Let's talk about doing a triple turbo setup
03-06-2008, 08:45 PM
#31
Registered User
I don't want to belabor the obvious, but since I didn't notice it mentioned yet - even though the drive pressure will essentially remain the same (not counting efficiency differences between turbocharging schemes) - the drive energy to each turbo in divided secondaries will be halved, so those tiny turbos may not spool as quickly as you'd think...
Ideally, the exhaust header tubes should converge on a concentrically located cone/flange into the turbine housing; this would also allow for equal-length tubes, and 'charger packaging constraints could be dealt with by altering the exit angle of the header flange.
If I read the concerns about hanging turbos off the head in relation to stud strength - you could lift the whole truck by the head studs w/o exceeding their tensile capacity!
Ideally, the exhaust header tubes should converge on a concentrically located cone/flange into the turbine housing; this would also allow for equal-length tubes, and 'charger packaging constraints could be dealt with by altering the exit angle of the header flange.
If I read the concerns about hanging turbos off the head in relation to stud strength - you could lift the whole truck by the head studs w/o exceeding their tensile capacity!
03-06-2008, 09:38 PM
#32
Registered User
Thread Starter
Join Date: Apr 2003
Location: Cummins Technical Center, IN
Posts: 6,564
Likes: 0
Received 6 Likes
on
5 Posts
I think the stud question was manifold studs
-- my manifold is held on with bolts.
Yes, I accounted for the drive ENERGY in the spoolup comparisons. This is where the turbine maps are useful, and why I choose Garrett for these little exercises.
Looking at the turbine map for the Stg 3 GT3788, we can see that at a PR of 2, the turbine map shows 30lb/min corrected gas flow. For the stg 2 with the tighter (.89 vs .99 A/R) housing, the map shows about 28lb/min at a PR of 2.
See here:
Now, one would be initially tempted to say that to preserve the same spoolup with two smaller chargers, just cut those numbers in half; i.e., use two turbos with 15# specs at a PR of 2 to equal the stg 3 or use 14# turbos to equal the tighter stg 2.
That's true from a pure drive ENERGY standpoint. This is boost THRESHOLD, or where the engine will start to make boost in absolute terms. Still, with reduced heat loss, the smaller chargers should make more of that energy available to the turbines.
Most importantly, once they are at their threshold, they can response virtually instantly to any change in drive energy.
Imagine driving the Smarty on #9 and flooring it and seeing almost no smoke-- instead, that smoke was converted to power by the nearly lagless response of the secondaries.
Now, since the secondaries are responding so fast, what would that do to your ability to light the primary? Oh yeah! The sooner the secondaries light, the sooner the primary lights!
In keeping with my theories of twins (the primary should flow double the flow of the secondaries at a PR of 3), that means a 90lb/min primary only needs 22.5lb/min from each of the small secondaries (22.5x2x2=90).
Thus, for a turbo like a GT4202 as a primary, we could choose as secondaries a turbo like the GT2252, with its tiny little 40.2mm inducer.
The compressor map looks like this:
So we can see that 22.5lb/min is good for this compressor.
Most importantly, the turbine map looks like this:
At a PR of 2:1, this thing is only 15lb/min, so spooling it will be very easy and it will spool early.
Still those might be a little TOO small, and you'd probably want to consider a charger with specs more like the GT2560R shown here:
http://www.turbobygarrett.com/turbob...R_707160_9.htm
The GT2560 is still only 15#/min at PR of 2.0, so you're still getting early spoolup, even in a larger charger than can support a 120lb/min primary like something in the GT4508 class (80mm inducer range).
There are a lot of options on the smaller end. I personally like the GT22s because they are cheap (<$700) journal bearing units that are small (<7" long) and still pretty efficient.
jmo
-- my manifold is held on with bolts.Yes, I accounted for the drive ENERGY in the spoolup comparisons. This is where the turbine maps are useful, and why I choose Garrett for these little exercises.
Looking at the turbine map for the Stg 3 GT3788, we can see that at a PR of 2, the turbine map shows 30lb/min corrected gas flow. For the stg 2 with the tighter (.89 vs .99 A/R) housing, the map shows about 28lb/min at a PR of 2.
See here:
Now, one would be initially tempted to say that to preserve the same spoolup with two smaller chargers, just cut those numbers in half; i.e., use two turbos with 15# specs at a PR of 2 to equal the stg 3 or use 14# turbos to equal the tighter stg 2.
That's true from a pure drive ENERGY standpoint. This is boost THRESHOLD, or where the engine will start to make boost in absolute terms. Still, with reduced heat loss, the smaller chargers should make more of that energy available to the turbines.
Most importantly, once they are at their threshold, they can response virtually instantly to any change in drive energy.
Imagine driving the Smarty on #9 and flooring it and seeing almost no smoke-- instead, that smoke was converted to power by the nearly lagless response of the secondaries.
Now, since the secondaries are responding so fast, what would that do to your ability to light the primary? Oh yeah! The sooner the secondaries light, the sooner the primary lights!
