Failed Holset experiment (HX40 rant)

[jkimes]

I should probably go ahead and apologize now for this long-winded post.. I'm sorry, but I have to get this off my chest.

I tried and tried to give Holset turbo's the benefit of the doubt. I've read repeated posts from multiple people on this forum who have made good numbers with them (especially the hx35's) and for those people who did, I'm very happy for you. However, I didn't have quite the same "warm-fuzzy" experience with my Holset that some others have claimed.

I always tried my best to stand up for my budget friendly DSM brethren who opted to try something unorthodox, and I didn't want to believe all of the haters on multiple message boards (including Dsmtuners) who talked smack about Holsets compared to similar Garrett, FP, PTE, or BW turbos. However, as with most things in life, it seems (at least in my case) that you tend to get what you pay for and if you don't pay a premium for quality components when you start a project, you wind up paying even more in the the long run.

Take my HX40 experience for example, I spent months reading and researching compressor maps and dyno graphs to find the theoretically perfect Holset HX40 with the exact compressor and turbine wheels that I wanted, followed by weeks searching the web to find someone selling the exact model Holset I was looking for (60mm 7 blade billet HX40 pro) only to later have to immediately ship it to the shop for service when it wouldn't work properly (the first time it needed a new snap ring as well as re-clocking and balancing). By this point I was already starting to get incredibly frustrated. Everything really started to boil over later after I started hearing more and more about Holset failures due to oiling problems. So I spent even more precious time researching oiling requirements and solutions and shelled out even more money for a larger drain line and fittings, as well as an inline restrictor plus an additional inline fitting and oil pressure gauge (to carefully check my oil pressure at the turbo inlet), not to mention, hours spent meticulously porting my oil filter housing (to relieve any excess oil pressure) and welding in a larger bung to the oil pan for a bigger drain.

The kicker came much later after I sent the turbo to the shop for a second time only to discover that not only did the turbo need a complete rebuild after only 500 miles, but I had also been sold the incorrect turbine housing and needed to drop another 250 bucks and several more days time to track down the theoretically perfect turbine housing for my application (17cm^2 divided T4) along with even more time and cash for custom exhaust work to make the new housing fit properly and NOW... after all that time, money, and effort... THE TURBO WON'T MAKE ANY BOOST PRESSURE OR POWER!

I could have just saved myself a mountain of time, money, and headache and simply bought a shiny new ball-bearing Garrett, FP HTA, or PTE turbo off the shelf and to top it all off I probably would have made more power with faster spool to boot. When Polk originally tuned my car with the HX40pro 2 years ago it made a respectable 455whp on pump gas (after I had it shipped off for repair the first time because it arrived D.O.A) but later when they tried to tune for higher boost on C16 the turbo refused to boost any higher than 23psi (possibly due to the huge housing) so I gave up and towed it home. I only drove the car a handful of times after the pump gas tune. It pulled really hard at high RPMs but the spool sucked on the street. The car sat for almost a year while I slowly upgraded the fuel system, ignition, and other parts in order to swap over to speed density and attempt to make some big power numbers at higher boost levels on E85. I finally finished the car a few weeks ago and took it to the shop to have it dynoed again but no matter how hard they tried MKC couldn't get the turbo to build boost over 10psi and it never made more than 240whp even when revving all the way to redline.

So, I sent the turbo to the shop once again (this time to Justin) and I was told my HX40 appeared slightly worn but not too bad (likely due to oil contamination from leaky valve stem seals that were just replaced). Justin also noted that my hotside was way too big, so I had him rebuild the turbo and found a more appropriately sized housing (The original seller told me my Hx40 had an 18cm hotside and I never bothered to check.. turns out, not so much.. it was actually a massive 22cm). Justin completely rebuilt the turbo and put it back together for me with the new 17cm housing (he says the turbo is now as tight as the brand new HX40's he sells). It spins freely and doesn't smoke or make any noise. MKC had to fab up a new downpipe to make the new housing fit my exhaust, and now after all that time, money and work.. IT'S STILL DOING THE EXACT SAME THING!! The shop has triple-checked for boost and exhaust leaks and swapped in a different wastegate and BOV but the results were still almost exactly the same as before. They finally swapped out the HX40 with another divided t4 turbo they had in the shop (a used journal-bearing PTE 6265 w a divided 1.15 T4 housing) and the boost and spool immediately picked up and the car started making good power. WTF?? I GIVE UP!!

