Jump to content
SAU Community

Recommended Posts

Blasphemy!!!!

Eh?

Noise is inefficiency, as it takes energy to make the noise. Mine is almost silent.

This one is for Lithium - :P

The GTX is a carry over from OEM applications where noise control is very very important. While it is a newer aero package, it does make sacrifices to be a quiet as it is. Typically all full blade designs are used for targeting a narrow band of operating conditions on a compressor map, and we initially investigated doing this as well in 2005 when we began our design proccess for the HTA wheel aero package. Ultimately we abandonded it due to the limitations that the full blade style puts on the width of the operating map. The HTA (actually HTA15) surpasses the performance of the full blade variants on both the surge and choke flow ends of the compressor map, oddly enough these are the exact 2 regions of the map that most people operate in without even knowing it and this is why the HTA outperforms the GTX when compared head to head with matching inducer diameters. In my trip the the MHI turbocharger research laboratories in Nagasaki in the summer of 2011 I was introduced first hand to the extreme effort these OEM put into noise control for aero on compressor wheels, it is their number 1 concern - this shocked me. They do not focus on 5:1 pressure ratios, they dont focus on surge line compressor efficiency, they dont focus on flat RPM lines and wide maps - they want the turbo to be quiet - end of story. Mercedes doesnt want to buy loud turbochargers.

The GTX is not a blank page design for high power small engines, it is something Garrett already created for an Motor Company and it was billet, so it fit the marcket needs. For example it does not take the blade thickness down to the minimum, it places a higher priority on million mile powertrain warranty performance. Ever had a race car that ran for a million miles on the same turbocharger?

The HTA was designed as a blank page effort targeting ONLY high pressure ratio efficiency (5:1), surge line efficiency, and flat RPM lines thru as much of the map as possible. To this day we do a steady stream of GTX conversions to HTA, if you know anyone that needs any pull-off GTX wheels, give me a ring.

It sucks that FP don't get taken more seriously by more people when there isn't really any reason not to. They have been doing the HTA turbos for ages, with proven results all around. I love the fact that (and this has come up a few times) the FP3582HTA and the GTX3076R apparently quite similar in response, it makes quite a few turbos redundant.


Compare (hypothetical) performance of a few turbos in that size range

Power potential

1) GTX3582R

2) FP3582HTA

3) GT3582R

4) GTX3576R

5) GTX3076R

Spool:

1) GTX3076R

2) FP3582HTA

2) GTX3576R

4) GT3582R

5) GTX3582R

Being second best on both lists kindof puts it on a different level in terms of balance of power vs response really.

There is an increase in price by my calculations though, and the thread was comparing the Garretts only I thought.

I will definitely be looking at the FP3582HTA for the Evo 7 stroker build I have coming up, hopefully it lives up to your expectations of it Lith. Is there a split external housing available?

I like many others are sceptical due to the inflated dyno figures coming out of the US, with local results I would be more confident I guess.

copy n pasted this

"Did an SR20 last week with a GTX2863 (with 0.64 AR rear). S14 SR20DET in Datsun 1600 on Dyno Dynamics dyno running 98 RON fuel. Impressive little turbo that delivered exactly what I was expecting – significantly more power than the T28 with only a slight lag penalty. The impressive part is how much more power! We saw 230rwkw (= about 380 hp at crank) on 17psi from an S14 SR20 with stock cams.

Comparison based on other SR's I've tuned with std cams on same dyno (I use 15psi as my standard full boost figure):

std T28 = 200rwkw on 15psi, making 15psi at 3300 to 3500
GTX2863R = 230rwkw on 17psi, making 15psi at 3750
GT2871R (52t) = 215rwkw on 17psi, making 15psi at 4000 to 4300


What remains to be seen is how it performs once cams are fitted and what happens at higher boost levels. Cams traditionally give 20rwkw gain on ANY SR20. So that’d make 250rwkw on 17psi from this little beasty! That’s really worth talking about
! "

Basically a disco spud on steroids, couldnt believe the smaller 71 made more then the gt2871. Another guy from Adelaide on NS put the same gtx2863 on his s15 with step 1's, made exactly 250rwkw on 17psi, he reckons he wil get 260rwkws.. but the gtx2867 would be even deadlier

Edited by discoPumpkin

There is an increase in price by my calculations though, and the thread was comparing the Garretts only I thought.

