requires more exhaust energy to drive the turbocharger. When this occurs there is more exhaust back pressure on the exhaust stroke of the cylinders. Less vane angle corresponds to more open vanes; less exhaust energy is required to drive the turbocharger. To see the significance of this, imagine the turbocharger as a variable supercharger, with the exhaust system being the equivalent of the drive belt on a supercharger. In the case of a belt-driven supercharger, the energy used to drive it comes from the crankshaft. Some of the combustion energy produced goes to drive the supercharger through the crankshaft and belt. A turbocharger has a similar parasitic effect, yet it is more efficient because it uses some of the waste heat energy in the exhaust system. However, the turbo still requires combustion energy from the exhaust stroke to generate boost, which ultimately steals some mechanical energy from the engine, much like the belt on the supercharger.
Consider the turbocharger vane angle as a rough measure of the exhaust energy being used to drive the turbocharger. Removing the restrictive LLY turbo inlet and replacing it with the free-flowing LBZ turbo inlet should make the turbocharger easier to drive, like removing a restriction from the supercharger. The data from the test runs supported this conclusion: the actual measured vane angle for all three prior runs to maintain peak boost levels was 68%. For the run with the LBZ turbocharger inlet duct, the vane angle was only 45% to maintain the same boost level. A 23% reduction to maintain peak boost is a huge difference in terms of the variable-vane turbo. That reduced the combustion energy (a result of reduced exhaust back pressure) required to drive the turbocharger by a large proportion. In all four runs, the load on the truck was generally the same: maintain 65 MPH with a constant mass and the same incline. The horsepower required from the crankshaft to drive the wheels is basically the same for all tests. However, the combustion energy underwent a measurable reduction with the fourth run, as less combustion energy was required to drive the turbocharger as measured by the vane angle. The data logs revealed another interesting detail that reinforced the fact that the truck was indeed making power more efficiently. For the length of the third run, made with the LLY turbocharger inlet duct, the truck spent 22% of the time in fifth gear. For the fourth run with the LBZ inlet duct, the truck spent an astounding 70% of the time in fifth gear. That, in itself, was an impressive change. It clearly demonstrated that a large parasitic load from the restrictive LLY turbocharger inlet duct had been removed. It also confirmed another suspicion that I had for a long time. There is a very short run of LLY engines in early 2006 that had all the LBZ parts on them. Their horsepower and torque ratings were the same as my truck, but they always seemed to feel stronger. Now I understood the primary reason why.
This change also has a dramatic impact on the cooling system. Less combustion energy equals less heat generated in the engine for the cooling system to accommodate. Consider that more efficient turbocharger operation reduces the heat workload of the intercooler as well, reducing the heat supplied to the airstream in which the radiator resides, improving radiator efficiency. All this results in a compound blessing: reduced airstream temperature that improves radiator performance as well as reduced engine heat load for the radiator. This creates another pleasant effect: reduced cooling fan operation, which became very apparent on the return trip home. On the way to the test hill, the fan engaged 23 times. On the trip back, it only engaged six times, and for shorter durations. A single part reduced the waste heat generated by the engine and turbocharger by a large margin, with a corresponding increase in power to the ground. Amazing, isn’t it?. When you look at the difference between the two parts (page 46) – it’s no wonder the turbocharger inlet duct was redesigned.
Which brings up another question: why doesn’t GM include this part with the service bulletin? The new, cold-air intake, by itself, does slow down the possibility of heat feedback, but the LBZ turbocharger inlet duct buys much more protection against overheating, not to mention increased power and efficiency. Add to that the increased longevity of the turbocharger itself due to a lower operating RPM: a win-win situation, no matter how you look at it. Surely GM engineering is aware of the benefits; after all, they redesigned the part. My recommendation for anyone who has had GM’s service bulletin applied to their truck: get the new LBZ turbocharger air inlet duct installed – even if you have to pay for it. It will pay you back in fuel savings alone, especially when towing.
One reason that GM would not include the LBZ turbocharger inlet duct in its service bulletin is that the 2004 Chevrolet Silverado with the LLY Duramax has a hood clearance that is much lower than the 2005 Chevrolet Silverado. The LBZ turbocharger inlet duct is somewhat taller than the LLY and so there may be a hood clearance issue with that specific truck. I spent considerable time trying to find a 2004 Chevy to investigate this, but they are few in number. Expect an update on this point. (This part will also fit the LLY-powered 2004-2005 GMCs and the 2005 Chevrolet Silverados.)
Unfortunately, the LBZ turbocharger inlet duct will only fit with the rest of the LBZ air intake system. It will not retrofit to the LLY. That means that the whole system will need to be upgraded to the LBZ specification. There is an alternative, however, if you choose to purchase an effective cold-air aftermarket intake, get an ‘06-’07 LBZ complete intake kit. That will bolt-up to the new LBZ turbocharger inlet duct, and will likely cost less than purchasing the entire GM system. A caution worth repeating: in order to stop the feedback loop, which is the other problem in this intake system, you need to purchase a system that completely isolates the underhood air from the intake.

