Join Date: Oct 2008
Garrett Turbochargers and the Truth
I'm posting this on behalf of Garrett who's post count is low, so their ability to make a new thread is not possible at this time.
It's a very long read, but very detailed - Garrett will be in this thread to answer any questions you may have from the below novel.
Hello everyone. Before I begin, I believe it is necessary for me to introduce myself so you all understand who I am and my intention for starting this thread. My name is Justin Ocello and since 2008 I have held the position of North American Sales Leader for Garrettís Performance Product line. That probably sounds like a big corporate title that may not mean much to you so let me explain. Essentially, I am in charge of everything that goes on in the Performance aftermarket for Garrett. When you see Garrett release a new turbo, it was my team that developed that product. If you are a sponsored racer, it is my signature on your contract and my team that is working with you. I decide who will be a distributor and, in Precision Turbo and Engineís (PTE) case, I also decide who will no longer be a distributor. Basically, the Performance aftermarket is my world at Garrett.
Now that you know who I am and what I do, I hope that I can clarify many of the misconceptions that have been expressed here over the past year or so. We make it a point at Garrett to never get involved in on-line debates, but after following the saga over the last year and seeing our name dragged into PTEís problems; we felt it necessary to put up this post to respond. Iím not starting this thread with the intention of slinging mud. I hope you will find that my statements are factual, hence the reason for me including names and part numbers.
Sometime around June 2011 we were notified of the first PTE turbo that blew up in the field. We didnít know at that point in time that it would just be one of many.
While we tried to figure out what was happening, we started seeing all of the posts on this forum with plenty of misinformation being spread around. It then became very concerning when we saw the results of Lynchís failure. From what we know, Tim Lynch was scared for his life that day (for good reason) and we were extremely relieved to hear that he walked away from that turbo failure with little injury.
In mid-October 2011, Harry Hruska worked through our distributors to send a message that he would like to speak to us. Garrettís Technical Manager and I called Harry directly. Harry has mentioned this in his previous posts and I can confirm that it is a true statement. The conversation was civil and we listened to everything we were told. It was assumed at that point in time by Harry that the Garrett turbine wheels he was using failed at the weld joint due to poor quality welding. Before you jump all over Garrett for this, keep in mind there is very little detail or information regarding the failures of PTEís own wheel Ė ďWith the first failure we switched back to a Garrett turbine wheel and that is when all hell broke looseÖWE had not used a Garrett turbine wheel for several months on ProMods when the 1st containment issue came upĒ found at PTE Wheel Failure. So failures began before the Garrett turbine wheel was re-introduced and continued after the Garrett turbine wheel was re-introduced. What is the only constant here?
Back to the Garrett wheel; bear with me while I explain a bit of history here. Our aftermarket group used to manufacture preproduction turbine wheel assembly part number 451768-0005 for use in our GT55 journal bearing aftermarket turbos. In 2009 we switched this wheel to an equivalent production part number 730595-0002 and notified all our customers (as a point of reference, PTE was cancelled as a customer at the end of 2008 ). It is our understanding, based on Harryís own admission, that Garrett part number 730595-0002 is what was being used on the PTE units that failed in the field. Again, by Harryís admission, PTE was modifying the Garrett shaft at the compressor end to fit his product design. Also, I must remind everyone that this turbine wheel assembly was designed for use in a journal bearing application. If it is to be used in a ball bearing application, the shaft would have to be machined to accommodate the different bearing design/layout. That means that the finished shaft was being modified in some way to fit into PTEís ball bearing turbo. I donít think I need to explain how tight the tolerances are on these parts and how difficult it would be to keep uniformity while re-machining a shaft. In all fairness, Harry did state that the turbine wheel assembly was being re-balanced after these machining steps. Also, keep in mind that both journal bearing and ball bearing turbos have been failing.
Now, the big question is: why are these turbos failing? Based on the posts Iíve read from Harry Iím being led to believe it is a shaft weld failure. In our experience (keep in mind weíve been doing this since the 1950ís) to truly understand the problem, and even help solve it, we asked Harry for some hardware to evaluate. During the conversation I mentioned previously, we were promised a new journal bearing unit as well as some failed hardware for our review. Harry would not ship us a new ball bearing unit because of his ďproprietary technologyĒ that he didnít want us to see. To this day, we have yet to receive any hardware from PTE directly for our evaluation. In addition, we have yet to see any returned turbos or CHRAs from our OE customers or our aftermarket customers that have failed turbine wheel assemblies using part number 730595-0002.
I should mention that the turbine wheel assembly used for our GT55R and GTX55R turbos is part number 769211-0002. This part was designed specifically for use in our ball bearing product line. So when people wonder why you arenít seeing Garrett units fail in a similar fashion, perhaps itís because we are using the parts in an application they were designed for. If for some reason those parts did fail, we can confidently say our housing will contain. Read our White Paper 2 to find out why containment is so important if you value your personal safety (see Garrett White Papers).
Letís compare some turbine wheel assemblies shall we? Below we have three parts: 1) Garrett turbine wheel assembly 730595-0002; 2) PTE turbine wheel assembly out of a Pro Mod 94 CEA; 3) Garrett turbine wheel assembly 769211-0002. Notice that the first two are small diameter shafts compared to the more stout 769211-0002 from our ball bearing GT/GTX55R. Below that is a comparison of the shaft lengths. We did not put calipers on the hub length, but take a look at the differences. DISCLAIMER: the PTE unit was supposedly built July 2011 but there was no identification tag to verify.
