Pressure vs flow 1

[Mark911]

I'm trying to understand why some say that on the same exact engine a turbocharger which has a high flow capacity at a relatively low pressure ratio actually flows more air than a supercharger at the same boost level? I understand how it can produce more power as other variables come into play. I understand that the pressure/density is generated in different manners (internal vs external compression) and this affects the volumetric and thermal efficiencies. However, everything else being equal (air temps in the manifold via intercooling, pressures, etc), how do you explain this statement?

 

[patprimmer]

I can't explain the statement and therefore I do not accept it until I see a satisfactory explanation.

I would expect that a turbo charger would heat the air less and therefore flow a higher mass of air at the same pressure, but if the temperatures and pressures were equal, WHICH THEY WOULD NOT ACTUALLY BE, then the airflow would be the same as it would be controlled by flow through the head. If anything the supercharger would flow more air as more boost would be lost through the exhaust valve at TDC overlap, as the pressure differential between the inlet and exhaust manifolds would be higher due to the exhaust pressure being lower. Therefore more air would be required from the supercharger to maintain the same temperature and pressure.


Regards

eng-tips, by professional engineers for professional engineers
Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[Warpspeed]

As Pat already has pointed out, a supercharged engine runs with minimal exhaust back pressure, whereas a turbocharged engine will always develop considerable exhaust back pressure with which to drive the turbine. If you had two identical engines running with identical boost pressures and induction temperatures, the supercharged engine will always end up having higher mass airflow and less thermal stress.

Throttling the exhaust reduces engine flow just as effectively as throttling the induction.

Both engines run at a disadvantage. The supercharger subtracts mechanical drive power from the crank.

The turbocharged engine runs with very high exhaust back pressure which can be extremely detrimental to engine operation in a whole variety of ways.

 

[mattsooty]

One point to note is that, there should be very little or no valve overlap in this type of engine application.

Overlap is used more in NA engines, since a degree of scavenging is required to allow for higher VEs. In a FI engine there is no need for this as you have a higher pressure in the intake manifold to fill the chamber with.

In the comparison stated there should be equal amounts of torque (power) produced if all other variables remain equal. Especially:
- actual engine load (airmass/stroke)
- spark timing
- AFR
- Intake air temp

As far as engine load goes, this is self compensating for the back pressure of the turbocharger/exh manifold. If this backpressure becomes great enough that there is significant amounts of internal EGR displacing the fresh charge then the actual airmass consumed will fall - in which case everything ceases to remain equal and you are comparing apples with oranges!

MS

 

[globi5]

As patprimmer already mentioned if pressure, temperature and humidity are the same so is the massflow (airflow). (The question is not what produces more power or is it not?)

The turbo has a centrifugal compressor which is adiabatically more efficient (less temperature increase and therefore higher air density at the same pressure (= higher massflow)) than a mechanical supercharger such as a roots blower (which was originally not intended to be used as a compressor anyway), G-charger (VW) or even a screw type compressor.
However, if you were to use an axialflow compressor (supercharger) which has a higher adiabatic efficiency than a centrifugal compressor you'd end up with a higher massflow than you would with a turbocharger at the same boost level.

http://www.axialflow.com/

 

[Mark911]