The blue line in the graph is the original HX40 dyno pull with the 22cm housing, the red line is the newly rebuilt HX40 with the 17cm housing, and the green line is the used PTE 6265.

You must be logged in to view this image or video.

 

[viperlp01]

That was my Hx40 turbo housing and I had no issues at all with it. I spooled the exact same hx40 7 blade never rebuilt with no regard for oil pressure control, 10an drain, custom 3in exhaust on a 2.3 with my custom exhaust manifold, 27 psi at about 4900 rpm with no boost controller. I added a boost controlled and turned the boost up to 29 psi and I hit that right around 5k. What wastegates are you running. It seriously sounds like a wastegate issue. I have seen some ####ed up turbos make serious boost pressure and with yours in damn near perfect condition, it should spool.

Seriously I have my hx40 still sitting here. If you are still up to it I would send it to ya to test(pay shipping and if it breaks its yours, it made multiple 10 sec passes and is in good shape) and you can figure out what the issue is.

 

[turboglenn]

I REALLY REALLY want to read t but can you break it down into some paragraphs, i am not being an ass saying this but i go cross eyed trying to keep track of where i'm at with that much text jumpbled up but i'm interested in hearing what's up because i've followed a lot of your stuff in the builds and such and respect your opinion, thoughts and will take your test results as serious and not just "oh well an idiot owned it" type thing

sorry to say it but i do wanna read it, and do have respect for ya bud, just PC text that big messes with my eyes

 

[Dyesuperman]

I REALLY REALLY want to read t but can you break it down into some paragraphs, i am not being an ass saying this but i go cross eyed trying to keep track of where i'm at with that much text jumpbled up but i'm interested in hearing what's up because i've followed a lot of your stuff in the builds and such and respect your opinion, thoughts and will take your test results as serious and not just "oh well an idiot owned it" type thing

sorry to say it but i do wanna read it, and do have respect for ya bud, just PC text that big messes with my eyes

In an nutshell...he had to buy and switch more parts than he thought he would have to with the hx40 and then when he thought he had it all set and good to go, the turbo wont build much boost (10psi) even after having it rebuilt and switching turbine housings. The shop dynoing his car switched on another turbo and said the other turbo built boost fine and the car made power. With the hx40 on though it would hardly build boost and make no power

 

[dexterholland04]

The hotside you were using is HUGE! No wonder why you didn't build boost at streetable RPMs. Why didn't you go with a smaller turbine housing...heck even a bolt-on BEP housing?

 

[airmail96]

He already did with no luck, he went from a 22 to a 17cm2 housing and those have been shown to spoo lbetween 4500-4900 rpm

 

[1990talontsi]

theres a wastegate problem and it got fixed when they swapped turbos
bet money on it

 

[airmail96]

If he is running a divided t4 manifold tough, isn't the wastegate pre turbo? So the only thing they could have fixed was how it was plumed?

 

[JusMX141]

I should note that the turbo came to me with zero wheel damage, but did have a bit of excessive up/down play and showed some turbine seal leakage prior to disassembling...no doubt a result of the excessive play. The amount of play that existed should have in no way rendered the turbo unable to build boost...hell I've witnessed turbos with enough play that the wheels rubbed yet they still made full boost, just a little slower than normal.

Upon teardown, both journal bearings and the thrust plate were scored, showing sign of foreign material in the oil. Of course there's no way for me to tell how long they have been like that...the damage could date back to the previous owner. No sign of heat-scoring, no sign of oil deprivation in any way.

