The Flying Kiwi (another build thread)

[Uniuno]

Nice work! Best build thread ever! It appeased my instant gratification need like no other! Never would have seen the triple turbo setup coming.

[AdamVR4]

The hot side makes sense. The part of this that makes my head hurt is feeding the compressor outlet of the T04 into the compressor inlets of the 9B's.

Any concern about over-spinning the 9B's? As you increase the system's overall boost, won't the 9B pressure ratio stay pegged? Having a hard time wrapping my head around what each compressor wheel will be doing. With different waste gates controlling the T04 and the 9B's, can you say with any certainty that ALL 3 of them will ever actually open? It really seems like you would want 2 EBC's.

[anyonebutme]

The hot side makes sense. The part of this that makes my head hurt is feeding the compressor outlet of the T04 into the compressor inlets of the 9B's.

Any concern about over-spinning the 9B's? As you increase the system's overall boost, won't the 9B pressure ratio stay pegged? Having a hard time wrapping my head around what each compressor wheel will be doing. With different waste gates controlling the T04 and the 9B's, can you say with any certainty that ALL 3 of them will ever actually open? It really seems like you would want 2 EBC's.

If you want I can send you a lot of info about compounding to get your head around it.

[Roybatty]

Love lotr film setting, Kimbra & your builds =)

Sent from garage

[Blackmount]

Subd. First awd n/a gto ive seen in the build threads. Excited for dyno results

[AdamVR4]

If you want I can send you a lot of info about compounding to get your head around it.

Can't post the highlights here?

[anyonebutme]

Can't post the highlights here?

This is going to be difficult without derailing the thread, so I'll be brief here. Charles seems to explain it better than me in typing so I'll just use his words.

"They don't work any differently than a single. You just use one compressor to alter the incoming "atmosphere" of the second. But neither charger actually "knows" it's in a compound setup. The first stage just thinks it's putting light boost into a large engine. The second stage just thinks it's putting light boost...at a few thousand feet below sea level."

You're not compounding pressure, but pressure ratios, which is why your end boost is not boost1+boost2, but P/R1*P/R2 (IIRC, I forgot if it's addition or product). That's how you can get 100+ psi out of 2 turbos huffing well under 40 psi. Gates control the speed of the turbo, therefore the boost of each stage. Since we are talking gas engines, I'll leave out discussion of "drive pressure", which is the pressure in the exhaust manifolds.

Yes your gates for the primary and secondary will have to be controlled independently, but an EBC is overkill for this application for either stage. You set the "high pressure" turbo first to it's optimum operating PR, for a 9b would be something like 5-8 psi, then add pressure from the "atmosphere" turbo until you reach it's optimum or the boost capability of the engine. It will be the latter for our engines. If your turbine flows enough (turbo size is nearly dead on) and your gates are large enough, you can actually do this with internal gates and/or no boost controller at all.

Have questions, just ask.

[J. Fast]

This is going to be difficult without derailing the thread, so I'll be brief here. Charles seems to explain it better than me in typing so I'll just use his words.

"They don't work any differently than a single. You just use one compressor to alter the incoming "atmosphere" of the second. But neither charger actually "knows" it's in a compound setup. The first stage just thinks it's putting light boost into a large engine. The second stage just thinks it's putting light boost...at a few thousand feet below sea level."

You're not compounding pressure, but pressure ratios, which is why your end boost is not boost1+boost2, but P/R1*P/R2 (IIRC, I forgot if it's addition or product). That's how you can get 100+ psi out of 2 turbos huffing well under 40 psi. Gates control the speed of the turbo, therefore the boost of each stage. Since we are talking gas engines, I'll leave out discussion of "drive pressure", which is the pressure in the exhaust manifolds.

Yes your gates for the primary and secondary will have to be controlled independently, but an EBC is overkill for this application for either stage. You set the "high pressure" turbo first to it's optimum operating PR, for a 9b would be something like 5-8 psi, then add pressure from the "atmosphere" turbo until you reach it's optimum or the boost capability of the engine. It will be the latter for our engines. If your turbine flows enough (turbo size is nearly dead on) and your gates are large enough, you can actually do this with internal gates and/or no boost controller at all.

Have questions, just ask.

I'll add to what Chris said Adam.

In early 2010 I was considering fitting a compound turbo. I consulted with a bunch of diesel guys at Turbo Forums since coumpounding is quite prevalent with them. What I quickly discovered was it was quite cost prohibitive and the piping was very difficult to fit. The biggest prohibitive factor was heat. The heat produced in a compound setup was thermodynamically proven to be extraordinarily high. It was going to to be quite difficult to manage because Air to Air intercooling capability is significantly lower than I'd initially estimated. After doing some planning I ditched the project, sold my exhaust flanges off to Frostedbutts, and decided to keep things simple (like Ninja mentioned). That's when I heeded some good advice for once and swapped to the KISS method. I chose to install an IPS Evo III kit instead.

I was collaborating with a guy on Turbo Forums named RyanMayo that had recently fitted a Nissan 300zx with a compound turbo setup. He helped me size turbo's for a 1000bhp engine build and confirmed my math to be correct.

Here's an excerpt from that thread for the information you seek:

The initial starting point is where I'm hung up... I'll try and do the math and hopefully you guys can double check it for me and then once it can be confirmed it's right we'll go from there. So to do this I ineed to calc the volumetric airflow requirement, the stage 1 turbocharger compressor outlet temperature, the stage 1 density ratio, the stage 2 compressor outlet temperature, the stage 2 density ratio, the combined density ratio, and the intercooler density change?

