Educate me on different set-ups

[AkumaNoZeta]

Ok, I understand the big single being best for peak horsepower applications. Now here's where I get confused. Twin turbo-using two smaller, quicker spooling turbos with lower boost. So lets say with an L6, you have 3 cylinders pushing 10 psi, then another 3 cylinders pushing another 10 psi, do they end up at the manifold as 20 psi? And the sequential has all the exhaust gas pushing like a 10 psi turbo and then all the exhaust gasses the leave that turbo enter a second 30 psi turbo so you run at 10 psi at super low RPMs until you get into the higher RPM range where you push the 30 psi? Can you get the peak horsepower of a single turbo set-up out of the sequential?

[1 fast z]

Stick two air hoses together from two seperate air compressors with each at 100 PSI. Guess what, you dont get 200 PSI, but rather 100 PSI. They do not add, but rather average out. If you had one at 100 and the other at 150, considering the SAME volume, you would end up with 125 PSI at the junction.

[AkumaNoZeta]

I've never worked on any turbo car in my life and reading about stuff without having any experience does make it harder to get it. So if you're running two 10 psi turbos in a twin system its still 10 psi? Whats the point then, just more volume? Running like a 10 psi turbo with a 20 psi turbo in a twin set-up would make more sense but then when it averages out it would just be 15 psi, the sequential set-up (if my explaination on how it works is correct) would be a better choice in my opinion.

[MONGO510]

A sequential system, if I recall, uses the first turbo to spin the second. This allows incredible boost pressures. I don't know of any gasoline engines using this setup. Detonation is the big problem. Diesels use this system to push 60+ lbs of boost. Tractor pull engines! Correct me please if this is being done on gas engines!

[jc052685]

I think this is where wikipedia may be of some help.

 

http://en.wikipedia.org/wiki/Twin-turbo

[jc052685]

A sequential system, if I recall, uses the first turbo to spin the second. This allows incredible boost pressures. I don't know of any gasoline engines using this setup. Detonation is the big problem. Diesels use this system to push 60+ lbs of boost. Tractor pull engines! Correct me please if this is being done on gas engines!

 

supra 2jzgte

[AkumaNoZeta]

Well, this is what got me thinking about this more http://forums.hybridz.org/showthread.php?t=145379 and I read the wiki description before post this thread. I just cant fully understand it. Was there a system that took the boost from the smaller turbo's compressor housing and ran it through the turbine housing of the bigger one? This is what I thought the sequential turbo system was but the Toyota one in the thread's link I pasted seems to make more sense as the sequential system.

[1 fast z]

RX7's and Fourth gen supras use sequential. How it works, is one turbo gets all 6 cylinders, untill a desired RPM, or boost pressure. Then a gate opens up and allows all the 6 cylinders to go through two turbos. This is why I built my twin turbo setup with twins rather than one big single. A correctly sized sequential setup will always have more area under the torque and HP curve than a big single.

[ShaggyZ]

1_fast_z is the only one who knows what's going on in this thread to date. However, he's not being very forthcoming (that's fine, by the way). I don't know a great deal about turbocharging but I can say that Grim should start with the "basics" of pressure and fluids (air) and then just with simple single turbochargers and the ideas behind them.

 

Compound turbocharging is NOT equal to sequential turbocharging - this hasn't been explicitly stated. Think of sequential turbos as steps, as 1_fast_z outlined above. Compound setups are still a bit elusive to me, but they basically reciprocate duties rather than waiting the duty to be handed off from the other (they do at times come on together but do not compliment each other). The compound setups feed each other.

[kiwi303]

A sequential system, if I recall, uses the first turbo to spin the second.

Correct me please if this is being done on gas engines!

 

Rover V8 cross bolt blocks, bored and stroked to 5.2 Litres, twin squential systems, one per bank. Used for a Hydroplane racing boat engine. 900Hp+

 

Ford Sierra 2.0L engine, squential turbo system, my old bosses Targa NZ motor for his Ford Escort panel van. 600Hp+

 

theres plenty of examples around, but they take serious engineering to make work properly.