The shaft was perfectly straight, and once rebuilt the play was definitely within spec. In fact once I filled the journals with low-pressure oil through the feed flange, the shaft play all but disappeared after the rebuild.

Honestly, I'm stumped. I've never seen such a situation where a certain turbo just wouldn't work. At first I though it was that monster turbine housing which is actually used to regulate boost on diesel engines four times the size of our gassers...now I'm clueless.

 

[jkimes]

theres a wastegate problem and it got fixed when they swapped turbos
bet money on it

It doesn't appear to be a wastegate problem. The first time they tried to dyno tune the car (blue line) was with my practically brand new 44mm tial wastegate. I also immediately blamed the wastegate so they swapped it with a donor 44mm gate off another car in the shop and got the exact same results. They then swapped in the PTE6265 turbo and it immediately started making good power and boost pressure. After that, I finally agreed to send my HX40 off for service (even though I couldn't believe the turbo could possibly have a problem). Justin rebuilt it and put on the smaller turbine housing, they slapped it back on the car and the results are almost exactly the same.

 

[forcefed86]

Did you do a compression test on the motor? Split your oil filter and look for debris? Soundsd like you have other issues to me...

 

[jkimes]

Did you do a compression test on the motor? Split your oil filter and look for debris? Soundsd like you have other issues to me...

Compression is good. This engine only has around 2500 miles on it. The valve stem seals were leaking because my machinist used the wrong size seals causing oil to puddle on cylinder 2 but I just had them replaced with brand new seals along with a brand new mls head gasket and L19 studs. The oil and filter were just changed with fresh 10w15 high zinc/phosphorus racing oil and an OEM filter.

Sell that Holset to me. Then it will be my problem.

The turbo is for sale actually. I'd hate for anyone else to deal with the crap I'm going through but maybe it's something simple we overlooked. I'm looking for another turbo to replace the Holset as we speak. Hopefully I'll have better luck this time around.

 

[bradikus7]

Sounds like you have had quite the rough time, however your original post really seems to be blaming Holsets and aiming towards you now believing they are useless? Maybe I'm wrong, but I absolutely love my Hx35 and at 24PSI, it's the fastest my DSM has ever been.

EDIT:
< That time is not on my Holset, that's T25 with stock clutch.

 

[knochgoon24]

Is the compressor cover possibly not fully seated? Have you done a boost leak test that would reveal a leak there? It's a bit of a stretch, but I have no other clue.

 

[JusMX141]

I put the same thin bead of RTV around the sealing surface of the cover that I did on Dave's HX40, and it's holding 38psi just fine.

The only thing questionable on the cover itself is that someone hacked the v-band completely off the outlet instead of just grinding it down and leaving a little sealing lip. Due to the taper of the outlet, I was afraid this was causing a giant boost leak but he claims they've checked for leaks.

 

[turboglenn]

Man, weird how most have great luck with them and then those who have bad luck seem to have horrible repeasted bad luck. Not sure what it could be from and wether it be more user setup and usage or something out of spec in the units parts or CHRA from the factory

wanna sell a failed old hx-35??

 

[jkimes]

Sounds like you have had quite the rough time, however your original post really seems to be blaming Holsets and aiming towards you now believing they are useless? Maybe I'm wrong, but I absolutely love my Hx35 and at 24PSI, it's the fastest my DSM has ever been.

