Posible divided scroll turbine housing for GT35R right off Garrett's shelfs?
01-17-05, 11:55 PM
#1
Full Member
Thread Starter
Posible divided scroll turbine housing for GT35R right off Garrett's shelfs?
This has really been bugging my mind for a while:
The Garrett GT35R ball bearing turbos have the same size/trim turbine specs as their GT35 non-BB turbo. But this later unit comes with divided scroll turbine housings on 1.18 (standard) and 1.05 (optional part# 714690-13) A/Rs. So, my questions are: Do the turbines on these two different turbos have the same shape? (Size and trim don't tell you anything about the turbine's actual profile, and these are very different turbos: BB shafts don't fit on plain bearing centers nor vice-versa.) If they are, can the GT35's divided scroll turbine housings be on some way adapted on to the GT35R by machining or making an adapter plate or a combination of both, etc? If they are not, might it be posible to take an un-machined, raw casting GT35 divided scroll turbine and machine it to fit the GT35R? To better understand what's bugging my mind, please check the specs on both GT35 and GT35R turbos by downloading Garrett's catalog from here:
http://www.turbobygarrett.com/turbob.../downloads.jsp
Why all the fuss? Mazda engineers eons ago (back when they were hacking more power out of the Turbo II, read SAE paper# 900036) found out that a divided scroll turbine housing is the best way to make a single turbo setup spool faster and breathe better at high rpm. How's that hack done? A divided scroll isolates the exhaust pulses from each rotor, so that when one port opens its "pulse" doesn't get lost feeding into both the turbine housing and the other port which has its rotor face on the middle of its exhaust stroke. This has the double bonus of:
a. Concentrating the full power of the exhaust "pulse" (the loud bang so unique to a rotary's exhaust port and the source of, perhaps, 95% of the energy that moves the turbine) on pushing half the width of turbine blades (each port will be exhausting through half the overall A/R of the housing, instead of the full A/R on a single scroll housing like the GT35R's), which is great for making the turbo come on boost sooner at a much lower rpm, and
b. Reducing backpressure by not allowing this "pulse" to push back the exhaust gases into the other rotor which is trying to push its remaining exhaust gases out, and therefore improving the overall exhaust efficiency of the turbo.
What gives? If this divided scroll thing is so good, then why Garrett doesn't make them available in any of their small frame BB GT turbos (nor does IHI [Apex'i] for that matter)? Are they not good for BBs? Do they do more harm than good on BBs?
Can someone please aproach a Garrett representative with this thoughts? It is nearly imposible for me to do it, given that I don't have any contact information of nobody there and I've seen that many of you guys know and work directly with some of them. I really love the GT35R, and I plan on using one on my FD3S, but I will really like to squeeze every bit of juice it can give, and the single scroll housings available for it are a hindrance to its giving out its full capability.
The Garrett GT35R ball bearing turbos have the same size/trim turbine specs as their GT35 non-BB turbo. But this later unit comes with divided scroll turbine housings on 1.18 (standard) and 1.05 (optional part# 714690-13) A/Rs. So, my questions are: Do the turbines on these two different turbos have the same shape? (Size and trim don't tell you anything about the turbine's actual profile, and these are very different turbos: BB shafts don't fit on plain bearing centers nor vice-versa.) If they are, can the GT35's divided scroll turbine housings be on some way adapted on to the GT35R by machining or making an adapter plate or a combination of both, etc? If they are not, might it be posible to take an un-machined, raw casting GT35 divided scroll turbine and machine it to fit the GT35R? To better understand what's bugging my mind, please check the specs on both GT35 and GT35R turbos by downloading Garrett's catalog from here:
http://www.turbobygarrett.com/turbob.../downloads.jsp
Why all the fuss? Mazda engineers eons ago (back when they were hacking more power out of the Turbo II, read SAE paper# 900036) found out that a divided scroll turbine housing is the best way to make a single turbo setup spool faster and breathe better at high rpm. How's that hack done? A divided scroll isolates the exhaust pulses from each rotor, so that when one port opens its "pulse" doesn't get lost feeding into both the turbine housing and the other port which has its rotor face on the middle of its exhaust stroke. This has the double bonus of:
a. Concentrating the full power of the exhaust "pulse" (the loud bang so unique to a rotary's exhaust port and the source of, perhaps, 95% of the energy that moves the turbine) on pushing half the width of turbine blades (each port will be exhausting through half the overall A/R of the housing, instead of the full A/R on a single scroll housing like the GT35R's), which is great for making the turbo come on boost sooner at a much lower rpm, and
b. Reducing backpressure by not allowing this "pulse" to push back the exhaust gases into the other rotor which is trying to push its remaining exhaust gases out, and therefore improving the overall exhaust efficiency of the turbo.
What gives? If this divided scroll thing is so good, then why Garrett doesn't make them available in any of their small frame BB GT turbos (nor does IHI [Apex'i] for that matter)? Are they not good for BBs? Do they do more harm than good on BBs?