[cygnusx1]

Two smaller turbos will spool much quicker than a single large turbo. Less lag.

[1 fast z]

Two smaller turbos will spool much quicker than a single large turbo. Less lag.

 

Actually no.

 

If you take one big turbo, that has the same EX volume in the turbine housing, as two twins, the single will spool faster. It has to do with the ineria of the turbines themselves. If you have two turbines that flow the same as a single turbine, the single will spool quicker. The mass of the two (inertia) is more than the single, given equal flow.

[Brad-ManQ45]

Actually, Hugh MacInnes book Turbocharging states that the combined polar moment of inertia of two smaller turbines is less that one bigger turbine capable of flowing the same amount of air...

 

The mass of the two may be more, but most of that mass is much closer to the center, whereas with the bigger turbine, there is a lot more mass further away, and when you are talking increases as distance squared, you pick up a bunch more PMI.

[WizardBlack]

Actually, Hugh MacInnes book Turbocharging states that the combined polar moment of inertia of two smaller turbines is less that one bigger turbine capable of flowing the same amount of air...

 

The mass of the two may be more, but most of that mass is much closer to the center, whereas with the bigger turbine, there is a lot more mass further away, and when you are talking increases as distance squared, you pick up a bunch more PMI.

 

Sorry, but that is completely dependent upon specific turbos. Mr MacInnes doesn't consider the bearing drag of two turbos versus one and countless other real world things. Two turbos that flow the same as one larger one is difficult to find. In the real world, you get something close. Some cars are a better fit for a single turbo and others for twins. This is from a PEAK HP point. In my experience, rotational inertia is less of a deciding factor as bearing design and finding turbos that fit the performance envelope of the engine.

A sequential turbo setup is nifty because the engine can run on one turbo to spool it at lower rpm and has some low down grunt. If you want the big power, you rev it up higher and it will open up both turbos. When you are rev'ing up to hit both turbos, you will have two spools; one as soon as you roll on the throttle (at low rpm) and another when the ECU opens up the second turbo. After that first gear, you just keep the revs up above that set point for the second turbo as governed by the ECU. Just as 1 fast z has said, some ppl prefer the sequential setup and some don't want the double spool, etc. Like most things, it's a preference, I suppose.

 

Sequential turbos are usually ditched in favor of a simpler system. Note that not all twin turbo systems are sequential. Some just have two turbos (I usually refer to them as parallel). Likewise, not all sequential systems use two turbos of the same size.

 

If you want to look at the Supra system, compare the HKS twin setup. It doesn't really spool any faster. It is more expensive (heavier) and cooler (to some, I suppose; I know someone who bought them for that reason).

 

Again, in real world, you have a limited set of turbo choices and sometimes twins are a better fit. IMHO, it is very rare that twins are better on an inline engine.

[letitsnow]

I've never worked on any turbo car in my life and reading about stuff without having any experience does make it harder to get it. So if you're running two 10 psi turbos in a twin system its still 10 psi? Whats the point then, just more volume? Running like a 10 psi turbo with a 20 psi turbo in a twin set-up would make more sense but then when it averages out it would just be 15 psi, the sequential set-up (if my explaination on how it works is correct) would be a better choice in my opinion.

 

I'll comment on this, volume is what makes horsepower, boost pressure is irrelevant. You can have 2 engines both flowing the same total volume of air, one has a ported head and a big cam, the other is totally stock. They both make the same horsepower, engine 1 doing it at 12psi and engine 2 at 25psi of intake pressure. The only time boost pressure is relevant is when it is the only thing being changed.

[thehelix112]

http://hyperphysics.phy-astr.gsu.edu/hbase/mi.html#cmi

 

Note the radius is always squared. E.g.

