[GrumpySteelMan]

...most people that have a single large turbo vs. twins need the power and are ok with the increased spool time because they can afford a few seconds for a very rewarding pay off down a straight away where as usually a tt setup the spool time is decreased and in my scenario the two turbos would be smaller so they would not yield as much top in power but will provide it in a shorter time.

A really huge single will be laggy, but pretty big twins will act the same way, unless they are sequential, similar to the OE RX-7 TT setup.

I still think a properly sized single makes the most sense no matter what, unless packaging is the issue. The quoted post about the formulas and pumping losses, etc of single vs twin is a well written post...but they're stuck in academia for the most part IMHO. In practice you have lower overall efficiency with twins compared to a single, and they agree with that.

Putting those twins so close to each other may not be better than a single, but it is pretty flipping cool.

I'm going to see about picking apart the academia real quick. I would like to say ahead of time that I am in no way attempting to insult "Master T."

Some people argue "one turbo gets all the cylinders' energy, twins gets half per turbo and significant exhaust pulses, which make the twins more inefficient". This is mostly invalid because modern turbine housings are radially split and tangential, so the exhaust pulses are calmed upon entry to the turbine, and more effectively handle unsteady energy flow. Also, inertia for the most part prevents any major transience with (heavily) pulsed exhaust energy.

Lacking in facts. Modern turbine housings are generally tangenital, but the radial split (also known as a divided inlet) is NOT common by any means. In fact MOST modern turbochargers for small (that is not large consumer/commercial diesel) engines are NOT divided at the inlet. Furthermore the split inlet on the housing is only advantageous if the turbo manifold is properly designed to take advantage of the housing design...most are not.

You get more exhaust pressure with a single than twins, which spools it faster. This can be eliminated because exhaust pressure is not really dependant on the turbo, moreso the manifold. Additionally, exhaust pressure does not have a direct effect on exhaust energy. While it IS an indirect effect on exhaust pressure, nothing is going to add energy to the exhaust flow, especially making the engine pressurize the exhaust more. This not only makes the engine and turbo less efficient, it is useless. Looking at the thermodynamic equation for energy in a fluid flow:

(((Formula Removed)))

Turbines (the exhaust driven side of the turbo) is driven by heat and the temperature differential across the blades. The exhaust 'pressure' argument is invalid for all intents and purposes.

There are two reasons to argue a single vs. twin turbos. If you compare the mass moment of inertia for a solid disc, you get 1/2*m*R2 on the axial axis based at the center of the circle (fig. 1) (call this z) and 1/4*m*R2 on the radial axis (x and y). For a circular cone (or circular cone section), the mass moment of inertia is 3/10*m*R2 along the axial axis centered at the center of the base circle (fig. 2), is 3/10*m*R2. The same is true for a cone sectioned by cutting it along the z axis with a plane parallel to the XY plane. Now observe the mass. In such a cone, with a constant density and a constant trim turbine wheel, the relationship of radius to mass is:

(((Formula Deleted)))

That means the large single turbo has 16 times the inertia. Inertia is the property of mass that is defined as "resistance to change in movement". Like the old saying goes: "An object in motion tends to stay in motion, and object at rest tends to stay at rest".

This makes single turbos look slow and laggy.

Instead of actually taking into account turbine flows and actual mass, they used a lot of formulas to estimate the size and mass difference between a 'large' and two 'small' turbines. Totally useless if you ask me. They should have taken an example like a Q-trim turbine's mass, flow, and area. Then found a pair of smaller turbines with similar combined flow characteristics. Once the did that, then fancy calculations about surface area, drag, mass, thermodynamics, etc. would have had some merit.

Lets talk efficiency. Tolerances in a turbo are pretty much the same regardless of size. What this means is that the error can be twice as much area, which can be twice the percentage of error. This affects smaller turbochargers a lot more than larger one. This, as expected, results in 4 times the efficiency loss due to clearances. The big single makes "better boost" then the twins.

Hence, the twins will spool faster, but once spooled, the single is more efficient. If you're driving a street only car and spool is your first concern, then twin turbochargers will suit your application better. If you are racing and know that you will not fall out of the powerband, and thus spool, the single will be a better choice.

This is outdated and possibly plain wrong. Ballistic Ball Bearing turbos are far superior, and this is one of the reasons.

Nothing in the previous paragraphs say WHY the twins spool faster. In fact the bearing tolerances argument would seem to point to twins being less efficient spooling AND at peak efficiency, right?

Also keep in mind another adage, "spool only exists in 1st gear and there is [among other methods] a simple solution in a bottle!" The use of tuning to decrease spool time will be addressed in another article, as well as actual turbine performance and efficiency.

