Advertisement

Today’s Tech Tip: Triple Turbos

Interesting | May 31st, 2011 by 32
P90045747

I swore off turbo-charged cars after the mid 1980s, I’d had enough of the lag until 3500 RPM and then the boost would come on …

I swore off turbo-charged cars after the mid 1980s, I’d had enough of the lag until 3500 RPM and then the boost would come on with a rush (though it was fun), the nasty side affects of a turbo-charged FWD car pulling wildly under power when exiting corners, and the all-around amateur feel of the OEM turbo-chargers on affordable cars of that era. And then the siren song of the N54 bit me . . .

Twenty years after my last turbo-charged car, I bought another. This one equipped with BMW’s N54B30 (‘N’ is for ‘New’, 5 = the cylinder count (no, it’s not 5, but because 8 cylinders is represented by BMW as a ’6′, then 6 cylinders has got to be something else, in this case ’5′), ‘B’ = ‘Benzine’, and ’30′ = displacement in deciliters). It also employs direct gasoline injection and two tiny turbo chargers in parallel.

If there’s one knock on this type of turbo setup, it’s that the twin turbos have to be throttled back at higher engine RPMs in order to avoid spinning themselves to death. Those little turbos have to spin much too quickly at higher engine RPM to produce the same boost pressures they did at lower RPM. And that’s why the torque curve on the N54B30 is a tall mesa between 1400 – 5000 RPM, but falls off the face of the earth afterwards.

Todays Tech Tip: Triple Turbos

BMW TwinPower Turbo shown here

So, how can that be fixed, we’ll take an approach known as compound supercharging. In this case, utilize the two small turbos for low RPM operation and employ a third turbocharger for higher RPM operation. A BMW tri-turbo engine can be done in a couple of ways.

One is to plumb the turbos to operate sequentially. This will run the smaller turbos up to a specific engine RPM, boost level where a transition to the third turbo will take place and the intake track is
supplied from the output of the bigger, third turbocharger at high RPM. The drawback to this system is the Rube Goldberg plumbing necessary to make it work (and the system tweaking needed to make sure the transition is seamless).

Another way of accomplishing it is through staged turbo-charging. This would use two turbos (or a combination thereof) of similar size to provide a multiplying effect on boost. If the first (set of) turbo(s) produces three times compression of inlet air and the second turbo is fed the output of the first stage, and then boosts that three times further, you’ve effectively squared the boost (the turbos deliver a total of nine times the input pressure). The drawbacks of this setup are lag, and an unresponsiveness to fairly rapid RPM changes. This type of turbo-charging was used in piston engined aircraft.

So of the two types of compound turbo-charging the type that makes the most sense for an N54 derivative would be the sequentially engaged turbo-chargers, if they can package the plumbing needed to do it in the existing space.

Now if they just made the next M3 FWD, they’d be able to cram the plumbing for the sequential turbos in the nose with no problem . . .JUST KIDDING!!!!

  • Pingback: Today’s Tech Tip: Triple Turbos | BMW Blog

  • Manny Antunes

    Excellent write up Hugo! 

  • Nnnn

    Hugo, I remember Horatiu talking about a quad-turbo I4 a number of times way back, but it wasn’t clear where that rumor was coming from or what the technical reasons would have been for it.  It’s definitely interesting, but I still don’t see many turbos as an effective strategy.

    While the small turbos help mostly at low RPM, I think their lighter mass accelerates the overall spin-up so it’s really about responsiveness.  The N54 use that small turbo + big turbo design.  So, the small turbo spins up faster, giving quick response and — in turn — creates more exhaust sooner to spin up the large turbo sooner.   I’m guessing that the small turbo is quickly reduced to an average speed so that it can respond to further throttle blips, meanwhile the large turbo pushes the real volume.   And then you have the twin-scroll turbos, which spin up more quickly anyway, thanks to more even exhaust flow enabling a lighter spindle/fan.