 

Radius1 = 40, I1 = M1.1681

Radius2 = 20, I2 = M2.400..... *2 as there are two smaller turbos: I3 = 2(M2.400) = 2.M2+800

 

So lets say the larger turbo wheel weighed twice as much as the smaller:

 

M1 = 2M2

so I3 = M1+800.. which is definitely less than I1, which = M1+1681.

 

Lots of assumptions, but hopefully that makes sense.

 

Re bearing friction, http://www.roymech.co.uk/Useful_Tables/Tribology/Bearing%20Friction.html offers a couple of equations for this, but both of them are linearr with regards to diameter. Therefore as the diameter increases, so does the coefficient of friction for the bearing.

 

But thats kinda besides the point as chances are relatively good that you can get two turbos with very similar sized shafts, and hence very similar sized bearings.

 

All this is starting to get very complicated. A large factor in the bearing friction is the axial or radial loads. I think the axial load in a turbocharger may be quite high, but how does this change between two turbochargers? Are two smaller turbine housings producing the same axial loads on two shafts?

 

Head hurts, stopping.

 

Dave

[AkumaNoZeta]

Yeah, my head is starting to hurt to. How about some book/dvd suggestions to learning so I can learn more from a start to end kinda way instead of this going back and forth that's happening

[AkumaNoZeta]

Know what? I'm just gonna design a new turbo. It will be so efficient that it makes 10 psi WHILE cranking, 15 psi at idle, 30 psi at 3000RPMs and 50 PSI at redline. I'll have it designed and manufactured by tomorrow, lol.

[letitsnow]

http://hyperphysics.phy-astr.gsu.edu/hbase/mi.html#cmi

 

Note the radius is always squared. E.g.

 

Radius1 = 40, I1 = M1.1681

Radius2 = 20, I2 = M2.400..... *2 as there are two smaller turbos: I3 = 2(M2.400) = 2.M2+800

 

So lets say the larger turbo wheel weighed twice as much as the smaller:

 

M1 = 2M2

so I3 = M1+800.. which is definitely less than I1, which = M1+1681.

 

Lots of assumptions, but hopefully that makes sense.

 

Re bearing friction, http://www.roymech.co.uk/Useful_Tables/Tribology/Bearing%20Friction.html offers a couple of equations for this, but both of them are linearr with regards to diameter. Therefore as the diameter increases, so does the coefficient of friction for the bearing.

 

But thats kinda besides the point as chances are relatively good that you can get two turbos with very similar sized shafts, and hence very similar sized bearings.

 

All this is starting to get very complicated. A large factor in the bearing friction is the axial or radial loads. I think the axial load in a turbocharger may be quite high, but how does this change between two turbochargers? Are two smaller turbine housings producing the same axial loads on two shafts?

 

Head hurts, stopping.

 

Dave

 

 

I don't believe that a wheel that flows twice as much would weigh twice as much, after all, we should be talking about flow, not size.

 

 

Also, I think you need a few more integrals, the weight changes as radius increases across the wheel. I believe this could make a significant change.

[letitsnow]

Here is some math to try and validate what I said above, any integrals could be ignored if we assume that the bigger wheel is a scaled up version of the smaller ones(they have the same basic shape). It's not optimal, but it should be close enough to compare without hours of measuring and calculation.

 

If we assume flow is directly proportional to area, twin 40mm turbos are equivalent to a single 57mm turbo(given by sqrt(2*r1^2)=r2).

 

Moment of inertia is given by I=(m*r^2)/2 for a thick rod about it's axis, that's not a compressor wheel but it is a scalar multiple of what the formula for a compressor wheel would be, so

 

40mm: I(total)=(m*800*2)/2=m*800

 

57mm: I = (m*1568)/2 = m*784

 

 

I would certainly bet that 2 40mm wheels would weigh more than a single 57mm wheel. Twins also have twice the bearing drag. If I've made any errors, feel free to correct me.

 

There are some reasons to go twins, but spool time shouldn't really be one.