Turbo engines

[Tony]

I believe turbos have been used for air cycle cooling by using the turbine as an expander, and a much smaller than normal compressor to further boost the air going into the intercooler.
As far as I know it has only been used in constant speed, constant load applications.
Specifying the rotors and housings would be a fairly complex undertaking, because all the operating conditions would be far removed from normal turbo operating parameters. Normal flow maps would be useless.

While theoretically possible, I can also visualise all kinds of dynamic problems with "lag" and suddenly changing pressures and temperatures when fitted to a sporty type engine and driven seriously.

The advantage of using a mechanical screw type expander are that it is geared directly to the crank along with the supercharger and is essentially locked to engine rpm, giving much faster and more predictable response to speed and throttle changes.
Kind of positive displacement supercharging in reverse.... (undercharging ??)
I would expect air cycle cooling would be much simpler to design an implement than trying to do it with a turbo.

[jimbo39]

would it really be worth all the effort for a street motor though tony?

[Tony]

Probably not.
But for a competition engine operating in extreme heat, it could make for an interesting development path.

[tt911er]

Greetings from Finland!

This is from latest Porsche 997 gt2 intake:
"The expansion intake manifold uses the principle of oscillating air in the intake manifold during the cooler expansion phase, keeping the temperature of the fuel/air mixture lower than in the 911 Turbo. This, in turn, means a significant increase in all-round efficiency, with fuel consumption down by up to 15 per cent under full load despite the increase in engine output."
Could someone explain how this works?

As a first idea was planning to put air/air coolers channelled inside rear fenders/bumper like latest 911 turbo do but that would need rear fenders to be widened, maybe to be widened anyway... After SC room inside engine bay is limited so there I need to go for a air/water and use that for turbo cooling as well. Expansion intake sounds interesting, but I have to admit I don't understand how it works.

[jimbo39]

I don't fully understand it myself, but basically, when you compress air it heats up (we all know this), the reverse happens when it is allowed to expand & when air is cooled the pressure drops. This is why you get pressure drop across an intercooler, it allows the air to expand AND cools it at the same time.

I'll leave the detailed explanation to Tony

[Tony]

That doesn't really sound like an engineering description written by a real Porsche engineer.
More like sales and marketing hype written by some young business school graduate totally high on yippee beans, ego, and weed.

The more I think about it, six of those Laminova air/water cores (each rated for 150Hp) located in the induction plenumns would be the way to go.
There could then be multiple small heat exchangers located wherever they will fit, to cool the hot water.
Getting the heat into the water is only half the battle, probably the easiest half.

My own once only experience with air/water systems is that they tend to rapidly heat soak, and are then very stubborn things to get back down anywhere near ambient once warmed up.

[tt911er]

About the backpressure. Have anyone tried to keep Bp down by using external wastegate for that purpose and controlled by Bp. I have played idea of using just spring loaded (adjustable) old exhaust valve. Of course it need to be tested when there is enough energy to spin the wheel and over that pressure it will start to open? Depending on power level, size of the valve may need to be matched. Of course "normal " wastegate can be used if it is big enough and ecu can read the Bp.
Boost pressure is wastegate controlled as usual but this would be top of that, just releasing the extra exhaust energy that turbine is unable to flow?

[Tony]

During spool up, you need all the available exhaust flow and pressure to get the turbo accelerating up to full boost speed as quickly as possible.
Bleeding off exhaust is going to reduce both back pressure (good) and boost pressure (bad) and slow down the turbo response (very very bad).

With twincharging the exhaust turbine a/r will be quite large, because the exhaust pressure ratio will be low.
A lot of flow but without much total exhaust back pressure to drive that flow.

The wastegates will also need to be unusually large, because they too need high flow, without there being much pressure to drive that flow.

If you expect to reach full boost at say half full rpm, then you can assume that at full rpm the wastegates will be flowing roughly about the same gas volume as the turbines.
If you plan to reach full boost at a third of full rpm, at full rpm the wastegates will be flowing twice the exhaust volume of the turbines!
That situation is going to be very easy to achieve with twincharging, and its a very nice situation to have.

Source your turbos first, then eyeball the throat area of the exhaust housings, and then get wastegates with an appropriate flow area to do the job.
They are going to be super sized wastegates.
What we are doing is rather different to just straight turbocharging, all the turbo and wastegate pressures are quite low, but flow everywhere is going to be relatively high.

[tt911er]

About the backpressure was more general thinking about the issue. Bleeding exhaust flow BEFORE target boost is very bad, that can happen also with poorly controlled wastegate if you allow the increasing pressure to open (even fraction of opening can hurt a lot of spooling) WG before target.
That "extra" WG was mentioned to start open in area where full boost is achieved but Bp is still going to rise towards redline. I believe there is something to be found to get better Boost/Bp ratio, specially when only turbo charged engine wanted to have wider power band/ faster respond