    It seems like they already have what they need to get responsiveness and maximum boost, it just requires some fine tuning, including adjusting size/mass of the turbos, the plumbing, and ECU ramps.  What does 3 (or 4) turbos get you beyond this?  What does two same-sized small turbos get you, versus having a small/medium/large configuration?   What about the loss in efficiency from driving more turbos from fewer cylinders (less even exhaust flow per turbo) ?

    I still think they need to use a single twin-scroll turbo and replace the small turbo with a small supercharger for immediate, linear response.  Someone please nudge BMW about this.

  • Nnnn

    Hugo, I remember Horatiu talking about a quad-turbo I4 a number of times way back, but it wasn’t clear where that rumor was coming from or what the technical reasons would have been for it.  It’s definitely interesting, but I still don’t see many turbos as an effective strategy.

    While the small turbos help mostly at low RPM, I think their lighter mass accelerates the overall spin-up so it’s really about responsiveness.  The N54 use that small turbo + big turbo design.  So, the small turbo spins up faster, giving quick response and — in turn — creates more exhaust sooner to spin up the large turbo sooner.   I’m guessing that the small turbo is quickly reduced to an average speed so that it can respond to further throttle blips, meanwhile the large turbo pushes the real volume.   And then you have the twin-scroll turbos, which spin up more quickly anyway, thanks to more even exhaust flow enabling a lighter spindle/fan.

    It seems like they already have what they need to get responsiveness and maximum boost, it just requires some fine tuning, including adjusting size/mass of the turbos, the plumbing, and ECU ramps.  What does 3 (or 4) turbos get you beyond this?  What does two same-sized small turbos get you, versus having a small/medium/large configuration?   What about the loss in efficiency from driving more turbos from fewer cylinders (less even exhaust flow per turbo) ?

    I still think they need to use a single twin-scroll turbo and replace the small turbo with a small supercharger for immediate, linear response.  Someone please nudge BMW about this.

    • JohnJay

      The N54 uses parallel turbos, both somewhat small in size.  Each is fed by exhaust from 3 cylinders and are manifolded together on the intake side.  And the twin-scroll, as used on the N55, is a single turbo with split exhaust housing.  This keeps the engine exhaust plumbing split into two banks which aids in cylinder scavenging as exhaust pressure from one cylinder doesn’t create back pressure on another.

      Neither idea presented in this article is a very good one.  The parallel small turbos in sequence with a larger turbo does create a nightmare of intake and exhaust pipework.  One would need no less than three valves to operate such a system.

      As for staged turbochargers, they work well in aircraft applications, not so much automotive. An aircraft engine tends to operate at a constant engine speed and high load.  This covers up the fact that such a system would have terrible throttle response.  It also doesn’t work as described.  The first turbo in the chain takes power from the exhaust stream, leaving less for the second.  You also don’t want the absurdly high inlet pressures you could create.  A piston aircraft at 40,000 feet has to deal with 2.7 psi atmosphere (as opposed to 14.7 psi).  At ground level, you’ll always have >10 psi atmospheric pressure, unless you are competing in the Pikes Peak International Hill Climb.

      • http://www.facebook.com/profile.php?id=100000477069969 Hugo Becker

        Yup!

      • Nnnn

        Why would staged turbocharging have lousy throttle response?   Actually…. I don’t think these are staged, technically, because they are – as you say – blown in parallel.  Their behavior is staged (in effect) but there’s no valve system opening up additional turbos per RPM or throttle demands.   Plus, there is the effect of the small turbo’s earlier boost creating more exhaust pressure for the larger turbo — earlier than it would get on its own.

        The scavenging idea I haven’t heard before.  That’s a better explanation of the need to have even exhaust flow (pulses) over the turbine, and explains things like flat-crank V8s.  But doesn’t the I6 have even exhaust pulses to begin with (given timing/firing order) ?   And on a twin scroll system you’re merging them into the same turbine anyway, unlike the N54 (or NA design) with completely isolated exhaust systems.    My understanding was that the twin scroll design lessened asymmetric pressures on the turbine, spindle and bearings, allowing it to be made more lightweight.

