Borg Warner R2S Image Repo Thread

[Fred]

I'll let the pictures do the talking, but I LOVE this setup and valving and badly want to do one on a small petrol engine and get torque everywhere and epic spool etc.























Enjoy! :-)

Fred.

[Nick_nl]

Looks impressive, but you might also look into the Lancia Delta S4.
They combined a turbo with a supercharger, the supercharger did the lower rpm, the turbo the higher RPM. This gave no turbo lag.
The only issue was reliability, or the lack of it. Volkswagen does use this technology in current models and have made it reliable.

[Fred]

Sure, however all super-chargers suffer from terrible efficiency which turbos do not. Thus I'd still prefer the above. There is a sweet Honda twin charge setup around. Maybe I can find a link for it :-)

[thebigmacd]

The check valve setup adds unneeded complexity in a compound turbo setup. If they just had the larger compressor feeding the inlet of the smaller compressor it would work just fine. There is no need to bypass the smaller compressor effectiveness-wise.

The advantage to the check valve I suppose could be piping packaging.

[Fred]

And much less restriction at high flows! With the small turbine bypassed the small compressor is not compressing and is just a blockage in the pipe. And a low diameter blockage at that. Without the small turbine being bypassed you're going to over-speed it. Without the turbines being inline with each other you're going to get a nasty two-phase response instead of a nice blended response. I have to say, I'm completely sold on the borg warner setup, and have been for 5 or so years. It's an elegant and efficient design. The extra complexity is negligible really, and serves the purpose perfectly. Ignore me :-)

[thebigmacd]

You would think that the smaller turbine is a restriction and will be overspun, but in actual fact (1) it still performs some work as the wastgate does not divert 100% of exhaust gasses around it, and (2) the volume flow of fresh air through it is pre-compressed and as such the actual velocity of the turbo is not nearly as high as one would assume for a given mass flow rate. Multi-stage compression with intercooling between stages is the absolute highest-efficiency method of compressing gasses, including industrial compression. The sweet spot is roughly 2:1 "size" difference between the turbos.

The big gotcha is that a turbo cannot be overspun from pushing extra air through the compressor, only through the turbine, for the same reason that the aspect ratio of the compressor housing makes negligible difference in spool time, surge resistance, or flow capacity. You can run an engine without the turbo spinning at all.

Because the small turbo is still performing work, it adds "negative resistance" to air flow via suction. It is actually trying to pull more air than is being fed to it. All without overspinning.

[thebigmacd]

I'll add a second post so I don't keep editing the first...

Turbo maps are always shown in lbs/min vs pressure ratio, but in actual fact this is only true when the compressor inlet is at standard temperature and pressure. So at first glance, you would think "hey, if I shove twice as much air through the smaller turbo as it is rated for, it will overspeed". In actual fact, if you precompress the air to 1 bar at the inlet, and cool it to standard temperature, the lb/min capacity of the turbo at any given point in its map has actually been doubled.

Why is this? Because the density of the inlet air has been doubled. For a given volume flow rate and pressure ratio the mass flow rate will now be doubled. The efficiency/rpm coordinate on the compressor map must now be referenced at half of actual mass flow.

So a turbine with inlet pressure of 1 bar, outlet pressure of 2 bar, and flow of 50 lb/min will actually be sitting at 25 lb/min/pr1.5 on the compressor map. Adjust for inlet temp as needed of course. So now you can see why intercooling is so important: if you can get the inlet temp down to ambient temp you have doubled the mass flow capacity of the smaller turbo by feeding it with a larger one.

This is where the term "intercooler" came from...most "intercoolers" are actually aftercoolers. Real intercoolers are between compression stages.

The only way to effectively gauge how a turbo will flow with compressed inlet air requires the compressor map to be in terms of volume flow, not mass flow. Mass flow is handier with single stage calculations of course.

[Fred]

Yes, you've understood from what I wrote that I don't understand what you've now written. That was a false assumption. What you've written is all true, but in no way negates what I said. Except the part about positive work from the small turbo once the turbine is bypassed, which I'd argue meant your bypass was really shit.

[thebigmacd]

Yes, you've understood from what I wrote that I don't understand what you've now written. That was a false assumption. What you've written is all true, but in no way negates what I said. Except the part about positive work from the small turbo once the turbine is bypassed, which I'd argue meant your bypass was really shit.

If you don't want the smaller turbo to be a restriction on the compressor side, you have to allow some exhaust to flow through its turbine. As long as the wastegate on the large turbo is open, there is excess exhaust energy available to use to keep the smaller turbo spooled. Yes you have less pressure available to the larger turbine, but now it has to do less work because the smaller turbo is doing some of it. The magic is in the equilibrium between the two.

Good intercooling allows the most efficient compression possible when both the large and small turbo are operating in their most efficient range...which is rarely attained when one turbo is doing all the work. Lower pressure ratios, lower shaft speeds, more efficient compression, etc. Peak efficiency is at the middle of the map, not the top right.

[Fred]

Yes, I think most people on this site can read a turbo compressor map just fine...

I think most people on this site would also rather that they were in the upper right at peak RPM/Boost. If not, you've got something laggier than need be. Which is the entire point of the above design. Minimal sized turbos for minimum lag and maximum output in the same package. The point isn't to obtain high pressure ratios, it's to obtain high pressure at low rpm and low restriction at high rpm. It's NOT a compound setup.