Twincharging meets Compound Turbos - Calling the Experts! 3

[drxlcarfreak]

Hey all, this is my first post on here, so go easy on me! I have been working on a theory of mine that I finally think is ready for other eyes to to see. Everything right now is all in theory stage, but I have extensively researched every portion that I could think of on the subject. I have read and absorbed every document/forum I could find on both compound turbos (Primarily on the 2JZ engine and 4G63 engine, but read about the diesel compound turbos as well) and twincharging (The most useful ones on this website).

I have begun the VERY early stages of this project by sourcing an older Eagle Talon TSI with a 4g63 engine and installed a stand alone type ECU for tuning, as well as completely upgraded the fuel system to support about 500hp. I will be getting a second block soon to build a stronger block with forged internals to withstand much more power.

Ok, enough lead up. I think it may help if I explain the evolution of my theory… Feel free to skip this section.

I began reading about compound turbo systems installed on a 2JZ and was amazed at the incredible HP levels obtained as well as the widened powerband. On the particular setup that I saw 7-800 HP was made from a 3.0L and I believe full boost was available at about 5k RPM. My goal was to get to these power levels, but try and increase the powerband even more by getting full boost even lower in the RPM band. I would like to have enough boost to get about 600-650HP from a 2.0L engine (4G63), and have this boost level as low in the RPM range as humanly possible. The ultimate goal is having a torque curve of an 8.0L engine, but still able to get somewhat decent gas mileage at low RPMs/partial throttle.

My original thinking was to better a compound turbo by adding a 3rd smaller turbo in series to get the high boost levels, but thought that by the time the exhaust goes through 3 turbine wheels there wouldnt be enough exhaust heat energy to properly spool a 600HP turbo. I dont have any proof to substantiate this, and really cannot find any information about it.

From this I moved to adding a centrifugal supercharger to overcome the lowered exhaust energy, but it was quickly dismissed as the minimal gain wasn’t worth the expense and complexity. So then I moved onto positive displacement superchargers.

End of evolution…

So now that I have decided that I need 2 turbos and a positive displacement supercharger I need to figure out what order they go in.

Sidenote: I made an excel file that calculates the curves and interactions between chargers as close as I could get them in excel without nasty circular references. The file can be downloaded and messed around with here (It is pretty big, so be patient!):

https://www.box.net/shared/ezqzysu8tj6tii2ph7y7

Supercharger First: Atmosphere>SC>Big TC>Small TC>Engine>Small TC>Big TC>Exhaust

My thought was to have two intakes with the supercharger on one and a wastegate on the other. When the turbos begin to overtake the supercharger the wastegate opens allowing air to flow freely into the system bypassing the supercharger. Playing with the numbers of the supercharger this can give an almost perfect curve, but in reality in order to get the pressure levels needed the supercharger would have to be massively oversized for a 2liter engine, which would take too much power from the engine to spin that it wouldn’t be feasible. Also at a boost level over 40PSI, even for a short duration is sure to just be spitting amazingly hot air out.

Supercharger Last: Atmosphere>Big TC>Small TC>SC>Engine>Small TC>Big TC>Exhaust

My next thought, was put the supercharger behind both turbos at lower boost levels. This didn’t really move the boost curve to the left enough to make it worth it. I did play with the idea of trying to fit a CVT system into it to have higher SC boost early and lower it as the TC came online (I want to do this so each charger in better in its efficiency range creating lower IATs, plus there is less drag on the motor from the SC pushing 5psi rather than 20psi. While this does give the overall best result, I haven’t seen any companies other than Fallbrook even attempting this, and they are still in prototype stage (

http://www.fallbrooktech.com/03_CVAD_boost.asp).

Supercharger Second: Atmosphere>Big TC>SC>Small TC>Engine>Small TC>Big TC>Exhaust

This brought me to having the supercharger between the turbos. While the ducting of this is going to be the most complicated, I think right now until CVT drives for superchargers are released it seems like the best setup. It isn’t the full boost off of idle that I was hoping for, but there would definitely be a lot more boost available from the get go. I know that most positive displacement SC have PR limits of about 2.2, but that is rated for throughout the RPM band. Does anyone know if a positive displacement charger could reliably peak higher PSI for a part of the RPM band and then settle to a lower point?

I know that on a compound turbo system I could use a smaller high pressure turbo to reduce boost lag, but then the small turbo would be overpowered by the large turbo and be choked out. I could also up the boost of a supercharger and increase the size of the turbo on a twincharged system, but the supercharger would be creating a lot of hot air and taking away more power from the small engine.