        • http://www.facebook.com/profile.php?id=100000477069969 Hugo Becker

          Your boosting boost, so any throttle reduction will require more time to bring it up to the multiplied boost levels you were at previously.

          Yes exhaust pulses and scavenging are important (and more of an dark art than science). The X5/X6 M turbo layout mimics a flat crank V8 (as do the old ‘bundles of snakes’ exhaust or the infrequently seen 180 degree exhausts.

          The I6 has one power stroke per 120 degrees of crank rotation. And that’s nigh on optimum (besides being dead smooth).

  • Nnnn

    Oh, and I loved your FWD ///M idea.  BMWblog should definitely run with this rumor.   It’ll be fun :)

  • PIERRE ROSSOUW

    hi a n54 is a six cylinderas the n55 is only a improve ment the nxx is only a engine code no a designation of cylinder count

  • PIERRE ROSSOUW

    hi a n54 is a six cylinderas the n55 is only a improve ment the nxx is only a engine code no a designation of cylinder count

  • Pingback: Today’s Tech Tip: Triple Turbos | labmw.com

  • Ali Guermachi

    Does really the turbo damager the engine. And the engines without turbos last very long compared to those with turbos?
    Thx.

  • Otto

    Quantity is the enemy of quality.
    Piling up turbos is not the solution. (increased moving parts, more intercoolers, more heat to dissipate, etc)
    VW found the solution years ago : supercharger + turbocharger : one for the low revs and the other for high revs.
    Well, they weren’t the 1st ones to come up with this setup (I believe Gemballa released a Testarossa with such duo in the 80s) but they were the 1st to mass produce it.

    • Ali Guermachi

      Does really the turbo damager the engine. And the engines without turbos last very long compared to those with turbos?
      Thx.

      • Otto

        Turbo Diesel engines are less reliable than N/A ones for sure – although there is not much N/A Diesel engines in the market if none.
        The new downsized petrol engines are quite new for a long term comparison but as I wrote, more moving parts = less reliability.

        • Ali Guermachi

          Thanks. I’m quite new as a BMW lover :) I leave in Europe, and here the diesel engines are farly used. I think to buy a BMW, but I worry about realibility. So I hesitate between, diesel or petrol engine. Now ther’s also petrol engines with Turbo. Are they as reliable as the petrol engines without Turbos? And why Turbo diesel engines are less reliable, ’cause I thought that Turbos give more power for less component: you can buy a 4 cylinder turbo charged engine that deliver the same if not more power than one with siw cylinders and without turbo.
          Thx in advance

      • http://www.facebook.com/profile.php?id=100000477069969 Hugo Becker

        The reliability issue is with the speed of the turbine on small turbo gas engines (not the engine itself). Rod Millen in his 1999 or 2000 Pike’s Peak run (can’t remember which at the moment and to lazy to do a search ;-) blew up his turbo mid-way up the hill. Turns out he wasn’t using a wastegate to limit boost, but rather an optical sensor that got oiled down and the turbo just spun itself up to stratospheric RPMs and committed harakiri (oh my!! ;-). As David Hobbs sez, “it blued up”.

        Are turbo diesels less reliable than NA diesels? I wouldn’t say that, but rather a variation on that. Turbocharging in gasoline or diesel engines does introduce more stress on the engine and if the manufacturer doesn’t account for that, they will be less reliable.

        And for anyone wondering why diesels don’t rev as high as gas engines, it’s because diesel fuel has a slower burn rate than current gasoline – and that limits max RPM if you want decent fuel economy. Diesel also gets better mileage than a gas engine, but the bulk of that is due to diesel’s higher energy density than gasoline.