I don't at all believe Holsets are useless. In fact, I'm a fan of Holset and I've been very impressed with the numbers that several people have made using them on the 4G63 platform (especially the Hx35). The point I'm trying to make is simply this;

Most DSMers will opt for a Holset because of initial cost. However, by the time you drop (typically) $400+ on a used holset (and risk it being potentially D.O.A like mine was), PLUS spend the time researching, shopping, buying, and shipping all of the necessary fittings, adapters, and lines PLUS finding a manifold that will fit and the restrictor needed to run it, PLUS the time and labor required to port your oil filter housing AND weld in a larger drain AND fab up an oil pressure gauge connection to double-check exact oil pressure at the turbo inlet, AND purchase an additional turbine housing that will fit your manifold & optimize flow for your specific application AND weld custom exhaust to fit, PLUS pay for a rebuild if the turbo has a problem only to risk the turbo still not working properly, how much did you really save?

On the other hand, you could have spent roughly the same amount of money and a lot less time and hassle by bolting on a new turbo that was designed and tested to work on similar platforms that only requires minimal extra parts and labor to install plus you get a manufacturer's warranty. As far as I'm concerned, time is worth even more than money and I've wasted way too much of both trying to get my used holset to work properly only to wind up empty handed. I got into this hobby because I love to drive fast, not because I love spending all of my time cobbling parts together and trying to fix broken sh*t all the time.

I knew the risk involved when I started this project and I know many people wound up with much better results than I did. I'm happy for those who were successful, however, after all of my experiences modifying this car, one single nugget of truth always winds up slapping me in the face; "Do things right the first time around, and don't be a cheap-ass". Period.

 

[silicosys4]

I like holsets. The option of a DIY budget turbo build is always awesome to have, but its true that when all is said and done hx35's aren't any more than a few hundred dollars cheaper than a comparable bolt on turbo, and usually are around the same price. The diesel-specific applications, engineering. reliability, application on our platform, blah blah...all mute points by now, its pretty proven they work just fine on our cars. Somehow in the last few years though, they have gone from a good budget diy dsm turbo to better than any other, according to most fanboys. The truth is, it costs money to get them to work optimally, they are either VERY reliable, or VERY unreliable, depending on how much you research your oiling needs, and in the end, 52lbs/min isn't all that spectacular for what amounts to a used $800 turbo.

 

[Rockyp]

Thermodynamics studies heat transfer and effects of the heat transfer in a given space. If a Holset is made of the same materials as Garrett or FP, thermodynamics have very little effect on the different performances in the temperature range discussed in our cars/trucks. At that point, it's about the force spinning the turbine and the shape/balance of the turbine.

The only thermodynamic law that comes into play is the first. The Zeroth law is common throughout science, the second law is two separate spaces interacting and becoming the same temperature, and the third is about a crystal. Newton's laws of motion and the formula F=ma would trump thermodynamic laws in turbo applications.

The oil in Cummins engines that lubricate the turbo is thicker and designed to lubricate a lower temp slower moving engine. Our engines flow thinner oil used to help the temperature difference as well as allow the engine to move more freely than a diesel. This means that the oil in a Cummins won't cool the turbo as well as our engines by design.

To describe this,

Heat Supplied (A) - Work performed (B) = internal energy increase (C)

Heat supplied (A) would be higher on our cars do to firing temps. Subtract that by the work performed (B), which is also higher on our system. Without actually researching the number required, lest say our engine's math would be a 5-4=1. The diesel would be 2-1=1. This is not factoring in the cooling differences meaning that our (A) would actually be smaller than 5 leaving less of a gap between (A) and (B), thus making (C) a smaller and safer number. Either way, both engine's thermodynamic prinicples should make a Holset perform at a minimum of equally.

With basic math, you could actually argue that if thermodynamics had a large say-so in the application differences, this turbo should perform better on our cars than a Dodge due to the modifications we make to our systems and the housings, but sacrifice the life of the turbo due to the higher centripetal forces.

Sorry about the long post. I just finished my third and thankfully last thermodynamics course last month and I wanted to get at least one use out of it throughout my life.

