My compound turbo set-up

[Tony20tgen]

Whats up Kevin, Im a fellow follower of u and paul. I have a compound turbo kit project swell. Please check out my build thread. I really learned a lot from u guys and ## thread on yellowbullit has been instrumental in figuring out my kit. I have a T3 57 trim with .48 ar and stg 1 exhaust wheel, and a gt4202 with 1.05 ar split T4. I am going to use two wg tubes off the collector instead of 1 due to collector could not be posted for the 60mm tail wg. As it sits now I only have 1 40mm tube off the collector. Check out my build and hope you can cheer in on our forum.

Tony20tgen Stage 3 build thread - Page 25 - Genesis Forum: GenCoupe Hyundai Genesis Forums

 

[tkelly27]

Agreed. I just was really curious to know what spool times are on a "stock" holset...either hx35 or 40 honestly. I know they are "awesome" with the BEP housing, but I already have an Evo 1 16g...no reason not to take advantage of it and then just bolt on big power By bolt on, I mean...fabricate, build, text, and cross fingers for everything else Thanks again for all your help, both here and on HRCS.

Two of my friends run the big hx35 with twin scroll manifolds with twin scroll 12cm^2 housings and they are both getting full boost closer to 3000rpm than 3500rpm. Peepers can get it to surge brake boosting on the freeway at 2200-2500rpm. The one will be taking back pressure data with bigger turbines, but it doesn't feel down any power from the bigger housings he was running.

If you were going to go through the trouble of fabricating and all you wanted was hx35 power you'd be better off building a TS manifold. If you had to use a smaller turbo, a T25 or 13g would be able to help spool it. A 14b or 16g wouldn't offer a huge amount of spool up increase. I don't have much information on the hx40.

 

[dragonov_elite]

Two of my friends run the big hx35 with twin scroll manifolds with twin scroll 12cm^2 housings and they are both getting full boost closer to 3000rpm than 3500rpm. Peepers can get it to surge brake boosting on the freeway at 2200-2500rpm. The one will be taking back pressure data with bigger turbines, but it doesn't feel down any power from the bigger housings he was running.

If you were going to go through the trouble of fabricating and all you wanted was hx35 power you'd be better off building a TS manifold. If you had to use a smaller turbo, a T25 or 13g would be able to help spool it. A 14b or 16g wouldn't offer a huge amount of spool up increase. I don't have much information on the hx40.

Not to thread jack. You should convince your friends to upgrade to 16cm^2 so i can see what kind of spool response is ran with that setup because that's what I'm putting together .

BTW awesome build 99gst_racer. keep the great work up.

 

[DashLaflash]

Two of my friends run the big hx35 with twin scroll manifolds with twin scroll 12cm^2 housings and they are both getting full boost closer to 3000rpm than 3500rpm. Peepers can get it to surge brake boosting on the freeway at 2200-2500rpm. The one will be taking back pressure data with bigger turbines, but it doesn't feel down any power from the bigger housings he was running.

If you were going to go through the trouble of fabricating and all you wanted was hx35 power you'd be better off building a TS manifold. If you had to use a smaller turbo, a T25 or 13g would be able to help spool it. A 14b or 16g wouldn't offer a huge amount of spool up increase. I don't have much information on the hx40.

I'm curious to know how compounding compares to twin scrolling. I've never heard of an hx35 spooling by 3000rpm. That's insane . If paired with a t25 would the holset reach full spool before 3000rpm? Twin scroll would definately be easier than compounding especially since people already build twin scroll manifolds and there would be less space issues.

Is it possible to generalize yet? Like with strokers we say you generally gain 500rpm of spool but lose out on max rpm. I know kevin said he doesn't need nitrous anymore but how much of a gain in spool is that? I'm unfamiliar with nitrous.

 

[Mr Peepers]

Not to stray too far off topic, but I see 20 psi a hair before 3000 in 4th gear with a 12mc^2 twin scroll hx35. Brake boosting can obviously be a good bit sooner but it surges when it hits 25 psi or more before 3k.

http://www.dsmtuners.com/forums/tur...mplete-installed-systems-2.html#post152484959

I personally don't see a reason to try compounding an hx35(as the larger turbo) with the twin scroll turbine housing already there.