        Now, I mentioned that diesel burns slower than current gasoline. Gas wasn’t always as good (or of a relatively uniform quality) in the firat half of the twentieth century as it is today. In fact BMW built and used a diesel radial aircraft engine for that reason. It never got into full production because of significant improvements to the quality and burn rate of gasoline in the late 1920s, early 1930s.

        • Ali Guermachi

          And than I go for Diesel or Petrol? :)

        • Ali Guermachi

          I think the common rail systems are the main cause of the reliability problem in diesel engines. The fact that the pression is huge on the engine, it would’nt last long. Does it true?
          ++

        • JohnJay

          The diesel cycle is inherently less efficient than the Otto cycle, and diesel fuel has less energy density (a lower heat of combustion).  The reason that diesel engines achieve greater efficiencies in practice is because they can handle much higher compression ratios.  A spark-ignition engine would knock if it were run at compression ratios anywhere close to what diesels are run at.  Drop the compression ratio on a diesel down to that of a gasoline engine, and it will have less thermal efficiency.

          • http://www.facebook.com/profile.php?id=100000477069969 Hugo Becker

            Double check the energy density of diesel. It has greater energy density than gasoline, however, you hinted correctly that diesel is less volatile than gasoline and that’s why diesel can be squeezed to higher pressures without blowing things up.

            Check this out: http://en.wikipedia.org/wiki/Gasoline

          • JohnJay

            The table in the linked Wikipedia article is a mess.  It takes numbers from different sources and is inconsistent about them.  Gasoline has greater higher heating value (HHV) and lower heating value (LHV) than does diesel fuel.  These numbers are in MJ/kg, so depending on the density you assume for the two fuels, you can calculate a greater MJ/L figure for diesel.

        • Otto

          About Turbo Diesel engine reliability, it’s a fact that statistics show more problems with them than N/A engines, petrol or Diesel. Back in the 80s, it wasn’t rare to see a Diesel MBs with 500K+ kilometres proudly displayed by the odometer.
          About Diesel itself, it’s a fact that it’s more efficient than petrol : for a given output, they use less fuel and therefore have lower CO2 emission. Only issue was the NOx emission, solved by the NOx filter.
          Lower revs is not an issue, Audi and Peugeot proved it with the R10 TDI and the 908 HDI.

    • http://www.facebook.com/profile.php?id=100000477069969 Hugo Becker

      Probably done in an aviation engine around the time of WW II. I’d have to do some digging, but I seem to remember compound supercharging mentioned in that time frame.

      One reason pre-war European race cars used Roots type blowers (superchargers) was the wide RPM range they ran at. In America, the centrifugal supercharger (developed with assist from Dr. Stanford Moss of GE) showed up in the ’20s on the board tracks and Indy. The centrifugal charger was less effective at low RPM but then the race cars on the board tracks only were at low RPMs during pit-out.

      • Nnnn

        Why would they be less effective at low RPM?   It would be a fixed ratio to the engine RPM, and presumably a ratio affording maximum boost at any particuular RPM.   They should certainly be better than turbos at low RPM.

        • http://www.facebook.com/profile.php?id=100000477069969 Hugo Becker

          Because of the size of the centrifugal blower – the blowers on the old board track cars were basically the intake side of a turbocharger (there are superchargers available now like Paxton that are centrifugal). They needed maximum charge air at maximum RPM (which wasn’t all that much in the ’20s and ’30s.

  • FanofBMW

    FWD M3…. LOL! *Jumps off cliff*

  • Pingback: RoundelTable 57: Nanny Nanny Boo Boo, Better Luck Next Year

  • Pingback: Rumor: New BMW M3 with tri-turbo engine and 450 hp

  • Pingback: Rumor: New BMW M3 with tri-turbo engine and 450 hp | labmw.com

  • Pingback: Rumor: New BMW M3 with tri-turbo engine and 450 hp | BMW Blog

BMWBLOG

NEWSLETTER