Twin turbo setups are just extra complex if you ask me. You also need to think about making sure the wastegates are set identically. If I need to explain why, just lmk.

The examples used here are simplified for ease of calculation, but will work for any scenario. You probably aren't going to have a turbine wheel twice the size of another, but interpolation and some calculations give a result that says a turbine wheel about 15% bigger has twice the inertia.

I am glad they gave this a nod here, but I think the physics lesson was the focus of the post, and not the actual application of the ideas.

 

[Vic Cardenas]

I'm no turbo expert here but I will try to help anyway. I think you want the twin turbo setup for the benefits of having no lag and boost at high and low rpms right? How will you accomplish this with two of the same sized turbos? My understanding of twin turbo is to have a small and a big one for those benefits.

Concentrate on your supercharger damnit. You're all over the place man.

P.S. make me those chips. How about a 4, 6 and 7 and I'll give you back the stage 2 and the gold star.

 

[reboticon]

Twin turbo setups are just extra complex if you ask me. You also need to think about making sure the wastegates are set identically. If I need to explain why, just lmk.

Can you explain why? I don't doubt you are right, i just dont know the reasoning. I would see it if each turbo were coming straight from the manifold of its respective head, but if coming from the Y pipe, why would it matter?

Since you started picking apart the academia of the original post, I thought i would continue.. I may be wrong but this is what i noticed when i first read it.

It states PV=nRT where P= pressure,V = volume n= moles of gas, R = gas constant, .0821, T = temperature. It then reduces that to (PV)/T = constant. Should be (PV)/T = nR. It would appear to not be constant at all.

@Telion
There was a company that did custom headers that would send you plastic bendy tubes and a plastic collector. You then bent the tubes to the ideal positioning for your application, and mailed them back. They would use that a template and and weld you identical headers. I can't remember the name of the company, but last i checked they were quite reasonable in price and only slightly more than a new set of OBX headers would cost you.

 

[Telion]

Now we're cookin thank you everyone for the wonderful dissertations (good) A lot of the formula stuff I do not nor will I pretend to understand so to have it broken down is very useful for myself and many others I am sure.

Yes Vic I am kinda allover the place while at the same time very focused with the intent to make FI happen for my car in a successful way. but none the less keep the great info and setup ideas coming and hopefully before long I will have some pipes bent and welded up to get to the next step.

I do have an important question though, running one 3" pipe in to feed the two turbos and another 3" pipe for the exhaust side of the turbos and splitting it back out, what would be the positives and negatives of this?
And yes I know some of you may go back to the train of thought why not just run one. Grump said it best it would be cool. (amungst other theories)

Reboticon thank you for the info, it is pretty much the plan but not with the headers (yet). For now I am going to mock it up with PVC and weld it locally.

 

[BUZZARD]

We use a twin turbo setup for two reasons.

1. Engine layout. The general system layout is better than a single setup.

2. Fast spool. Two small turbos with the associated low inertia give fast spool and excellent boost response. Overall package provides a wide torque band with good mass flow for high rpm power.


Chris

 

[Telion]

We use a twin turbo setup for two reasons.

1. Engine layout. The general system layout is better than a single setup.

2. Fast spool. Two small turbos with the associated low inertia give fast spool and excellent boost response. Overall package provides a wide torque band with good mass flow for high rpm power.

Chris

Hey Chris I have a question for you, actually two. Do you run one turbo per exhaust side, with each exhaust header feeding each turbo?

are you running a custom manifold where each turbo feeds each head or do they join and go into a manifold that feeds both banks?

 

[Telion]

Ok I am going to kill two birds with one stone on this and this will be the Visteon radiator install set of pics and the pics to show the newly acquired space for the turbos mounting. Hopefully this will support my theory that mounting them up front won't be the issue and that the piping in and out will be. And as always please excuse the current condition of the car; I gotta hack it up before I fix it up.

In this shot you can see the custom top mounts that I made for the radiator.

Here is a great top view of the working room

Available space on the engine side

and on the passenger side ( the wet spot is water and the other thing is my CTS quickfix)

Good shot of the space saved by the sexy SPAL fan, man does it work!

 

[reboticon]

I had thought that running two turbos off of 1 pipe would cause a lag in spool time. My reasoning for this is one of the formulas originally stated. For finding the pressure and volume of gas pushing each turbo would be
P1 x V1 = P2 x V2
P(exhaust) x V (exhaust) = ([P(turbo1) x V(turbo1)] + [P(turbo2) x V(turbo2)])

I get lost here though. If I understand Grumpys post correctly, The temperature is what is responsible for spool time. Even if that is the case, I still think that the pressure and the volume have to be considered. We see from the equation for gasses that

PV= nRT. Reordering this equation we get T= (PV)/nR ; V= (nRT)/P ; and P= (nRT)/V . Basically this states that Pressure Volume and Temperature can all be expressed as the relationship between the other variables.