 

[Blurred Talon]

All this Holset talk, what about an HX52 or HX55? Now where getting into big turbos. Coupled with a Evo 3 16G maybe?!

 

[pj91gsx]

Not to stray too far off topic, but I see 20 psi a hair before 3000 in 4th gear with a 12mc^2 twin scroll hx35. Brake boosting can obviously be a good bit sooner but it surges when it hits 25 psi or more before 3k.

http://www.dsmtuners.com/forums/tur...mplete-installed-systems-2.html#post152484959

I personally don't see a reason to try compounding an hx35(as the larger turbo) with the twin scroll turbine housing already there.

That surprises me for a holset turbine housing.

 

[bastarddsm]

Has anyone thought about a remote mount atmospheric turbo?

 

[Kevin Jewer]

Yes, under the bed of my Mightymax. Or my F250.

 

[hurricanecris24]

All this Holset talk, what about an HX52 or HX55? Now where getting into big turbos. Coupled with a Evo 3 16G maybe?!

Im using a hx52 with a small 16g for my setup.

 

[TimIacobucci]

After seeing some really nice remote mount setups and how much they don't affect the engine bay setup that route really made allot of sense to me for compounding.

Leave the stock location turbo as is and optimize it as you normally would, then run the exhaust to a big remote mount in the back. Especially a big holset.

I think this would yield the same results and minimize the complexity and cost when you start trying to package big big turbos for the larger turbo on a compound setup into a cramped ass dsm engine bay. That is to say, when you don't have the luxury of a drag rwd conversion stripped engine bay and holes in your hood like some people.

1. MORE ROOM!

2. Minimized fabrication. No need to mess with the small turbo layout that's already there.

3. Weight distribution. Bigger big turbos like hx52 and up have very heavy turbine housings, where would that weight help more, in the back or added out beyond the front wheels like every other damn thing on a dsm?


There are new problems like compressor ducting though. Need to get air to the big turbo and send it back up front.

Also not sure about the exhaust sizing between the turbos. I imagine you might loose some energy feeding the big turbo with a 3" pipe. Though I'm not sure how to size that pipe really. Any suggestions on that remote compound or no? Theres a calculation or rule of thumb for most every other exhaust sizing, whats the word on this one? Diesel guys must have an idea?

 

[99gst_racer]

Some friends and I have discussed remote mounting with compound turbocharging. I feel it would work better than most of them think. Only one way to find out though.

I don't think remote mounting would be less complex. You'd be pumping oil to and from the rear of the car. Not to mention the overly long charge pipe. But it would free up some room under the hood as well as reduce under hood temps - both of which would be HUGE benefits.

Talking about it really makes me want to see someone try it.

 

[Kevin Jewer]

I really want to try it on the MightyMax, but I can make all the power that truck can handle with one turbo. Being able to use the bed makes rear mouting make even more sense IMO. Room for a standalone oil system, AW intercooler, air intake/filter up off the ground, etc. Several drag trucks have done rear mounts recently for these reasons. You could do the same with a trunk/hatch of course but then it's all inside the car.

After I try a few things at the beginning of this season I'm making the first change to my initial turbo setup. My broken wrist is killing my motivation to write it all out, but all of the calculations I've been doing over the last few weeks lead me to expect another 13lbs/min and 15 psi less back pressure at the same boost. I'm hoping that gives the car a kick in the ass in time for the shootout.

 

[TimIacobucci]

Cool to see I'm not totally out in left field with that idea and the two of you have also thought about trying it as well. Or maybe all of us are out in left field together. Either way I think I'm in good company. Usually I say stuff like this and people go, "oh, I get it, you're crazy".

I don't think remote mounting would be less complex. You'd be pumping oil to and from the rear of the car. Not to mention the overly long charge pipe.

I do realize it's more if a trade off of complexity more than an overall reduction. It's just that allot more people have been doing remote mounting and the solutions and parts, ie oil return pumps, are sort of worked out already. Running it all up front and making at all fit and flow well and managing the heat like you say is sort of a whole new unique set of design challenges.