I will definitely be looking at the FP3582HTA for the Evo 7 stroker build I have coming up, hopefully it lives up to your expectations of it Lith. Is there a split external housing available?

I like many others are sceptical due to the inflated dyno figures coming out of the US, with local results I would be more confident I guess.

Other people were discussing it so I figured I'd add my 2c. And while people turn their brains off about dynos overseas, at the end of the day the results measured on the same dyno will be relative. Turn the numbers off and pay attention to the relative curves if you are have beliefs of what numbers are realistic based off your own ideologies.

I have a dyno graph with a back to back of a GT30R and GT35R I did on my RB25. Ill post it up over the weekend.

Basically showing what an awesome turbo the GT30R is, only giving away 23rwkw at the top end but could potentially be less if boost held (GT30R was maxxed).

GT30 hitting full boost 600rpm earlier.

std T28 = 200rwkw on 15psi, making 15psi at 3300 to 3500

GTX2863R = 230rwkw on 17psi, making 15psi at 3750

GT2871R (52t) = 215rwkw on 17psi, making 15psi at 4000 to 4300

Basically a disco spud on steroids, couldnt believe the smaller 71 made more then the gt2871. Another guy from Adelaide on NS put the same gtx2863 on his s15 with step 1's, made exactly 250rwkw on 17psi, he reckons he wil get 260rwkws.. but the gtx2867 would be even deadlier

I did some tuning on a 4cylinder which previously ran a 52trim GT2871R which has since upgraded to a GTX2867R - it previously was tuned to 16psi on BP98 (which was at the limit of it's fuel system) and hitting 12psi by 3600rpm. We took it out for initial testing and road tuning and It now hits 12psi by 3200rpm, and the fuel system was at it's limit on 13psi at peak power rpm with the same AFR targets so safe to say it flows quite a lot better. Not sure if the T28 hotside will start becoming a limiting factor if/when the boost gets wound up, it is still on a .64 hotside.

I did see a reputable Turbo manufacturer taking the piss out of them? Not a good indication! lol

I did see a reputable Turbo manufacturer taking the piss out of them? Not a good indication! lol

lol they say on there site that they replaced a fp green turbo on a evo 9 with their atp turbo and made 30kw more on the same boost.

Ive still got my heart set on the hta3586

It sucks that FP don't get taken more seriously by more people when there isn't really any reason not to. They have been doing the HTA turbos for ages, with proven results all around. I love the fact that (and this has come up a few times) the FP3582HTA and the GTX3076R apparently quite similar in response, it makes quite a few turbos redundant.

Compare (hypothetical) performance of a few turbos in that size range

Power potential

1) GTX3582R

2) FP3582HTA

3) GT3582R

4) GTX3576R

5) GTX3076R

Spool:

1) GTX3076R

2) FP3582HTA

2) GTX3576R

4) GT3582R

5) GTX3582R

Being second best on both lists kindof puts it on a different level in terms of balance of power vs response really.

Where do we think a 6262 would fall in here?

Am trying to plan a few things, your opinion would be valued and I think that would be a good addition to this topic anyhow.

Thats a hard one with the amount of comparable data out there (or not) however from a personal opinion standpoint I feel more comfortable about the idea of a GTX3582R. I reckon if someone tried both back to back on the exact same car the spool/power difference (or balance of) will be very close however with the Garrett you are getting an ACTUAL OEM quality unit you know you can count on, while the Precision turbos are not shite they just don't have quite the reputation of the Garrett.