Can someone please aproach a Garrett representative with this thoughts? It is nearly imposible for me to do it, given that I don't have any contact information of nobody there and I've seen that many of you guys know and work directly with some of them. I really love the GT35R, and I plan on using one on my FD3S, but I will really like to squeeze every bit of juice it can give, and the single scroll housings available for it are a hindrance to its giving out its full capability.
Last edited by EFINI_RX-7_RZ; 01-18-05 at 12:01 AM.
01-18-05, 09:56 AM
#3
It will not work due to different center sections I can't totally remember but it has something to do with not enough material to swap the cartridge. I have talked to garrett a lot on this. I have been in the process for quite awhile on getting them to make me something very similar to what your talking about.
01-18-05, 11:18 AM
#4
Full Member
Thread Starter
Zero R, I just saw stevecod's pictures of the manifold you made for him. Will this work well with a GT35R with a divided scroll turbine housing? Will you be replicating this design for future sale on your kits? It sure looks like a much thought through design than you current offerings.
Back to the topic on hand, guys like ATP Turbo have made different turbine housings for the GT35R but no one has come up with a divided scroll design, that's why I'm asking here if a divided scroll will harm the BBs, since not even IHI/Apex'i have one for their small frame BB cartridges (like on their RX6). So Zero R, from your comments deduce that the turbines on the GT35 and GT35R have exactly the same design/profile. If so, I'm sure Garrett might be able to give you rough GT35 housing castings so that you can machine them to fit the GT35R. Or you can use it as a template and have it re-molded to make a new casting design that will fit the GT35R, then patent the design and make serious $$$ out of it. Gosh, why Garrett doesn't offer them right off the bat? And speaking of which, why Garrett chose to give the GT30R an anti-surge compressor housing and not make the one on the GT35R likewise? Just look how fat is the GT30R's map. I'm sure the GT35R's map would also benefit from an anti-surge compressor housing, and this will help prevent some members here from surging their GT35Rs. So Zero R, you be the man, you get them Garrett guys make us a perfect GT35R turbo, complete with divided scroll turbine housing and anti-surge compressor housing, not the 3/4 of a full eficiency GT35R they are currently selling (which by the way is a more than 5 years old design), you show them the path to enlightning!
Back to the topic on hand, guys like ATP Turbo have made different turbine housings for the GT35R but no one has come up with a divided scroll design, that's why I'm asking here if a divided scroll will harm the BBs, since not even IHI/Apex'i have one for their small frame BB cartridges (like on their RX6). So Zero R, from your comments deduce that the turbines on the GT35 and GT35R have exactly the same design/profile. If so, I'm sure Garrett might be able to give you rough GT35 housing castings so that you can machine them to fit the GT35R. Or you can use it as a template and have it re-molded to make a new casting design that will fit the GT35R, then patent the design and make serious $$$ out of it. Gosh, why Garrett doesn't offer them right off the bat? And speaking of which, why Garrett chose to give the GT30R an anti-surge compressor housing and not make the one on the GT35R likewise? Just look how fat is the GT30R's map. I'm sure the GT35R's map would also benefit from an anti-surge compressor housing, and this will help prevent some members here from surging their GT35Rs. So Zero R, you be the man, you get them Garrett guys make us a perfect GT35R turbo, complete with divided scroll turbine housing and anti-surge compressor housing, not the 3/4 of a full eficiency GT35R they are currently selling (which by the way is a more than 5 years old design), you show them the path to enlightning!
01-18-05, 12:31 PM
#5
Senior Member
Join Date: Apr 2001
Location: Seattle, WA
Posts: 528
Likes: 0
Received 0 Likes
on
0 Posts
Divided turbine housings are noteworthy for small displacement motors, but there are tradeoffs that have to be considered. I think there's a little bit of an advantage for this application (if done properly), but it wouldn't be huge if we stay with the T3-based housing. EFINI_RX-7_RZ clearly states the advantages, but we must be careful to consider:
The effect of rotary EGTs
turbine housing choice and turbine efficiency
compressor
That is, simply adding a divided housing to, say, a 1.06 exhaust GT35R MAY give you a boost threshold that is similar to a .80 housing, but you're not going to get more boost than the compressor will allow (also consider that the .80 T-3 is a pretty small exhaust housing) and will still be limited by turbine choice. The compressor maps indicate that 14 PSI can be attained around 3,300-3,400 RPMS, but a divided housing won't do better than that, unless you want to run your turbo in the surge area.
The 35R's turbine and housing appears to be a suboptimal choice for rotary exhaust. I'm not saying that is sucks (hell, I own one), but it COULD be better. Look at the 35R turbine efficiency maps and correct for 1400 degree EGTs. This, IMO, is the primary reason a T04S will ultimately make more power. Adding a divided housing to the current setup will still limit HP potential.
A good solution would be a large, divided 1.00 or so exhaust housing with a similar hot side as a T04S. This should provide threshold characteristics of the current .80 35R with the power potential of a T04S or 1.06 35R+.
Gene
The effect of rotary EGTs
turbine housing choice and turbine efficiency
compressor
That is, simply adding a divided housing to, say, a 1.06 exhaust GT35R MAY give you a boost threshold that is similar to a .80 housing, but you're not going to get more boost than the compressor will allow (also consider that the .80 T-3 is a pretty small exhaust housing) and will still be limited by turbine choice. The compressor maps indicate that 14 PSI can be attained around 3,300-3,400 RPMS, but a divided housing won't do better than that, unless you want to run your turbo in the surge area.