I really think looking at your setup Paul that with those size turbos and piping your about maxing out the space and heat potential of the fwd layout engine bay. Fitting bigger turbos with requisite bigger piping is gonna make a similar design allot tougher it seems to me. Unless you move to a tubular style front end and half radiator setup which is an option I guess.

You could do the same with a trunk/hatch of course but then it's all inside the car.

That's exactly what I was thinking actually. Naca duct in the quarter window to feed it and maybe water to air intercooler in the back. Still gotta sheet metal the turbo off from the passenger compartment but with some access plates it seems like it would be a hell of allot easier to get to and work on that way as well. Most pro drag cars running big engines and turbo(s) got the water to air intercooler running shotgun or in back now already. As long as it's only compressed air in the charge pipe it's not terribly dangerous running the piping through there back to the front. In fact with a compound setup you would be running less relative pressure in that pipe than a high power single.

After I try a few things at the beginning of this season I'm making the first change to my initial turbo setup.

Anything to add on what those changes might be?

Do either of you two have specific ideas on the secondary turbine exit to primary turbine inlet pipe sizing? Or wastegating? Or turbine sizing? Relative to compressor flow?

I am thinking relatively large wastegate flow and large pipe and turbine sizing on the primary turbo would really boost the overall efficiency of the system without much trade off in spool.

I really think this idea has proven it works. It just hasn't been pushed to the point is really displays dramatically how much of a difference this system makes vs traditional methods like the diesel setups do because they really push the boost pressures and flow efficiency of their turbos more often.

Using a higher flowing primary turbo at high boost and efficiency at much lower rpm is where this system will shine. Massive torque everywhere you want it without the super high rpm you usually need to run equivalent mass air flow on a traditional small engine single turbo.

I actually think this system if optimized for a lower redline would make the overall engine system, the engine build and turbo combination, less complex and more reliable overall, as crazy as that may seem sheerly because of reduction of max rpm. The less rpm the less crazy parts and wear and maintenance that is needed to run this engines at 10k+ rpm. It's just a matter of where the crossover is. At lower power and boost levels a well setup single twinscroll will get the job done for much less complexity. It's really the crazy high rpm high mass flow where this system will prove itself. That's what I don't think most people understand about what Paul has done here. Mainly proof of concept, and now that's checked off it just needs to be pushed harder to really show it's merits.

 

[CrawlinZX14]

I did a twin compound setup on a Dakota RT. The longer pipes helped to cool down the charge and allowed for no intercooler in my setup. I started on a twin turbo setup feeding into a single intercooler then single output to the TB. Lowered the boost pressures while increasing volume, but the lag was no fun. The compound setup was far more efficient. I used a CT-26 turbo from a supra as my small turbo, and a 70mm single scroll turbo for the large one. With that setup I was able to take my 427ci small block mopar and remote mount the turbo's under the bed. For oiling, I simply made a second oil box. I set the turbo's high so I could gravity feed the oil from the turbo's down into the tank, and ran a small oil cooler before the turbo's with an electric oil pump to supply the pressure. They made great power in that small block mopar. I would love to see someone do a remote mount setup in a dsm. would be interesting to say the least.

 

[TimIacobucci]

That sounds really bad ass. Have you got any pictures handy? Especially under the bed and oiling setup.

What pump was it exactly?

How did you route the charge pipe? Back underneath?

 

[CrawlinZX14]

Unfortunately no. I would have to hunt thru my records to figure out the pump I used for oiling. The charge pipe was ran all the way back up to the front along the frame rail, up the fire wall and to the 100mm TB. We actually cut out the bottom of the bed of the truck. Built custom braces to hold the bed in place. Made a custom skin for the top to act as a tonneu cover, except it was welded with X bracing for support. I only have 2 pix left of the truck, and nothing of the engine bay. I had a ton of custom work as there is not very much available aftermarket for the dakota RT's. Custom fuel rails, custom intake, custom exhaust and intake pipes. Fuel cell ran off A1000 pumps, ford 9" rear. 360ci R3 mopar motor. Bored and stroked out to 427cid. Made 723whp on motor only. never got a full pull on boost, it would not hook up. last pull was 1137rwhp and the tires broke loose on the dyno. The truck scared the holy hell out of me. After I sold it, the new owner did a full 4 link setup and made a low 8 sec pass with it. IDK what all he did after that. I wish I would have had more pix of it, but lost most on an old laptop that was trashed. I will see if I cant possibly find some more details on the setup tho. I have to search for the log records from that build.