I understand that a turbo is thermally more efficient, however, my original assumption is that the blower (or even a less efficient turbo) can recover the lost density via intercooling making an apples to apples comparison more realistic. So, an environment at X temperature and Y pressure captured within a manifold of Z volume should have the same density regardless of what device created that environment. However, as usual were not talking about a static condition. Therefore, I was curious if anyone knew of some unique characteristic of a turbo charger that would allow it to deliver more mass given the same manifold conditions (at least as measured by our crude methods) than a blower under dynamic conditions. Granted, boost is simply a bulk measurement of the average pressure in the manifold at any given time. Could it be that the turbo has a greater ability to “respond” flow wise to transient conditions existing under actual operating conditions? Or maybe the turbo creates mass flow of higher velocity and therefore can fill the cylinders faster. Both would result in a higher VE without resorting to mechanical modifications (porting, bigger valves, etc) and therefore “flow” more at the same boost levels. I’m stretching here, but the flow vs. pressure theory is touted by some pretty big names in the tuner business and I need knowledge to dispel any untruths.
These same “experts” will argue that the turbo of higher flow vs. pressure ratio will flow more mass than a smaller turbo at the same (or lower) pressure ratio on the same engine (thermal & volumetric efficiencies the same). To me, if the smaller turbo produces enough flow to produce X pressure the larger turbo will produce the same numbers but it won’t be working as hard in terms of turbine speed. I would expect an increase in HP simply due to less backpressure of the larger system (along with greater lag), but not due to more flow.
I base this on the model of an engine as a highly dynamic “variable” leak. The forced induction device simply tries to “fill the leak”. Any flow not consumed by the leak is manifested as excess pressure which increases the density of the air in the manifold which in turn increases the overall mass flow rate. I’m still a bit unsure if the actual port velocities change much due to the increased pressure differentials (you would think so based on flowbench data at different inches of water), but to a degree it would depend on how quickly the pressure is recovered inside the cylinder at various piston speeds. Anyway, exhaust backpressure aside (on turbos), one would think that the same engine will exhibit the same “leak” rates regardless of what’s feeding it. Therefore if one were to measure actual intake flow pre FI device I would expect the numbers to be fairly equal given the same conditions in the manifold regardless of what FI method is used. Where is the flaw in my logic?

 

[globi5]

The intercooler can recover the lost air density but it can't recover the work the compressor put in the first place. So a less efficient compressor will either increase parasitic loss (supercharger) or increase backpressure (turbo) both will result in a loss of power even if you 'recover' the lost density with an intercooler.
Besides an intercooler is also a restriction in the airflow.

Regarding massflow and higher velocity:
If p*V=m*R*T (ideal gas) and if massflow is m/t this means
massflow = (p*V)/(R*T*t)
or in other words you cannot increase massflow without increasing pressure or reducing temperature. (The volume is basically given by the displacement and rpm of the engine.)

Well, that's how I understand it.

 

[Warpspeed]

Fitting a higher flow rated centrifugal compressor cannot increase flow through the engine unless the air density at the engine rises somehow. Boost must increase or induction temperature fall, or both. Most likely it will just reposition the different map contours under the same operating point without any significant net change in flow.

I dispute that all turbocharger compressors are more thermally efficient than all superchargers. A given turbo may have a fairly high peak efficiency at one particular combination of flow and pressure, but everywhere else efficiency falls off rapidly. A screw supercharger can have similar adiabatic efficiency to a turbo, but over a far wider operating range which makes it more useful. For a practical road applications it would be far superior without intercooling. But as has already been said, a reasonable attempt at intercooling is a great leveler.

Saying a turbo compressor will respond faster to flow changes may actually be true, assuming it is already spooled up to the required compressor Rpm. That is a very big if.

How can a turbo already be up to the required operating speed BEFORE the throttle opens to gain this theoretical advantage ? While a supercharger will usually have to fill all the pipework on sudden throttle opening, at least the supercharger itself does not lag behind the engine.

 

[globi5]

Warspeed--
I don't question what you just said.
I just wonder why VW used this twin charging concept instead
of just one twin screw supercharger to supercharge the engine?

http://www.greencarcongress.com/2005/08/inside_vws_new_.html

They could have applied a miller cycle instead of using an additional turbocharger and all the added complexity.
Mazda reached 92 HP/litre at 5300 rpm without intercooling so I think 120 HP/litre doesn't appear to be completly out of reach for a miller-cycled and intercooled engine.
What am I missing?

 

[mattsooty]

2 Points about why compound forced induction makes sense.

1. The twin charging system allows a huge, efficent turbo to be used at high mass flows. With all of its inherit benefits in terms of recouping lost energy through the exhaust system.

2. A supercharger is used to allow the 'off boost' performance penalties to be negated because boost is available from idle onwards.

As for the use of the miller cycle in an automotive application - I have no real experience. However I would have thought that the precise air metering requirments will be unavailable. Condsidering you never know how much air is truly where.

Also, the fuel delivery will not be emissions optimal considering the amount of time that the inlet valve will remain open and and the homogenity of the charge (brownian motion!?!)