All formulas aside, I was viewing it like this. Imagine a garden hose that splits into a T at the end, with each end of the T being a water wheel. If the radius of the T hose outlets are the same size as the T hose inlet, then if the inlet is flowing its max rate, each hose would still only be recieving half of the water.

 

[BUZZARD]

Hey Chris I have a question for you, actually two. Do you run one turbo per exhaust side, with each exhaust header feeding each turbo?

Yes

are you running a custom manifold where each turbo feeds each head or do they join and go into a manifold that feeds both banks?

We are building two new engines and plan to trial both types of
intakes on the engine dyno. We previously used a single plenum intake that joined all port together.



Chris

 

[BUZZARD]

I get lost here though. If I understand Grumpys post correctly, The temperature is what is responsible for spool time.

Ever put a turbo in an oven at 200F ? 500F ? Did it spin?

Chris

 

[reboticon]

:rofl: No i have never put a turbo in the oven. I concede the point. My post should have read " the temperature differential across the blades"

It seems like feeding each head off of its own exhaust gas could set you up for some major complications. In the event of a partial system failure, half your cylinders would quit receiving boost.. Would a boost controller be able to open the gate on the other turbo fast enough to prevent catastrophic damage to internal engine parts? For race applications it might be worth it, but for just a sick car it seems kind of risky.

Something to consider -and im not sure how it applies in this situation, but worth mentioning- Dynos have repeatedly shown that in race motors with 3" pipes running from each bank straight to a muffler on either side, adding a crossover H or X pipe reliably adds 15-25HP by stabilizing the exhaust pressure across all cylinders.

 

[BUZZARD]

:rofl: No i have never put a turbo in the oven. I concede the point. My post should have read " the temperature differential across the blades"

I knew what you meant . We are using it in a race application and I will have a full ecu control strategy for component failure.

Chris

 

[B+O]

Ever put a turbo in an oven at 200F ? 500F ? Did it spin?

Chris

That's my Understanding as well...

Grumpy said everything i was gonna say . However, I've always believed this one true saying that works EVERYWHERE you try to apply it on an automotive...

"liquid and air move due to pressure differentials. Always from high to low"

a difference in temperature is, essentially, a difference in pressure. high pressure to low pressure. The lower the low pressure is, the more activity you'll have.

I'm a believer in Twin turbo over single turbo. I have my reasons, but mostly due to the "single is better" propaganda I see everywhere.

 

[Telion]

Excellent information thank you each of you. I think the first attempt should be to get enough pipe in and out of the engine bay in that respect then. Then if that fails then I can go down to twon in and one out so spool time didnn't suffer (not that lag is all that bad for our cars, but it would be a defeating purpose functionally to one turbo. And if that fails I can go one in and one out with 3" Pipe and Y's. Fail meaning not enough space without major work.

Reboticon thank you for the reminder I do need to start watching for X pipes as that is in the plan.

 

[gsridah]

a couple quick comments for telion...

I admire you for putting twins in your car, but i believe a single turbo would be easier to put in. Most of us that work on these cars are on very tight budgets so it would be better for us to see a single turbo because it would be more attainable for more of us. But either way you are cooking and I like the work....I look at this thread at least 6-7 times a day.

I know you know your ECUs...so im wondering what kind of managment system you are considering>>is this amount of boost something a chip can handle or will you need a stand alone managment unit.

And sorry to poop on your thread but one more question...
how hard is it to install standalone managment units?? I install car alarms, and all i do is find wires and jump into them, I imagine that managment units are similar to this?? <--by my previous questions you may have noticed i know very little if anything about this subject.

Thanks

 

[Telion]

No worries your not thread pooping, and believe me anyone who knows me knows I love building on a budget, and this project will be no different even with the twin setup and to be honest I have been comparing the price difference between twins and single turbos and really the difference is only a factor when you go with more expensive turbos, or double up on expensive blow off valves and wastegates, but if you stick to a budget (without cutting corners, bend but don't cut) then I would imagine this will be done at a fairly decent price, unless I end up having to get a stand-alone but the goal is not to just for that reason. Shopping and pricing ahead of time helps too.

I don't wanna comment too much on getting the ECU to support it in part because I don't know to what degree this will be successful other than to say that it has been done on a stock ECU so it should be a go. Great question, sorry for the vague answer. I will also be the first to admit I know very little about the AEM EMS but this would be the direction I would suggest going if you were to get a stand alone simply because there are members here with it who would be likely to assist you in troubleshooting or getting you started.