In keeping with my theories of twins (the primary should flow double the flow of the secondaries at a PR of 3), that means a 90lb/min primary only needs 22.5lb/min from each of the small secondaries (22.5x2x2=90).
Thus, for a turbo like a GT4202 as a primary, we could choose as secondaries a turbo like the GT2252, with its tiny little 40.2mm inducer.
The compressor map looks like this:
So we can see that 22.5lb/min is good for this compressor.
Most importantly, the turbine map looks like this:
At a PR of 2:1, this thing is only 15lb/min, so spooling it will be very easy and it will spool early.
Still those might be a little TOO small, and you'd probably want to consider a charger with specs more like the GT2560R shown here:
http://www.turbobygarrett.com/turbob...R_707160_9.htm
The GT2560 is still only 15#/min at PR of 2.0, so you're still getting early spoolup, even in a larger charger than can support a 120lb/min primary like something in the GT4508 class (80mm inducer range).
There are a lot of options on the smaller end. I personally like the GT22s because they are cheap (<$700) journal bearing units that are small (<7" long) and still pretty efficient.
jmo
03-06-2008, 10:14 PM
#33
Registered User
Join Date: Nov 2006
Location: Rochester, NY
Posts: 2,333
Likes: 0
Received 0 Likes
on
0 Posts
I know I know, it sucked. Tomorrow morning I'll get started on something realistic. That was just for a visual aid.
03-06-2008, 11:06 PM
#34
Registered User
I think the drawing was fine - just suggesting an improved design.
I needed to point out the drive pressure/energy qualification in relation to the turbos' moving parts, since the lunch isn't totally free; but it's still a win/win if you don't count cost.
For example, a 64mm charger has about 25% more surface area (and analogously, a corresponding weight increase) than a pair of 40mm units...
03-07-2008, 06:31 AM
#35
Registered User
Join Date: Nov 2006
Location: Rochester, NY
Posts: 2,333
Likes: 0
Received 0 Likes
on
0 Posts
OK - this is where I was led astray... ^^^
I think the drawing was fine - just suggesting an improved design.
I needed to point out the drive pressure/energy qualification in relation to the turbos' moving parts, since the lunch isn't totally free; but it's still a win/win if you don't count cost.
For example, a 64mm charger has about 25% more surface area (and analogously, a corresponding weight increase) than a pair of 40mm units...
I think the drawing was fine - just suggesting an improved design.
I needed to point out the drive pressure/energy qualification in relation to the turbos' moving parts, since the lunch isn't totally free; but it's still a win/win if you don't count cost.
For example, a 64mm charger has about 25% more surface area (and analogously, a corresponding weight increase) than a pair of 40mm units...
Starting a real model now, btw.
03-07-2008, 08:14 AM
#36
Registered User
Join Date: Nov 2004
Location: corpus christi texas
Posts: 491
Likes: 0
Received 0 Likes
on
0 Posts
you could get 2 manifolds, possibly ats 3-piece manifolds, put two turbo pieces together toward the back and the front one be an end piece. it should fit. or you could weld 2 stock manifolds together and cap the ends off if you want the turbos more toward the middle. but then your oil filter may get in the way and the welds may break. this is why i say toward the back. just an idea to think about.
or get 3 manifolds and have triplets with a big boy on bottom
or get 3 manifolds and have triplets with a big boy on bottom
03-07-2008, 02:05 PM
#41
Registered User
Join Date: Aug 2005
Location: Airdrie, Alberta
Posts: 7,780
Likes: 0
Received 0 Likes
on
0 Posts
you could get 2 manifolds, possibly ats 3-piece manifolds, put two turbo pieces together toward the back and the front one be an end piece. it should fit. or you could weld 2 stock manifolds together and cap the ends off if you want the turbos more toward the middle. but then your oil filter may get in the way and the welds may break. this is why i say toward the back. just an idea to think about.
or get 3 manifolds and have triplets with a big boy on bottom
or get 3 manifolds and have triplets with a big boy on bottom
Problem with your ATS idea is that the center sections have female flanges, end pieces have male flange. So you can't run two center sections beside each other. The other problem, if the first one wasn't an issue, is that if you had a a center section in the front, middle, and an end piece at the back, is that the rear turbo would be getting 3 cylinders all to it self, and sharing 2, while the front is sharing 2 cylinders, and only gets 1 to itself. So the rear turbo would be getting twice as much exhaust energy as the front one.
03-11-2008, 11:11 AM
#45
Registered User
Bracing of the manifold seems like it would be an issue. If what I'm visualizing is correct, each manifold (which only bolts to 3 of the cylinders, correct?) is being asked to hold two turbos. Some serious bracing would be required, otherwise I would venture a guess to say we could actually rip the studs/bolts out of the head.
I say that, because I'm not sure if we use studs or bolts on our trucks. Haven't had the pleasure of dropping one.....yet.
I say that, because I'm not sure if we use studs or bolts on our trucks. Haven't had the pleasure of dropping one.....yet.