Killer Hill Vanquished

For me, the question of whether or not the GM service bulletin is effective in reducing the possibility of runaway overheating has definitely been answered. The new LBZ intake assembly, installed according to the GM service bulletin, buys some extra cooling system capacity. As mentioned earlier, however, I still feel that the truck could be pushed to overheat when operated in more extreme conditions with only the LBZ cold-air intake installed. The real revelation in these tests is that the addition of the LBZ turbocharger inlet duct allows the whole air intake system to operate properly, keeping the turbocharger and the cooling system under much better control. That part in itself makes a truly dramatic improvement, one that I would not have fully appreciated until I installed and tested it. For this reason, I am somewhat disappointed that GM doesn’t include this part in their service bulletin. It really does make a difference.
For anyone running a power program or chip, installing the LBZ intake with the turbocharger inlet duct is highly recommended, as the air demands of the engine are even greater. In fact, if I was to recommend any modification to the 2004-2005 LLY Duramax, this now tops my list. The benefits in efficiency, cooler engine operation, longer turbocharger life and horsepower are definitely worth the price of admission. If your truck is still under warranty, you may even be able to get the dealership to install all these parts at no cost. If not, don’t worry about the cost - the new air intake system will pay you back. If you live or work in dusty areas, the longer air filter service life alone will be a benefit. You will be surprised at the performance improvement of your truck, especially with the new LBZ turbocharger inlet duct.
As it stands right now, my truck is a joy to drive, with quicker turbo spool-up and much-reduced fan operation. It also generates noticeably more power, and should yield better fuel economy while towing (fuel economy improvement will vary with load and conditions). Killer Hill no longer daunts me. I can tow in confidence, without worrying about blowing coolant all over the ground or stressing my turbocharger. The reduced fan operation alone is worth it and the memory of that 454 powered GMC no longer gets to me. With this simple modification, my Duramax will tow every bit as well and return better fuel economy in the process.


  1. This is great, I don't suppose you have any pictures of the installed parts?

  2. Great write up. GM offers a kit, part # 19167951 that I assume contains all the required parts. Do you have a part # for the turbo intake?

  3. The bulletin also indicates that the PCM needs reprogrammed. Was that not an issue with your power adder?

  4. I did 5 things recently on my 2005 GMC 2500HD LLY, 214,377 miles. It now has much more power and gets 23-26 mpg on hwy.

    1.) Changed air filter. Still using AC Delco A1618c, which is the best. Avoid K&N oiler filters to keep the MAF clean.
    2.) Cleaned the MAF using approved spray cleaner.
    3.) Installed LBZ turbo mouthpiece.
    4.) Installed 4 inch intake tubing kit (goes from LBZ MP to the filter housng).
    5.) Bought PPE Standard ECM programmer. using level 2, which is +90hp and +160 torque. Now running 400hp and 765 torque.

    This truck hauls ass! Runs great! I look forward to driving it everyday. :0)

  5. I just read your story.... Its an inspiration to the seasoned, keep up the good work. make America great again......

  6. This is brilliant. This means I will not have to sell my 07 and buy a 17. How can I get a list of the parts or how to do it or where to take it to get this done? someone can email me at if they have any idea what this costs to convert and yes - I want to do the full thing that he described.

  7. Since I do not know how to reach out to the author directly, does anyone know anyone that is able to do what he did? I don't mind driving to someplace and paying someone to do it the right way (within 3 or 4 states distance that is). I'm open to ideas but I'm pulling an 18,000 lb rig around the country right now and having to baby my truck to keep it from overheating and I REALLY want to do what the author did.

    1. I installed the LBZ turbo mouth piece but elected to use a fresh air intake box kit from Killerbee. This was due to cost and not having to reflash the computer because of a relocated maf sensor. It does help although it is not a total cure.

  8. Old Sawmiller, would you feel comfortable either posting on here a "how-to" install the LBZ (what parts you bought and step-by-step) or even emailing me at and telling me how. I'm either looking to pay someone (if I can find someone that knows how) to do it or try to do it myself (Im not very motorhead enabled) as I need to do this.

  9. it was a wonderful chance to visit this kind of site and I am happy to know. thank you so much for giving us a chance to have this opportunity. High Performance Cylinder Sleeve

  10. Thank you Soooooo much for this write up!

  11. Hi I really appreciate all the great content you have here about this Tub reglazing services topic. I am glad I cam across it!

  12. The read was good and like the other people would like parts and pieces list I have 2005 gmc 3500 daully 6.6 des. Crew cab Towing a 44 ft. Toyhauler

  13. I just recently installed the Cooler towing .com auxiliary radiator, without electric cooling fan. My fan clutch has not engage since the install, haven't towed yet but it's been over 100 degrees in the high desert with the AC on high climbing hills no full lock up on the clutch fan. Super excited about the performance.

  14. Much of the crashes including tow truck drivers occurred while they were filling lorries, altering tires, or carrying out various other jobs while dealing with the emergency roadside assistance of the roadway. The motorists might have been functioning or standing below impaired Lorries after an effect. They additionally might have been sidetracked as well as not taking note of their environments.