 

[GrumpySteelMan]

I think you want the twin turbo setup for the benefits of having no lag and boost at high and low rpms right? How will you accomplish this with two of the same sized turbos? My understanding of twin turbo is to have a small and a big one for those benefits.

As far as the compressor side of a turbo goes it is airflow and pressure ratio. A small turbo can create a high pressure ratio (boost) but at a much lower airflow (CFM, Lb/min, etc). So the idea is that two small turbos create the same pressure ratio and then you add their airflows together.

Can you explain why? I don't doubt you are right, i just dont know the reasoning. I would see it if each turbo were coming straight from the manifold of its respective head, but if coming from the Y pipe, why would it matter?

Two internally gated turbos whether the exhaust is merged or not share the intake tract and manifold. This is where they get the 'signal' for the wastegate actuator. So assume turbo 1's WG is set at 7psi and turbo 2's WG is set at 9psi. Once the boost passes 7psi and reaches 9psi the first turbo is running at reduced efficiency because the wastegate is closing in an attempt to regulate the boost pressure to 7psi, which isn't possible because of the second turbo.

Since you started picking apart the academia of the original post, I thought i would continue.. I may be wrong but this is what i noticed when i first read it.

It states PV=nRT where P= pressure,V = volume n= moles of gas, R = gas constant, .0821, T = temperature. It then reduces that to (PV)/T = constant. Should be (PV)/T = nR. It would appear to not be constant at all.

Once again, the post was more of a physics lesson than an applicable write-up.

I do have an important question though, running one 3" pipe in to feed the two turbos and another 3" pipe for the exhaust side of the turbos and splitting it back out, what would be the positives and negatives of this?

You probably want to keep the piping diameter on the small side and make sure to use heavy gauge pipe for the inlets. I would suggest 2.5" from each header and an X-pipe pre-turbo if packaging allows for it. I'd also consider wrapping those pipes, since you'll have relatively long runs between the engine and the turbos. You need to keep the temperature up!

3" for the exhaust side is probably about right.

All formulas aside, I was viewing it like this. Imagine a garden hose that splits into a T at the end, with each end of the T being a water wheel. If the radius of the T hose outlets are the same size as the T hose inlet, then if the inlet is flowing its max rate, each hose would still only be recieving half of the water.

This is a reasonable point. In some cases there are those that run smaller pipe diameters between the collectors and the turbos if there is a noteworthy distance between the two. Increasing the piping area on the far side reduces the temperature...and that's probably an oversimplification, but you get it.

Ever put a turbo in an oven at 200F ? 500F ? Did it spin?

Damn Son. That's all I can say.

I'm a believer in Twin turbo over single turbo. I have my reasons, but mostly due to the "single is better" propaganda I see everywhere.

I think that propaganda started with the Supra guys mainly. A properly set up twin turbo system is much more complex, in planning and execution, in most cases. Except of course if you need to for packaging reasons.

Still a single turbo is less likely to give you an ID-10T error on your engine management, if you don't really know what you're doing.

 

[reboticon]

Oh i see now, about the two turbos having the same wastegate settings. Naturally if they are sharing a common intake then pressure will equalize.:yes: So i guess, theoretically, it could be possible with some kind of crazy setup with backflow valves and dual pressure sensors but there would be no reason for any of that, just helps me visualize.

Telion, I hope you do go with the twin turbos, just because if you succeed, that will be an amazing feat of engineering on a budget. I think the whole single turbo being better thing mostly comes from the crazy amount of tinkering you will have to do to get them to work. Adding on 1 turbo is usually enough of a daunting task for most sane people!

Would it be possible to swap in a manual steering rack from a late 80s prelude or whatever honda offered it on? You could relocate the battery for the driver side turbo and get rid of the P/S pump + hoses to make room for the other?

 

[Telion]

Shouldn't be too crazy and I don't think I will need to go to that extent to be honest. At most I will relocate the battery. I ordered the bulk of the piping for the exhaust and the intercooler today. That will be the next step in seeing what I am really working with but as for space I think I will be close to good.

 

[BUZZARD]

Oh i see now, about the two turbos having the same wastegate settings. Naturally if they are sharing a common intake then pressure will equalize.:yes: So i guess, theoretically, it could be possible with some kind of crazy setup with backflow valves and dual pressure sensors but there would be no reason for any of that, just helps me visualize.

It is quite simple really, what we did was use a high precision air regulator. gave each waste gate actuator 10 psi and set with a 0.040 feeler gauge under each waste gate flap.

Our new setup will have pressure sensors on the compressor outlet and two boost control solenoids, the ecu will control each turbo individually.

Chris