Now back to twincharging. Tony, I'm not shure how you meant the supercharger relief valve (if that was the right name for it) to be connected? The charger I have has quite large (48mm) valve between chargers intake/exhaust. That will naturally release the charger to more or less freely rotate (windmilling) but it will increase temp every rotate that air is circulating, right? In twincharging it is more critical because air is already compressed/heated before it will arrive to SC -> cooling the turbocharged air need to be taken even more seriously.
Do you want to dump that supercharged air just to atmosphere (out of the whole induction system , just like most aftermarket dumpvalves do) because that amount of flow is just extra air you don't need/want on THAT moment?
Wondered first how Audi ended up using that size of the butterfly valve? But after calculating that 300 hp/450cfm flow, the 48mm orifice will mean more than 380 ft/sec! No need to wonder anymore!
Leading back to my project. Need to figure is it easier to fit two smaller relief valves or one quite a large one?
All the parts (pluming etc.) seem to grow when working that amount of flow, funny how that will surprise time after time

[Tony]

Once the turbo has accelerated up to rated boost pressure, the waste gates starts to open.
If you look at the speed lines on a compressor map, they are almost horizontal with only some slight droop.
That means that flow can increase substantially without any more turbo shaft rpm.
Between just reaching rated full boost, and full maximum power, turbine rpm might only have to increase slightly to compensate for the extra compressor shaft loading on the turbine, as lbs of air per minute increases.

Beyond initially spooling up to rated boost, virtually all the extra exhaust flow will be dumped through the wastegates, and the wastegates need to be large enough to handle that flow.
If they are undersized, boost will rise uncontrollably as the turbines will be forced into overspeed.
You don't need anything extra besides the wastegates themselves.

The wastegates you end up with will be able to control even better than on a straight turbo engine.
The reason being that the diaphragm and spring will be operating from 20 psi total boost pressure, but the poppet valve only has to fight against 8psi exhaust back pressure. On a turbo engine these pressures are usually about the same (if you are lucky) or exhaust back pressure very often exceeds boost pressure.
Wastegate control will definitely not be an issue with twincharging, just make sure the wastegates have sufficient flow area compared to the turbine housing flow areas.
Just looking will tell you if they are going to be large enough.

Eaton build a butterfly bypass into the supercharger itself, driven by a pneumatic actuator, which in turn is controlled by the electronic engine management unit.
That is very convenient for the vehicle manufacturers because the bypass will be under software control and they can do all kinds of clever things with it.
No boost with a stone cold engine, no boost if engine overheating, no boost if detonating, no boost in reverse gear, that type of thing.

For us, controlling that actuator becomes a bit more difficult, and it also has the disadvantage that it recirculates the air directly within the blower casing which can lead to temperature buildup and a hot running blower if run for lengthy periods at small throttle high engine rpm, as in freeway driving.

Some people have made it work, and others have not. I have not had much success with it myself.
The problems I had were very non linear flow versus opening angle, and a tendency to suddenly flop open or flop closed, or flutter.
But then again factory cars all use it, and some guys say it works just fine for them....
I just give you what is based on my own personal experience and very biased opinion for what it is worth.

I have had much more success with using an external turbo wastegate fitted with a very light spring as a supercharger bypass. It responds instantly, and gives very sensitive, smooth, proportional and stable control.
Everyone that has tried it has fallen in love with it, and it also has the advantage that you can bleed the air back to the blower intake from after the intercooler solving the blower overheating at small throttle problem.

The air bypass wastegate will need a very light spring that you will need to find through experimentation.
Something that opens the wastegate at maybe around 2.5 to 3.5psi actuator air pressure.
You can do it with a shorter stiffer spring that opens at 3psi, or a very long soft spring that opens at 3psi.
This changes where, and how quickly boost rises compared with accelerator pedal position.
Its a drivability issue, and wide open to personal preference.
I prefer to have no boost at all, normal n/a operation up to about 75% throttle, then very rapid proportional boost increase up to full boost at 100% throttle. This gives excellent fuel economy, is easy to drive and is a lot safer in the wet.

The two air lines connecting to the wastegate diaphragm (one above, one below) connect to either side of the throttle butterfly. It's as simple as that.
A high differential pressure developed at closed or partial throttle opens the wastegate and bypasses the supercharger.
At full wide open throttle the differential pressure will be zero, and the wastegate held fully closed by the spring, and you get full boost.
The choice of spring you fit, determines how much boost you get at various throttle openings, and it's a really fun thing to experiment with.

The supercharger bypass does not need to be very large, it's just a boost leak around the supercharger. There is no requirement for it to go to absolutely zero pressure drop when fully open. Anything less than about 0.5psi back pressure has essentially totally unloaded the blower. A 32mm wastegate should be more than sufficient to totally unload the blower at freeway rpm which is all you really need to have.

The only thing you need to be absolutely sure of is that the wastegate you use as a supercharger bypass has sufficient diaphragm area compared to poppet valve area. We only have about 3psi control pressure to control about 8psi of blower pressure. The control diaphragm needs to be about twice the poppet valve diameter, giving about four times the area.
Any area ratio less than that and you will not be able to find a spring that will be both heavy enough hold the wastegate shut against full blower boost, and light enough to open at part throttle differential pressure of around 2psi to 3psi.
The larger the ratio between diaphragm area and poppet valve areas, the wider and less critical the spring choice becomes.

This also imposes a limit on supercharger boost pressure. If you tried to run say 15 psi blower boost, it will probably just blow the wastegate bypass open. You will then need a much heavier wastegate spring which will keep the wastegate shut (and produce sudden high boost) at very small throttle openings.
That might be o/k, or it might be very hairy to drive with a very powerful engine on a slippery road.

So choose your blower bypass wastegate (and spring) with care and it will provide wonderfully smooth and controllable throttle response and power delivery.