Garrett Medium Frame Gtx Turbos

[jet_r31]

On 3/31/2017 at 9:30 AM, Lithium said:

Nice!  What changes does it have from when it had the 3586?   I wonder if it was hotside or coldside on the HTA which held it up.

The hotsides makes lot more with the 66mm upgrades..

Guessing its that unless,  the extra compression matt runs upsets things a lot

[Lithium]

17 hours ago, jet_r31 said:

The hotsides makes lot more with the 66mm upgrades..

Guessing its that unless,  the extra compression matt runs upsets things a lot

Yeah, that was my gut feel too - but quite interesting, as FP really pushed the .85 T3 housing they put on those things as being capable of flowing a heap.  The bigger hotsides on the GT35 turbines seem to really give the boost threshold a major hit as well, I remember thinking the first time I came across a car with a 1.06 T3 GT3582R that the lag it had (like ~5000rpm on an RB25 from memory) you'd may as well go to a larger frame turbo - but hopefully the Gen2 beast offers enough flow to justify whatever lag it introduces.

[jet_r31]

the 1.06 vs .82 on gt3582 deal is really not worth it. the 1.06 is really for 3.5l/4.0l motors to kill backpressure at same hp as the smaller motors

On a 2.5l to 3.0l there is a little bit of power gain on kill, but absolutely  f**kloads of lag..you might as well go up in turbos size as you will get the same lag..and heaps

more power...

Edited April 3, 2017 by jet_r31

[jet_r31]

7 hours ago, Lithium said:

Yeah, that was my gut feel too - but quite interesting, as FP really pushed the .85 T3 housing they put on those things as being capable of flowing a heap.  The bigger hotsides on the GT35 turbines seem to really give the boost threshold a major hit as well, I remember thinking the first time I came across a car with a 1.06 T3 GT3582R that the lag it had (like ~5000rpm on an RB25 from memory) you'd may as well go to a larger frame turbo - but hopefully the Gen2 beast offers enough flow to justify whatever lag it introduces.

Yeh im not sure, ive actually written my wording wrong...

I mean theres lots  more flow with the 66mm compressor upgrades, the turbine backpessure woudnt be pretty, but theres probally another 60+rwkw left(over a standard 3582 comp) of the turbine before it gives up.

Theres plenty of information around on this even via xr6 upgrades and vl turbos if you look, but i never take to much notice as its not the right way to build a turbo

but i know that those combos dont work on 98 oct and need e85 even on the xr6's with 4.0 of motor s to be a viable option

 

The problem with designing a turbo that way is..yeh it makes power..but you have a comp/turbine mismatch which does a few things...

Cant take timing all that well

doesnt have as wide powerband as it should 

has a very sharp power curve onto boost that hits like a sledgehammer and then gets pulled down due to backpressure quicker

turbos dont last as long

That was my initial worry about the gen2 gtx 3582, the powerband would be peaky and not hold up top and need a 65-66mm turbine

Edited April 3, 2017 by jet_r31

[Lithium]

^ I think we're on the same page

[whatsisname]

R33, built RB2630, GTX3584RS 1.01 making 548.9kW at the hubs and has ran 9.6 @ 145.97mph

 

Not sure how much boost nor response etc. Auto too so probably 560-570kW through a 5sp manual

 

 https://m.youtube.com/watch?v=HtwR2p9XZcw

 

 

 

 

 

 

[Lithium]

That's decent!   145mph says a lot.  Any word on when you get yours back/some dyno sheets with rpm on them?

 

[whatsisname]

Shaun is snowed under with work at the moment so unfortunately for me it's hurry up and wait.
Going to ramp boost on harder and add 1-2psi in the mid range but stick to 26psi up top.

[whatsisname]

Posted in the RB30 results thread but figure it probably should go in here given the back story resides here.

 

As per the plan, we added 1psi in the mid range, bringing it up to 27psi. It drops off to 26psi at peak power. Added a touch more timing up top as well.

 

HTA GT3586R 0.85 vs GTX3584RS 1.01

 

350-400rpm drop in response vs. near on 70kW peak gain and reasonable gains from 4200rpm upwards.

 

Roughly 870Nm at 4500rpm. 335Nm at 3000rpm.

 

As a comparison, a mate's VE SS with supercharger running 6psi, mild cam and usual intake and exhaust mod's made 375kW and 660Nm peak the night before mine on the same dyno.

[Lithium]

9 hours ago, whatsisname said:

HTA GT3586R 0.85 vs GTX3584RS 1.01

 

350-400rpm drop in response vs. near on 70kW peak gain and reasonable gains from 4200rpm upwards.

Decent result!  Do you have an rpm scale version of the dyno plot?   It looks a fair bit laggier than the HTA3586, without an rpm scale I'd have guessed that as more than 400rpm - but either way it clearly is capable of much more power so it's hardly a fail being laggier than something that at one point was probably one of, if not THE most responsive turbo capable of 75lb/min.

Any plan to hit the strip with it?  How does it drive?

[SimonR32]

3 hours ago, Lithium said:

Decent result!  Do you have an rpm scale version of the dyno plot?   It looks a fair bit laggier than the HTA3586, without an rpm scale I'd have guessed that as more than 400rpm - but either way it clearly is capable of much more power so it's hardly a fail being laggier than something that at one point was probably one of, if not THE most responsive turbo capable of 75lb/min.