The 35R's turbine and housing appears to be a suboptimal choice for rotary exhaust. I'm not saying that is sucks (hell, I own one), but it COULD be better. Look at the 35R turbine efficiency maps and correct for 1400 degree EGTs. This, IMO, is the primary reason a T04S will ultimately make more power. Adding a divided housing to the current setup will still limit HP potential.
A good solution would be a large, divided 1.00 or so exhaust housing with a similar hot side as a T04S. This should provide threshold characteristics of the current .80 35R with the power potential of a T04S or 1.06 35R+.
Gene
01-18-05, 12:34 PM
#7
Remember that Mazda also worked w/ Hitachi to develop the special "impact" style turbine blade that takes advantage of the strong individual exhaust pulses of the rotary by having a flat profile front and back instead of the typical cupped rear and open front of piston exhaust wheels.
Perhaps this impact wheel works best w/ the individual hammering of a fully divided housing.
You have chosen the new GTQ exhaust wheel of the GT line- I am willing to bet it benifits most from a smooth collection of pulses like if you had a long tuned length manifold collected at the turbo flange.
Mazda also designed the dual wastegate S5 turbo so that it is truely divided from exhaust port to exhaust wheel (when spooling w/ wastegate open.
You could simulate this if you ran a wastegate per exhaust runner, but it is unknown if the GTQ likes the hammering anyways.
On my S5 hybrid w/ 60mm HKS GT wastegate on the backside I ran it so that there was exhaust port communication at the wastegate and then later added a divider all the way up to the wastegate valve and I only noticed the improved spool at the VERY low end (better 800-2,000rpm spool, but nowhere else). Sure I could pull smoothly from 25mph in 5th, but it didn't help in performance driving at all.
Perhaps this impact wheel works best w/ the individual hammering of a fully divided housing.
You have chosen the new GTQ exhaust wheel of the GT line- I am willing to bet it benifits most from a smooth collection of pulses like if you had a long tuned length manifold collected at the turbo flange.
Mazda also designed the dual wastegate S5 turbo so that it is truely divided from exhaust port to exhaust wheel (when spooling w/ wastegate open.
You could simulate this if you ran a wastegate per exhaust runner, but it is unknown if the GTQ likes the hammering anyways.
On my S5 hybrid w/ 60mm HKS GT wastegate on the backside I ran it so that there was exhaust port communication at the wastegate and then later added a divider all the way up to the wastegate valve and I only noticed the improved spool at the VERY low end (better 800-2,000rpm spool, but nowhere else). Sure I could pull smoothly from 25mph in 5th, but it didn't help in performance driving at all.
Trending Topics
01-18-05, 02:29 PM
#9
To put things into perspective concerning my set-up of fully divided VS divided w/ pulse communication between exhaust runners.
It made a MUCH larger impact on turbo spool when I had to go from 3 1/2" exhaust down to 3" exhaust to stop my boost creep.
Datalogs showed over a 500rpm loss in boost response when I downsized to 3" and it was worst at the lowest rpms. Full boost was "only" 500-800rpm later, but boost threshould was about 500rpm later as well- so instead of 1-4psi boost at say 1,500rpm it was now 2,000rpm now. Made the engine feel alot smaller and couldn't pull smoothly in 5th much below 1,500rpm anymore. Of course, I didn't HAVE to since it was so quiet now...
So, perhaps you can put a small part of the effort it would take to make a divided GT35R turbo housing and put it into fitting the lowest backpressure exhaust you can under your car and have greater gains?
It made a MUCH larger impact on turbo spool when I had to go from 3 1/2" exhaust down to 3" exhaust to stop my boost creep.
Datalogs showed over a 500rpm loss in boost response when I downsized to 3" and it was worst at the lowest rpms. Full boost was "only" 500-800rpm later, but boost threshould was about 500rpm later as well- so instead of 1-4psi boost at say 1,500rpm it was now 2,000rpm now. Made the engine feel alot smaller and couldn't pull smoothly in 5th much below 1,500rpm anymore. Of course, I didn't HAVE to since it was so quiet now...
So, perhaps you can put a small part of the effort it would take to make a divided GT35R turbo housing and put it into fitting the lowest backpressure exhaust you can under your car and have greater gains?
01-18-05, 04:24 PM
#10
Corn-to-Noise Converter
iTrader: (6)
Join Date: Mar 2002
Location: The Elysian Fields (Texas)
Posts: 1,527
Received 386 Likes
on
154 Posts
I continue to like the way that you all think. I'm considering being an early adopter to just such a setup, as I discussed most of these points with Sean back in December.
So how's about it Sean, do you have your first customer lined up?! ;-) Just on Gene's recommendation alone, your products have become the lead candidate for my turbo upgrade.
So how's about it Sean, do you have your first customer lined up?! ;-) Just on Gene's recommendation alone, your products have become the lead candidate for my turbo upgrade.