Well, that was a very long winded lead up to my overall question. I know that it is going to be very complicated and tight to get everything working and fitting together and will not be at all cheap to do. In fact I could probably get a couple LS9 engines for the price this would cost, but I guess I want to be different. Do you guys think that a system like this would be at all beneficial? I would like to hear what input you may have! If you think it is a stupid idea by all means tell me, but please back it up with actual (or theoretical) evidence. Don’t just say that idea sucks and move on. I would love to hear the thoughts you guys have on the different setups as well.

 

[Lou Scannon]

Questions for you:
1. How much boost pressure are you aiming for?
2. What is the intended use of the engine?
3. What is the desired longevity of the engine?

I forgot what I was going to say

 

[drxlcarfreak]

Hey Hemi.

I am looking for as high as possible boost pressures, but have seen 47 somewhat reliably. I would like to aim around there, and possibly tune for just a bit higher for race days. It would be for a weekend warrior. Mainly street driven, but taken to the track ocassionally (Drag/Auto X etc). I think at about 600HP with forged internals would be able to last pretty long, provided that it isnt revved out all the time. That said, I would like to get a couple seasons out of it.

 

[drxlcarfreak]

I should add that while PSI is a good indicator of power, it is the desnity of the charge that I am really trying to get here, which is why for most of the powerband the supercharger is making lower boost so the heat gain is minimal and each turbo will be sized to be in its best efficiency range. In addition, I plan to have interstage cooling after each stage of compression. I haven't quite figured out the best way to cool the charge air yet whether it be Air/Air, Air/Water, Water-Meth Injection or a combination of all three. I haven't seen results from multistage cooling yet, so who knows the charge may be cool and dense enough that I may not need to hit 50psi for my goal.

 

[Lou Scannon]

Assuming you mean 50 psig, your goal is 65-ish psia. Like you say, a modern positive displacement supercharger stage should be able to deliver 2:1 PR. Assuming 14-ish psia atmospheric pressure, you're looking for 65/(14*2) = 2.3 PR from your turbos. So why do you need anything more than conventional twincharging?
Atmosphere>TC>SC>Engine>TC>Exhaust

I forgot what I was going to say

 

[drxlcarfreak]

Yes, sorry I was referring to gauge pressure, not atmospheric. My plan was to use the supercharger to increase the response of the turbos and then drop to a lower PSIG. My thinking is that a positive displacement supercharger at a PR of 2 will be running at a lower efficiency than a turbo (High 50s vs high 60s) and creating more heat. Plus I am thinking that lowering the PSI of the supercharger will reduce the load on the engine which should increase the net torque output.

Originally I was thinking of bypassing the supercharger altogether, but after reading around a bit about how the SC creates net negative back pressure which provides a cooler intake charge, hopefully reducing my chances for detonation!

 

[Lou Scannon]

I've read that the modern Eaton superchargers are pretty darn efficient, comparable to a turbo, up to their rated PR. And, BTW, a decent modern turbo that is well matched to the application should be running in the 70s at mid to high load.
I still don't see a case for all the complexity of your proposed system versus conventional twincharging. And if you do successfully make a case that conventional twincharging is inadequate for your goals, I will then start digging into the technical integrity of your proposed system, which I haven't done yet (one step at a time).

I forgot what I was going to say

 

[madmac666]

Some valuable info here on my 4G63 build:

http://www.eng-tips.com/viewthread.cfm?qid=267889&page=1

Hemi is right, i think you're over complicating your build when a simple twincharger system with turbo feeding SC will give you the results you're looking for.

 

[patprimmer]

I absolutely agree with two posts immediately above this post


Regards
Pat
See FAQ731-376 for tips on use of eng-tips by professional engineers &

http://eng-tips.com/market.cfm

for site rules

 

[drxlcarfreak]

Well, my reasoning for going the much more complicated route is 3 fold.

One is spool time. Say we are aiming for about 650hp at an overall pressure ratio of 4 (44.1psi). In both systems the low pressure turbo must be capable pushing this much air. In my case this would probably be something GT35R. On my triple compounded system I would only be asking this turbo to spool to about 13 psi, where on a twincharge setup this turbo will need to spool up to 29.4 psi. In my rough calculations with the impact of the supercharger/turbo interaction on each other in my setup I would be getting full pressure about 1,000rpms sooner (3,000 vs. 4,000).