So yeah, I consider the PT6262 and GTX3582R more or less interchangeable (or side by side) in those lists - money no object I'd take the GTX, PT6262 is the slightly more budget option.

Fair enough.

Bro's sigma may be getting a 6262 as it still falls reasonably within his budget (chancing a JB). The set up is too rugged for a finicky DBB item (particularly a garrett, hes killed a few).

At the same time I am starting to consider the higher end of the power scale for myself, but I can see your faith lies more in the 82HTA which I was expecting. I am just fearful of the price, at over $1,000 more landed its a big investment. There is a lot of love for the 6262 on an angry SR, but they are easily cracking 500kw; somewhat more than I am aiming for yet the lag difference between that and the smaller variants are more than acceptable.

Anyhow, back on topic. Get me on FB if you want to deliberate a little further.

Going from a ball bearing Garrett to a journal Precision for durability is a curious way of going... unless you are thinking in terms of the relative price of replacements?

Sent from my GT-I9505 using Tapatalk 2

  • Like 1

Repair and replacement is one of the issues, for sure.


The motor is just not great for the application and EGT is quite high when pushed, melting the plastic bearing cages with ease. Despite mega particle filters being run in line were also having issues with oil contamination. None of this covered by warranty or necessarily rebuildable. We are tired of buying cores.

Just going to opt for something a little more robust (in theory) and cheap to repair/replace. JB fits turbos fit the bill, PT is one of the decent options (to make some power).