Volumetric airflow(cfm) = RPM (X) Displacement (X) .5 (X) (volumetric efficiency/1728)

Absolute compressor outlet temp = Compressor inlet temp (*R) + (((Compressor inlet temp (x) Pressure Ratio) - Compressor inlet temp) / Compressor Efficiency)

Density ratio = (Inlet absolute air temp(*R) / Outlet temperature(*R)) (X) Pressure Ratio

Combined density ratio = 1st stage density ratio (X) 2nd stage density ratio

Intercooler density change = (Stage 2 compressor outlet / Stage 1 compressor outlet) - 1

Used my thermodynamics book for this so hopefully this equation setup is right.

Jeremy

I'll reiterate again that I'm not sure I know as much about this as I think I know, so I may be off and take what I write as an opinion. It's certainly not a definitive lesson on the topic.

So how I would start is the same way you have: by figuring the cfm of the engine. I'll just go ahead and ASSume 7000rpm and 90%ve for that, which equals 345cfm.

Now the question is what's the cfm I need to make 1000hp? I figure on 1lb/min equaling 10hp, so that's 100lbs/min. And 1lb/min equals 14.47cfm at 1 atmosphere (14.7psia) so we want to flow 1,447cfm to make 1,000hp.

1,447/345=4.19 round to 4.2
take the square rt of 4.2 = ~2.05 <-that's the pressure ratio I want each turbo to hit.

To find the flow rate needed for the second stage, I would just use this calculator: Not2Fast: Turbo Calculator

After plugging in all the engine variables and using 15.4psi, (roughly equal to 2.05pr,) it tells me the small turbo needs to draw 655cfm, or the equivalent of 49lb/min, if it were drawing from atmosphere. Although it's not drawing from atmosphere, I think it should work when looking at compressor maps.

So, first stage turbo should flow 100lb/min of air, and second stage should flow 49lb/min of air when looking at compressor maps. Both should be just above the 2:1pr line. Looking at Garrett maps, a GT4718 and a GT3582 would work well. As for the turbines, I don't even know where to start haha.

None of this takes into account the effect that heat has on the density of the charge, but there's no question that a GT4718 can make the power, and the 3582 would be a decent match to the motor as well. So if you were to use these turbos I'd plan on maybe having to gun a little more boost than what I've figured here. If I were to change anything I'd maybe go a little larger on the secondary turbo, but not by much.

Hope this helps you, and I look forward to seeing what you do with this. It's really interesting and I think I probably stand to learn a thing or two.

Haha that calculator is funny... I have this one saved in my fav's tab: Index page :: motorgeek.com

Oky so I Math'd this out, posting it would be a nightmare so maybe I'll scan my chicken scratch. I'm looking at some serious exponential factors.

Givens: psia elevation factor in Colorado 12.5, 6000rpm, 70* ambient temp, 90%V.E, 189 CID, 295cfm airflow(22.4lb per min)

Math I came up with:

Final discharge temp with heating effect: 481*F (no interstage intercooler)
Final intake temp with 80% intercooler: 96*F
Stage 1 Compressor flow demand: 2.2 Pressure Ratio, 37.4lb per min
Stage 2 Compressor flow demand: 2.2 Pressure Ratio, 62.5lb per min

Final pressure ratio: 6.1 = 76psia

Boost pressure 15 psi

I'll double check this again... the numbers are makin me headspin.

As you can see 2.2P.R. being fed into a 2.2P.R. at 15psi has a final net psia of 76psia. That's effectively 1400CFM at 15psi with a gt 47feeding a 35R... crazy! So I'm thinking about what box has going on... I haven't done math like that in a couple years (and I don't even know for sure if I setup the equations right) but I'll try to run a calc at 10psi for the 9b twins and the T04 and see what the final estimated psia comes out to. I'm guessing at 12psi its likely to have an equivalent CFM of somewhere around 30psi. Let me run some numbers.

To answer your question of what's got you hung up... Yes, the 9b's are a restiction... but they're also dividing the final P.R. back out. That's what makes this setup so cool. It's really difficult to calculate airflow. Hence why I was asking box for his AIT and kpa numbers.

Do you have that data, box? yYur AIT's and your GM Mas Hz numbers... or Kpa figures?

[box]

No numbers yet, will get some as soon as I start boosting

[AdamVR4]

This is going to be difficult without derailing the thread, so I'll be brief here. Charles seems to explain it better than me in typing so I'll just use his words.

"They don't work any differently than a single. You just use one compressor to alter the incoming "atmosphere" of the second. But neither charger actually "knows" it's in a compound setup. The first stage just thinks it's putting light boost into a large engine. The second stage just thinks it's putting light boost...at a few thousand feet below sea level."

You're not compounding pressure, but pressure ratios, which is why your end boost is not boost1+boost2, but P/R1*P/R2 (IIRC, I forgot if it's addition or product). That's how you can get 100+ psi out of 2 turbos huffing well under 40 psi. Gates control the speed of the turbo, therefore the boost of each stage. Since we are talking gas engines, I'll leave out discussion of "drive pressure", which is the pressure in the exhaust manifolds.

Yes your gates for the primary and secondary will have to be controlled independently, but an EBC is overkill for this application for either stage. You set the "high pressure" turbo first to it's optimum operating PR, for a 9b would be something like 5-8 psi, then add pressure from the "atmosphere" turbo until you reach it's optimum or the boost capability of the engine. It will be the latter for our engines. If your turbine flows enough (turbo size is nearly dead on) and your gates are large enough, you can actually do this with internal gates and/or no boost controller at all.

Have questions, just ask.

Ah, so it would be quite close to perfect if you ran the 9B's at waste gate boost and the boost controller just on the T04.