A second is the power taken to run supercharger. In order to keep the calculations somewhat simple, I had to pretend we are in a perfect world so say 1 lb/min is the equivalent to 10 hp and that the supercharger takes 30% of its gross output to spin it. I know it is very different for a compounded system but lets just quickly say at 6,000 RPM we are at 14.7psi for the twin charge and somewhere around 6 for the triple charged system. That means the lb/min that the SC is compressing in the twincharge setup is about 31.4 and the tricharge setup is 22.8, which results in roughly 315 hp and 230 hp net respectively. If we assume that the SC is taking 25% of its output to power itself that’s 105 and 75 hp respectively. Negating heat (which would increase the power more), that is about 30hp we can gain by having a turbo do the work instead of the SC.

Air intake temps are the third main reason for me wanting to try this setup. I did some quick calculations based on my theoretical spool up times and came up with the table below. I did a quick rough sizing of what I would need for my chargers and came up with a GT35R low pressure turbo, TD06-20G high pressure turbo and something like an M45 or M62 supercharger. Based on the pressure ratios of each, I grabbed the efficiency of each and plugged it into what each charger was producing at a set RPM. Even with the compounded inefficiencies of each charger it looks like I am able to get cooler intake temps due to the fact that I have an extra cooling charge, and am pushing a lower pressure ratio on the most inefficient charger. Not only does this help reduce detonation allowing me to attempt to get more boost out of it, but it increases that density of the charge resulting in more HP (Granted, it is only about 10-15hp).

Now, I know I am probably going to be ridiculed about assuming or missing this and that variable, but I understand that these are all theoretical values, and there is no way that I can account for every variable. There are about 100 that I know that I missed (such as boost isn’t rpm dependent, its load dependent), and probably about 1,000 that are so out of my depth that I couldn’t begin to comprehend them. For the sake of trying to get my point across though, I feel like the simple calcs I have done are somewhat adequate. If not, please let me know!

Madmac, actually your post is what brought me to this site and helped me better understand the principles of the supercharger side of things. It is a very informative read!

 

[patprimmer]

Hews Flash

It takes power to drive a turbo. Because there is no direct mechanical linkage it is not so obvious.

Regards
Pat
See FAQ731-376 for tips on use of eng-tips by professional engineers &

http://eng-tips.com/market.cfm

for site rules

 

[Lou Scannon]

drxlcarfreak, while most of your reasoning is qualitatively correct, the quantitative results are what really count. As you mention, your proposed setup is difficult to analyze due to the complexity. Even a twincharge setup is a tremendous undertaking, ask madmac666. If I were going to go down the path you're proposing, I'd first try to get a twincharge system working, and tinker with it until I have a fair grasp of the interactions. At that point, if I saw potential gain from adding a second turbo, I might take that additional step.

 

[drxlcarfreak]

patprimmer: I know that it takes power to drive a turbo as well, but it is a lower percentage than a supercharger, which nets more WHP/psi which is my overall end goal.

Hemi: I definitely will be taking the whole project step by step to completely learn the interaction between each charger and the engine. I just wanted to hear what people experienced in twincharging and compound turbocharging thought about my logic of the interaction between the chargers before I began the endeavor. For example, I know in compound turbo charging it is advisable to run each turbo at equal pressure ratios to keep backpressures low (low pressure turbo at 14.7psi and high pressure turbo at 29.4psi).

 

[madmac666]

Exhaust backpressure is one of the most relevant issues with turbos. The lower the backpressure, the easier the engine can breathe and the lower the pumping losses in the engine are. I've yet to see any compounded turbos that have very low backpressure due to the housings and the nature of the airflow through the impellers, and that's a bad thing for any engine. The power needed to drive a SC can easily be produced at the crank if the exhaust backpressure is low enough that the engine breathes easily. Intake pressure will be higher than exhaust backpressure so the charge entering the cylinders will push most, if not all of the spent exhaust gasses out of the exhaust valves and the opposing cylinder on its power stroke will put all of its torque to the flywheel instead of using a percentage of that torque to push the exhaust gasses out of the opposing cylinder.

As drxlcarfreak says, start with a conventional SC/Turbo compounded system and IF that doesn't produce the results you want then you can look at adding in your 2nd turbo. I feel you will learn enough from the conventional system and build your own picture as to how it works and where changes need to be made. Until you bolt it all together and fettle with it, it's very difficult to understand fully how each area of the system affects the engine performance. My system is very simple and we're seeing 900hp+ on a 2.3l 4-pot engine with great reliability from the system. I simply cannot understand why you would want to complicate it with 3 compressor devices + the weight and complexity of that when the conventional method will do exactly what you want to achieve. Walk before you run seems apt in this build!

 

[snoopnoon]

i would advise you to read through newengdsm.org/forums I used to belong to that club and there is endless amount of information for that engine. Also, there are guys on there making that power with the stock motor and single TC without to much sacrifice in powerband.