 

[Kevin Jewer]

Anything to add on what those changes might be?

I could ramble on a bit about it, I'll make another post soon with details.

Do either of you two have specific ideas on the secondary turbine exit to primary turbine inlet pipe sizing? Or wastegating? Or turbine sizing? Relative to compressor flow?

I am thinking relatively large wastegate flow and large pipe and turbine sizing on the primary turbo would really boost the overall efficiency of the system without much trade off in spool.

For WG size, you need a lot. About double what you would need for the small turbo alone, or more closely, what you would need with the small turbo times the PR of the big turbo. In my case 1 V44 wasn't enough, but two of them are. WG flow is enough if the turbo in question isn't boost creeping.

For secondary turbo turbine sizing, the main consoderation IMO is spool, since spool up is based mostly on that. Secondary consideration is turbine expansion ratio. Primary turbine drive pressure is multiplied by secondary turbine expansion ratio (or turbine PR) to equal total system drive pressure (EMP).

Anything to increase piping diameter from the secondary turbine exit will help spool up, and as long as it is merged into the primary turbine inlet without turbulence, won't affect spool up on the big turbo.

I really think this idea has proven it works. It just hasn't been pushed to the point is really displays dramatically how much of a difference this system makes vs traditional methods like the diesel setups do because they really push the boost pressures and flow efficiency of their turbos more often.

Using a higher flowing primary turbo at high boost and efficiency at much lower rpm is where this system will shine. Massive torque everywhere you want it without the super high rpm you usually need to run equivalent mass air flow on a traditional small engine single turbo.

I actually think this system if optimized for a lower redline would make the overall engine system, the engine build and turbo combination, less complex and more reliable overall, as crazy as that may seem sheerly because of reduction of max rpm. The less rpm the less crazy parts and wear and maintenance that is needed to run this engines at 10k+ rpm. It's just a matter of where the crossover is. At lower power and boost levels a well setup single twinscroll will get the job done for much less complexity. It's really the crazy high rpm high mass flow where this system will prove itself. That's what I don't think most people understand about what Paul has done here. Mainly proof of concept, and now that's checked off it just needs to be pushed harder to really show it's merits.

This is essentially what I do. I'm geared to only run to 9500 rpm, which requires no special maintenance to do reliably for long periods of time. I went two full seasons of 8 second passes on my last motor before pulling it for a freshening up.

My setup has gone high 8s consistently, but I'm hoping to push that further this year, since I've already done high 8s on a single 4294. The beauty of this setup though is that it spools on its own with the auto trans. With the right turbo selection and interstage intercooling, and knowing what I know now about this stuff, I'm certain I can make significantly more power with with the same spool.

 

[TimIacobucci]

This is essentially what I do. I'm geared to only run to 9500 rpm, which requires no special maintenance to do reliably for long periods of time. I went two full seasons of 8 second passes on my last motor before pulling it for a freshening up.

My setup has gone high 8s consistently, but I'm hoping to push that further this year, since I've already done high 8s on a single 4294. The beauty of this setup though is that it spools on its own with the auto trans. With the right turbo selection and interstage intercooling, and knowing what I know now about this stuff, I'm certain I can make significantly more power with with the same spool.

Yes you are absolutely right, I was even thinking about you while I was writing that and thinking I hope he realizes I was focusing more on Paul's setup and not your own.

I think maybe because your's is a race application people think a crazy setup makes more sense and the complexity and all is more justified. When in fact like you say the overall maintainability and versatility are improved on your setup using compounding. It's not to discredit what you have done but it feels like most people focus on Paul and if they think the results aren't impressive completely gloss over your car. The principles are the same on the street or strip, awd or rwd, and you have done exactly what I was saying.