 

[ed1380]

I don't at all believe Holsets are useless. In fact, I'm a fan of Holset and I've been very impressed with the numbers that several people have made using them on the 4G63 platform (especially the Hx35). The point I'm trying to make is simply this;

Most DSMers will opt for a Holset because of initial cost. However, by the time you drop (typically) $400+ on a used holset (and risk it being potentially D.O.A like mine was), PLUS spend the time researching, shopping, buying, and shipping all of the necessary fittings, adapters, and lines PLUS finding a manifold that will fit and the restrictor needed to run it, PLUS the time and labor required to port your oil filter housing AND weld in a larger drain AND fab up an oil pressure gauge connection to double-check exact oil pressure at the turbo inlet, AND purchase an additional turbine housing that will fit your manifold & optimize flow for your specific application AND weld custom exhaust to fit, PLUS pay for a rebuild if the turbo has a problem only to risk the turbo still not working properly, how much did you really save?

On the other hand, you could have spent roughly the same amount of money and a lot less time and hassle by bolting on a new turbo that was designed and tested to work on similar platforms that only requires minimal extra parts and labor to install plus you get a manufacturer's warranty. As far as I'm concerned, time is worth even more than money and I've wasted way too much of both trying to get my used holset to work properly only to wind up empty handed. I got into this hobby because I love to drive fast, not because I love spending all of my time cobbling parts together and trying to fix broken sh*t all the time.

I knew the risk involved when I started this project and I know many people wound up with much better results than I did. I'm happy for those who were successful, however, after all of my experiences modifying this car, one single nugget of truth always winds up slapping me in the face; "Do things right the first time around, and don't be a cheap-ass". Period.

You're a bit exaggerating. Used hx35's are in $250 shipped range. The stock 2g drain is big enough to use. Just cut it up and replace with a $5 piece of hose. You may choose the warranty, but I'd rather have my $50 rebuild kit. Manifold, couplers, gaskets, etc are the same with any turbo, so they're not considered. If you don't want to deal with a custom downpipe, the bep housing is $250.
I understand your frustration with constant bad luck, but hundreds of dsm's running holsets can't be wrong.

 

[red1993gsallmt]

Justin, could the compressor wheel ever possible slip on the shaft? Im sure you would of seen evidence of that tho

 

[bradikus7]

Yeah so my HX35 was like $165 shipped. I think no matter what turbo you are running it's a good idea to check the oil pressure so buying a $30 gauge and adapter fitting seems to be no big deal. Also as already pointed out, with a modified car you already have to buy couplers, different exhaust housings, down pipes, drain lines etc.

I opted for the BEP .55 setup so in reality I could have just stayed internal and with the housing be running a Holset for roughly $465. That's $250 for the housing, $165 for the turbo, $30 for the gauge plus $20 for whatever misc hoses or fittings. My turbo had 300 total miles on it (not rebuilt) I would have thought it was new if not for the minimal soot on the exhaust wheel when I got it.

I chose to have an external WG off the o2, FP manifold, ceramic coating on everything and custom drain line. All of those things would be reusable on any other DSM flanged turbo if I ever decided to change.

 

[JusMX141]

Justin, could the compressor wheel ever possible slip on the shaft? Im sure you would of seen evidence of that tho

It's possible, but will only happen if the nut comes loose or the wheel hits the housing. Regardless, newer Holsets like the O.P.'s are component-balanced from the factory so even if the wheel alignment were off slightly it wouldn't matter.

 

[jkimes]

Once again, maybe I overlooked something but so far I'm stumped and so is the shop that's trying to tune my car. I'm going to use a new PTE 6266 for now just to get the car running again and see how much power I can squeeze out of it. If I ship this HX40 back to you (Justin) or perhaps John (viperlp01) is there any way one of you can test it again for me to see if you can get it to make boost pressure or at least double-check for something else wrong that would prevent it from making boost?

 

[Morphius]

I've had nothing but good luck with them. I bolted my wh1c on in place of a 16g, and immediately dropped 1.5s and gained 14mph in the quarter. Not to shabby for a turbo/mani/wg combo I had less than $350 in.

You bastard... how dare you.