Any plan to hit the strip with it?  How does it drive?

At a guess (4.11s with 265/18s) 112 km/hr is around 4000rpm... So it's gone from 345kw to 230kw which is a massive wack.

I've now stopped look at graphs across the x axis for response and now look more to the difference in y axis as it shows a lot more about how a car will drive

[Lithium]

11 minutes ago, SimonR32 said:

At a guess (4.11s with 265/18s) 112 km/hr is around 4000rpm... So it's gone from 345kw to 230kw which is a massive wack.

I've now stopped look at graphs across the x axis for response and now look more to the difference in y axis as it shows a lot more about how a car will drive

Yeah I've kind of always looked at a mixture of the both trying to build a pic, as you indicate - saying there is 300rpm better spool doesn't sound like much but if you have >100kw gain at something like 4000rpm then you REALLY feel it.  In more recent times however - especially after experiencing the difference between a GT3076R and a T3/T60-1, and then a GT3582R and an EFR8374.... you can end up with dyno plots which seem comparable both on x and y axis and then get into the car and go uhhh, what has changed between these since the dyno?    

Beating a dead horse here, but transient response is something dynos really don't seem strong at demonstrating and I still don't quite know the most sensible metric for that.  Not sure if you can easily do that.  Given that the GT3584RS is a completely different turbo on both sides to any other Garrett unit, I'm really interested to know how it drives - the RS turbine could behave any number of ways... I've kind of wondered why it wasn't used on the Gen2 GTX3582R.

 

[whatsisname]

Hey mate. Shaun always leaves it set on road speed for some reason. He did tell me why at one stage but I can't recall. It is laggier than the HTA3586. No big surprise there, given the 0.85 vs 1.01 housing change and 62mm v 67mm comp wheel.

Running the numbers through a basic speed to RPM calculator (using gear ratio [4th = 1.000], differential ratio [4.111], tyre diameter [641mm or 25.24"], and vehicle speed) it shows that 10km/h equates to roughly 300-400RPM.

HTA peak torque arrives at approximately 120km/h which is just over 4000rpm in 4th

RS peak torque arrives at approximately 130km/h which is just over 4350rpm(4377rpm) in 4th

80km/h is near enough to 2700rpm

90km/h - 3000rpm

100km/h - 3400rpm

110km/h - 3700rpm (112km/h is a smidge over 3800)

The RS overtakes the HTA by approximately 4200rpm. I've got 7800rpm to play with.

I'm picking it up this afternoon, so I'll report back later tonight or over the weekend. The plan is to hit the strip once the TC/LC system is sorted - had to send the system back to the UK for reconfiguration and bench testing.

 

The graphs really don't do either turbo justice. Response with the HTA was exceptional. There was no big ramp up onto boost, it just spooled up and by 2500-3000rpm it was turning the tyres. I'm fine with a shift to the right by 350-400rpm, it means speed will be up and it should be less likely to want to swap ends.  

On the street I don't really care. As long as it can keep up with the traffic I'm happy. I'm old enough to know I don't have to prove anything to anyone. The car is too 'pretty' to see any serious track use. The drag strip is where it will be unleashed in full. However, I'm not aiming for any records, it's just a hobby and a bit of fun.                  

 

 

    

[Lithium]

Here is an interesting comparison, someone in Subaru world went from a 1.03a/r Gen1 GTX3582R to a 1.03a/r Gen2 GTX3584RS with everything else the same spec.  
 

 

Edited July 11, 2017 by Lithium

[blah_blah]

Am i reading it right?

Green is before red is after. So slightly less response but 20% increase in power?

[Lithium]

Yep, that's what I'm seeing

[whatsisname]

Nice find mate [emoji1303]

Can report that the car is great to drive. Still plenty of urge below 4000rpm, more than enough to get in trouble. Above 4000 in 3rd it will do a rolling burnout. TC is back from the UK and hope to have it running soon.

[infomotive]

Good to see good results coming through on the gen 2s.

I have some results to share on the GTW3884 v TO4Z.

Is this an appropriate place or should I start another GTW specific thread?

Thanks

[Lithium]

Hi Jason,

I can't speak for everyone but I suspect this is probably the best place for all things "modern" Garrett really.  I for one would be really interested to see how the GTW3884R compares with a T04Z.  Is it a 67mm GTW?   

Anything else (ie, if its the same exhaust housing a/r etc) would be awesome info too

For Oz folks reference, everything I've seen indicates Infomotive's dyno reads pretty comparable with the Dyno Dynamics dynos commonly used in Oz.  

[infomotive]

2 minutes ago, Lithium said:

Hi Jason,

I can't speak for everyone but I suspect this is probably the best place for all things "modern" Garrett really.  I for one would be really interested to see how the GTW3884R compares with a T04Z.  Is it a 67mm GTW?   

Anything else (ie, if its the same exhaust housing a/r etc) would be awesome info too

For Oz folks reference, everything I've seen indicates Infomotive's dyno reads pretty comparable with the Dyno Dynamics dynos commonly used in Oz.  

Hi mate.

Yes 67mm.

Always wanted to see what a billet wheeled version would do as I knew I was out of compressor.

So first testing I can share is identical bar the GTW compressor and results are as you would think looking at the comp maps.

Unfortunately boost is missing from first GTW runs but will be pretty self explanatory.

Will post once backups etc are done

Cheers