The other cool thing like you said about trying to push it harder now, is maybe the craziest highlight of this setup most don't recognize. How much more headroom do you have on the primary still? You're going 8's and not even leaning on either turbo at all yet huh? The biggest asset of this setup it seems to me is where you are right now, it's all built and working and your not revving to the moon and you can just tweak stuff and keep pushing it on that same setup pretty damn far just upping the boost. No bigger turbo or trade offs or increased rpm, just more boost and lower ets.

For secondary turbo turbine sizing, the main consoderation IMO is spool, since spool up is based mostly on that. Secondary consideration is turbine expansion ratio. Primary turbine drive pressure is multiplied by secondary turbine expansion ratio (or turbine PR) to equal total system drive pressure (EMP).

Anything to increase piping diameter from the secondary turbine exit will help spool up, and as long as it is merged into the primary turbine inlet without turbulence, won't affect spool up on the big turbo.

So let me get this right, your saying the more flow you can get before the primary turbo will make it spool faster?

I was wondering if modifying an existing 3" ss exhaust from the secondary turbo to a remote mount in the hatch would slow the gas velocity too much and hurt the primary spool. But if that's a happy size for the secondary turbine exit and I taper it back down smoothly to meet the primary turbo turbine inlet you think will just help?

I am thinking this works because the primary turbo spool time is basically just based on the gas energy increase from the secondary turbo spooling. The secondary like any turbo will spool faster with a larger pressure differential across the turbine so the least back pressure between the stages in the exhaust the faster the spool for the secondary and consequently the primary.

I guess it seems obvious now but it's sort of counter intuitive. That's pretty cool actually, with a normal setup pre turbine flow is usually a compromise for gas velocity and spool time, with this you actually spool the big turbo faster if the exhaust before it flows better.

Running with that idea I am wondering if it might actually be beneficial to open the primary turbo wastegate to bypass the exhaust restriction the primary turbine causes to spool the secondary faster, then have it close at a low boost pressure. I am thinking if the primary is sized properly it will be pretty large and need a decent amount of boost from the secondary going into the engine before it really starts to wake up, it would require some experimentation but I bet a delay to a relatively high boost pressure than most would anticipate would actually still spool the primary faster than not.

 

[Kevin Jewer]

How much more headroom do you have on the primary still? You're going 8's and not even leaning on either turbo at all yet huh?

I'm actually maxed out on the primary (big) compressor. People often ask why the car is only going 8.80s with the S475 when others are going mid 8s and in rare cases (kiggly) low 8s. The reason is that the PR is so low. If you look at the map, this ~105 lbs wheel maxes out at closer to 90 lbs at a low PR of 2:1. So from a PR perspective, the turbo is out for a Sunday drive (17 psi), but is giving all the flow it can. Running off the map over 8500 or so is evidenced in the turbine drive pressure spike above that rpm. Running this same compressor at a higher PR will move more air more efficiently. I'm hoping to step up to an 80 or 88mm compressor this year.

So let me get this right, your saying the more flow you can get before the primary turbo will make it spool faster?

I was wondering if modifying an existing 3" ss exhaust from the secondary turbo to a remote mount in the hatch would slow the gas velocity too much and hurt the primary spool. But if that's a happy size for the secondary turbine exit and I taper it back down smoothly to meet the primary turbo turbine inlet you think will just help?

Anything to reduce post secondary (small) turbine pressure will improve secondary turbo spool, which is also your overall turbo spool since it spools first. In the case of a remote mount setup you could step it up right at the turbine discharge, then gradually reduce it to a smaller size for the long run back to the big turbo if you're worried about velocity. Some OEMs are going to this type of downpipe to improve spool, look up expansion chamber downpipes, there is very little info out there, but there is some. As far as whether 3" all the way back or reducing it to 2.5" is better, I can't say.