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now



  • Latest Posts

    • Well, that's kinda the point. The calipers might interfere with the inside of the barrels 16" rims are only about 14" inside the barrels, which is ~350mm, and 334mm rotors only leave about 8mm outboard for the caliper before you get to 350, And.... that;s not gunna be enough. If the rims have a larger ID than that, you might sneak it in. I'd be putting a measuring stick inside the wheel and eyeballing the extra required for the caliper outboard of the rotor before committing to bolting it all on.
    • OK, so again it has been a bit of a break but it was around researching what had been done since I didn't have access to Neil's records and not everything is obvious without pulling stuff apart. Happily the guy who assembled the engine had kept reasonable records, so we now know the final spec is: Bottom end: Standard block and crank Ross 86.5mm forgies, 9:1 compression Spool forged rods Standard main bolts Oil pump Spool billet gears in standard housing Aeroflow extended and baffled sump Head Freshly rebuilt standard head with new 80lb valve springs Mild porting/port match Head oil feed restrictor VCT disabled Tighe 805C reground cams (255 duration, 8.93 lift)  Adjustable cam gears on inlet/exhaust Standard head bolts, gasket not confirmed but assumed MLS External 555cc Nismo injectors Z32 AFM Bosch 023 Intank fuel pump Garret 2871 (factory housings and manifold) Hypertune FFP plenum with standard throttle   Time to book in a trip to Unigroup
    • I forgot about my shiny new plates!
    • Well, apparently they do fit, however this wont be a problem if not because the car will be stationary while i do the suspension work. I was just going to use the 16's to roll the old girl around if I needed to. I just need to get the E90 back on the road first. Yes! I'm a believer! 🙌 So, I contacted them because the site kinda sucks and I was really confused about what I'd need. They put together a package for me and because I was spraying all the seat surfaces and not doing spot fixes I decided not to send them a headrest to colour match, I just used their colour on file (and it was spot on).  I got some heavy duty cleaner, 1L of colour, a small bottle of dye hardener and a small bottle of the dye top coat. I also got a spray gun as I needed a larger nozzle than the gun I had and it was only $40 extra. From memory the total was ~$450 ish. Its not cheap but the result is awesome. They did add repair bits and pieces to the quote originally and the cost came down significantly when I said I didn't need any repair products. I did it over a weekend. The only issues I had were my own; I forgot to mix the hardener into the dye two coats but I had enough dye for 2 more coats with the hardener. I also just used up all the dye because why not and i rushed the last coat which gave me some runs. Thankfully the runs are under the headrests. The gun pattern wasn't great, very round and would have been better if it was a line. It made it a little tricky to get consistent coverage and I think having done the extra coats probably helped conceal any coverage issues. I contacted them again a few months later so I could get our X5 done (who the f**k thought white leather was a good idea for a family car?!) and they said they had some training to do in Sydney and I could get a reduced rate on the leather fix in the X5 if I let them demo their product on our car. So I agreed. When I took Bec in the E39 to pick it up, I showed them the job I'd done in my car and they were all (students included) really impressed. Note that they said the runs I created could be fixed easily at the time with a brush or an air compressor gun. So, now with the two cars done I can absolutely recommend Colourlock.  I'll take pics of both interiors and create a new thread.
    • Power is fed to the ECU when the ignition switch is switched to IGN, at terminal 58. That same wire also connects to the ECCS relay to provide both the coil power and the contact side. When the ECU sees power at 58 it switches 16 to earth, which pulls the ECCS relay on, which feeds main power into the ECU and also to a bunch of other things. None of this is directly involved in the fuel pump - it just has to happen first. The ECU will pull terminal 18 to earth when it wants the fuel pump to run. This allows the fuel pump relay to pull in, which switches power on into the rest of the fuel pump control equipment. The fuel pump control regulator is controlled from terminal 104 on the ECU and is switched high or low depending on whether the ECU thinks the pump needs to run high or low. (I don't know which way around that is, and it really doesn't matter right now). The fuel pump control reg is really just a resistor that controls how the power through the pump goes to earth. Either straight to earth, or via the resistor. This part doesn't matter much to us today. The power to the fuel pump relay comes from one of the switched wires from the IGN switch and fusebox that is not shown off to the left of this page. That power runs the fuel pump relay coil and a number of other engine peripherals. Those peripherals don't really matter. All that matters is that there should be power available at the relay when the key is in the right position. At least - I think it's switched. If it's not switched, then power will be there all the time. Either way, if you don't have power there when you need it (ie, key on) then it won't work. The input-output switching side of the relay gains its power from a line similar (but not the same as) the one that feeds the ECU. SO I presume that is switched. Again, if there is not power there when you need it, then you have to look upstream. And... the upshot of all that? There is no "ground" at the fuel pump relay. Where you say: and say that pin 1 Black/Pink is ground, that is not true. The ECU trigger is AF73, is black/pink, and is the "ground". When the ECU says it is. The Blue/White wire is the "constant" 12V to power the relay's coil. And when I say "constant", I mean it may well only be on when the key is on. As I said above. So, when the ECU says not to be running the pump (which is any time after about 3s of switching on, with no crank signal or engine speed yet), then you should see 12V at both 1 and 2. Because the 12V will be all the way up to the ECU terminal 18, waiting to be switched to ground. When the ECU switches the fuel pump on, then AF73 should go to ~0V, having been switched to ground and the voltage drop now occurring over the relay coil. 3 & 5 are easy. 5 is the other "constant" 12V, that may or may not be constant but will very much want to be there when the key is on. Same as above. 3 goes to the pump. There should never be 12V visible at 3 unless the relay is pulled in. As to where the immobiliser might have been spliced into all this.... It will either have to be on wire AF70 or AF71, whichever is most accessible near the alarm. Given that all those wires run from the engine bay fusebox or the ECU, via the driver's area to the rear of the car, it could really be either. AF70 will be the same colour from the appropriate fuse all the way to the pump. If it has been cut and is dangling, you should be able to see that  in that area somewhere. Same with AF71.   You really should be able to force the pump to run. Just jump 12V onto AF72 and it should go. That will prove that the pump itself is willing to go along with you when you sort out the upstream. You really should be able to force the fuel pump relay on. Just short AF73 to earth when the key is on. If the pump runs, then the relay is fine, and all the power up to both inputs on the relay is fine. If it doesn't run (and given that you checked the relay itself actually works) then one or both of AF70 and AF71 are not bringing power to the game.
×
×
  • Create New...