All that said, we cannot comment as to what is happening with the failed units because it would be pure speculation. Before anyone believes what is posted, I would welcome any of you to contact us here at Garrett Ė either PM or through Garrett.Gearhead@Honeywell.com Ė and send us your failed hardware for our evaluation. Harry mentioned that he showed a Garrett engineer a wheel at PRI. Actually, what our engineer was shown amounted to a shaft remnant from a separated turbine-shaft-wheel assembly out of a failed PTE turbo. One piece of a broken turbo is never enough to tell the whole story. Harry knows that and now you all know that. To truly evaluate a turbo you need the full picture. Otherwise, itís like trying to reconstruct an accident with only one of the involved cars and no visibility to the scene of the accident or the other party involved. Harry, I welcome you to send every piece of hardware to us for our evaluation.
The possible root causes of failure resulting in the turbine wheel separating from the shaft are weld joint quality, overspeed, an out of balance rotating group, foreign object damage of the blades, over-temp or excessive bearing wear.
But letís remember, not every PTE turbo failure had a shaft-wheel separation. There also were natural bursts of the turbine wheel from what weíve seen and heard in the marketplace. The possible root causes of failure here are a severe casting flaw, over-temp or overspeed.
Unfortunately, without hardware we canít make a distinct conclusion so I will leave it only to the possibilities.
Moving on, letís address some of the other misconceptions Iíve read recently either directly aimed at us or mentioned in this forum:
- OE parts are used in Garrett performance product. TRUE: we do use OE parts because they are tested rigorously to limit failure and meet TS 16949 quality standards. Feel free to read this on our web site (see Why Garrett). If you wonder what this means to you, ask a racer. Bothwell Motorsports ran the SAME turbos on their 10.5 car for three years Ė they removed them once in those three years because they ingested some foreign objects into the turbine side and we needed to check them out. The GTX compressor wheel design originated on an OE application Ė we just applied it to our product and gave you 10-20% more flow while doing so. Oh, and a fun fact for you: billet compressor wheels were originally used by Garrett back in the 1940s so donít let others fool you, most of the product ideas and ďinnovationsĒ in this market originated with Garrett.
- Garrett isnít part of NHRA but all they need to do is submit a turbo for consideration. PARTIALLY TRUE: NHRA surprised us with their spec turbo ruling for Pro Mod. There was no indication they were going this route and many of our sponsored teams were caught off guard. We called NHRA two days after the rule was posted online (Dec 22nd) and finally spoke to the Technical Director in January. We didnít get a positive ruling in our favor and they will not consider our turbo so thatís where we stand for 2012.
- Garrett is not involved in drag racing. FALSE: we have been a sponsor of NMRA/NMCA and continue to be.Most recently weíve added sponsorship of ADRL and NMCA West. Instead of offering contingency, we sponsor racers with either completely free hardware or discounted hardware through our distribution network as well as provide money to the racing organizations to keep the series going. If NEOPM had returned our calls/emails we might also have been a sponsor of that series as well. We may not physically be present at every race, but thatís mostly because weíre also out supporting road racing, time attack, drifting, rally, Pikeís Peak, circle track, land speed, and many other disciplines where our drivers are winning and setting records.
- Garrett turbos cannot be rebuilt. FALSE: itís not that they cannot be rebuilt; itís that we choose, in most cases, not to. Our distributors have the ability to rebuild our turbos Ė just call them and ask. Our goal is to get the customer back on the road or track as soon as possible with fewer headaches. Harry mentioned in his post that a compressor wheel cost is $620 if you need it replaced. I donít know if labor is factored in there but if itís not, I donít know many shops or manufacturers that donít charge for labor. Since I donít know PTEís business model anymore, I wonít guess as to how they price this out. What I can tell you is that if you fail a GTX5533R with a 94mm inducer (and youíre not sponsored by us or a distributor) your out of pocket cost will be approaching $2,000. Now, you will say thatís way more than a compressor wheel from PTE. You are correct, but for $2,000 you are also getting a brand new CHRA from the factory where we guarantee the part. Remember, if you damage a compressor wheel there could be other damage that is not visible on the surface. A bent shaft, a slightly damaged turbine, etc. Donít take the $620 as your total cost. We feel that our policy is the best option for the consumer and this program has been working for more than five years now.
- GTX47R turbos cost $5,521.89. FALSE: a GTX4708R MSRP is $4,247.60 (did I mention Iím also in charge of setting prices?) but as TOM TOM stated, nobody pays list see Garrett MSRP. If you ever want to check that, just call a Garrett distributor and ask.
For those of you that have made it this far I commend you. Iíve given a very detailed response to over a yearís worth of posts and misinformation. The true cause of failure has still not been addressed in our opinion. In addition, the new PTE stainless housings are nice pieces; but are they guaranteed to contain (remember the White Paper I referred to earlier) should a failure occur? It is our firm belief that the turbine wheel assembly isnít the primary cause based on what little information has been shared with us. I really hope the new housings are stronger than the previous versions. Finally, Garrett has invested in the product, the racers and the racing organizations to support your sport. Weíve done the same for many motorsports in fact. We arenít limited to drag racing and our product speaks for itself.
I hope that this will clear things up for you. I will monitor this thread and attempt to answer any questions I can but donít expect me to speculate or debate with you. Best of luck to all you racers out there and be warned, Garrett racers are hitting the track hard and will be breaking records (great job Bill Lutz on the 5.97 at MIR).
Turbo by Garrett