I am thinking this works because the primary turbo spool time is basically just based on the gas energy increase from the secondary turbo spooling. The secondary like any turbo will spool faster with a larger pressure differential across the turbine so the least back pressure between the stages in the exhaust the faster the spool for the secondary and consequently the primary.

I guess it seems obvious now but it's sort of counter intuitive. That's pretty cool actually, with a normal setup pre turbine flow is usually a compromise for gas velocity and spool time, with this you actually spool the big turbo faster if the exhaust before it flows better.

Running with that idea I am wondering if it might actually be beneficial to open the primary turbo wastegate to bypass the exhaust restriction the primary turbine causes to spool the secondary faster, then have it close at a low boost pressure. I am thinking if the primary is sized properly it will be pretty large and need a decent amount of boost from the secondary going into the engine before it really starts to wake up, it would require some experimentation but I bet a delay to a relatively high boost pressure than most would anticipate would actually still spool the primary faster than not.

Until the big turbo starts to make boost, there is very little back pressure from it (expansion ratio near zero), so I don't think it gets in the way. This can be seen in the spool data I collected early on, timing the spool up on the small turbo with and without the big turbo installed. There was no noticeable difference.

 

[LandSpeed-DSM]

I am in the process of fabbing up the exhaust side of a compound system on a 1.5L SOHC Honda. Using a 61mm Garret in a 1.15 A/R open T4 with 3 inch V-band discharge as the primary and a MHI 14B in the 6cm2 housing with the wastegate blocked off..

Because I want to be able to run as little as single digit boost should I choose, and as much as 3 Bar or more on kill mode, I have opted to go with a 60mm TiAl off the exhaust manifold and a 44mm gate on the inter-stage pipe between the 14B and the Garrett.

The plan is to eventually move 55+ lbs/min and wind out as high as 11k rpm on a 309/292 cam.

Basically I am wondering if I have the wastegates laid out properly... With the bigger 60mm gate used to bypass all that exhaust around the small turbo, and the 44mm to regulate boost on the primary?

You must be logged in to view this image or video.

(Crappy hand-sketch because the stage I am at doesn't give much of an idea of the room/layout I have to work with)

I am curious to hear what others experienced in this field have to say.

 

[bastarddsm]

That looks like your headed in the right direction.

 

[Kevin Jewer]

Those WG sizes should work fine. I have two V44s on the manifold and one on the big turbo and they bypass enough for a 75mm compressor. Then again I don't run low boost, so it may not be a good comparison.

 

[bastarddsm]

I'm actually maxed out on the primary (big) compressor. People often ask why the car is only going 8.80s with the S475 when others are going mid 8s and in rare cases (kiggly) low 8s. The reason is that the PR is so low. If you look at the map, this ~105 lbs wheel maxes out at closer to 90 lbs at a low PR of 2:1. So from a PR perspective, the turbo is out for a Sunday drive (17 psi), but is giving all the flow it can. Running off the map over 8500 or so is evidenced in the turbine drive pressure spike above that rpm. Running this same compressor at a higher PR will move more air more efficiently. I'm hoping to step up to an 80 or 88mm compressor this year.

Is that really true that the compressor maxes out at lower flowrates at less psi? Its sort of counter intuitive to me. I know the maps make it look like you have less mass flow available at lower pr's, but I assumed that the choke line was vertical, and it just has shit for efficiency at low pr's.

I guess I was under the assumption that the wheel can only take in a certain volume of air (which has a mass determined by ambient conditions), and that is defined by the area of the compressor inlet. What comes out of the compressor is that same mass of air, but at a much reduced volume and higher pressure.

 

[Kevin Jewer]

When I say maxed out I really mean efficiency is in the toilet, drive pressure is climbing exponentially, and I don't want to push it any further since it's at the point of deminishing return. The maps only go to ~60-65% effciency, and they'll run lower than that. I'm already off the map. For comparison, my 75mm wheel is at 60%, an 80mm would run at 66%, and an 88mm at 77%! Add to that the fact that I'm expecting ~13 lbs/min more flow, and the current turbo isn't going to cut it. I'm also focusing on reducing drive pressure, so compressor efficiency becomes important.