MS

 

[Warpspeed]

Quite possibly the design goals were completely different.

The Miller cycle is really all about engine efficiency, particularly at small throttle openings. The high compression ratio and resulting higher expansion ratio can recover more heat energy from the fuel. If you want good specific power along with excellent fuel economy, the Miller cycle is an excellent way to get both.

If your goal is sheer power and performance from a small capacity engine, twincharging is hard to beat. A positive displacement blower in series with a turbo brings out the very best in both. The blower provides instant low rpm boost which really kicks the exhaust turbine. It will be very responsive. For sheer top end power, a turbo is going to maintain high airflow even when the Ve of engine and blower are starting to fall off.

If set up correctly boost pressure can be held just above total exhaust back pressure over a very wide Rpm range which gives wonderful exhaust scavenging with a zero overlap camshaft. The result is a very tractable responsive and powerful engine.

Several years ago I developed my own twincharge system on a small 4WD 1.6 Litre hatchback (Ford Laser in Australia). The results were beyond expectations. I can well believe VW are very pleased with their effort. Cost is the only real problem but VW were not the first to do it. Lancia and Nissan have built successful production twincharge engines.

 

[iolair]

>>Saying a turbo compressor will respond faster to flow changes may actually be true, assuming it is already spooled up to the required compressor Rpm. That is a very big if.

How can a turbo already be up to the required operating speed BEFORE the throttle opens to gain this theoretical advantage ?<<

Easy.....by placing the throttle in front of the compressor inlet. During part throttle cruise you can hear the turbine spooling up because the compressor isn't compressing much air. When the throttle is nailed, the boost gauge moves as fast as your foot moves the throttle.....zero lag/instanteous reponse.

 

[Warpspeed]

How much boost does this give at idle ?

 

[iolair]

>>How much boost does this give at idle ?<<

None...I simply responded to your prior statement about turbo compressors and in it you did not specify off idle. In a subsequent post, you did however mention instant low RPM supercharger performance.

 

[Warpspeed]

But idle may be from where you open the throttle, yes ?

If it is to have zero lag/instantaneous response as you claim, then it must be fully spooled up even at idle.

A supercharger will respond instantly, because it is mechanically coupled.

The engine may not, because of pipe and intercooler volumes that must first be filled, but those same (or similar) volumes can exist with a turbo too.

 

[iolair]

>>If it is to have zero lag/instantaneous response as you claim, then it must be fully spooled up even at idle.<<

Nope.....what part of part throttle cruise do you not understand? There is no "idle" involved in either the motor or the

turbo in my statement.

>>A supercharger will respond instantly, because it is mechanically coupled.

That can also be a problem with a supercharger.....a turbo does not suffer from the same fate of being mechanically

coupled/speed restricted, especially during part throttle cruise when the compressor is spooling up in a partial vacuum.

>>The engine may not, because of pipe and intercooler volumes that must first be filled, but those same (or similar)

volumes can exist with a turbo too.<<

In the real world, volumes are filled apparently much quicker than you imagine. The engine/turbo throttle response for all intents and purposes was pretty much as *instant* as it gets!

In 4th gear - 3.73 rear end - 12 psi

30 - 50 mph in slightly less than 2 seconds.

50 - 70 mph in slightly less than 2 seconds.

Photo circa 1984. Z28, 350 sbc, variable geometry manifold and Bosch K-Jetronic FI from a 6.9 litre Mercedes Benz.

 

[Mark911]

We’re getting off the topic with all this lag talk. When I mentioned “transient” conditions I didn’t mean on-boost off-boost conditions. What I meant was the dynamics of the intake process during steady state (engine rpm) conditions. For example, if one could show for whatever reason unique to the greater “flow vs pressure ratio” that the mass flow created by a turbo in the manifold, intake ports or across the valves is somehow greater than the mass flow created by a blower under the same conditions (density, temp, pressure, valve lift, rpm, etc) I be interested in hearing it. A good example of what I’m talking about is a recent article in a popular tuner mag where the express purpose was to keep upping the boost on a STOCK 4 cyl Honda eng until it blew and then figure out what broke (the weak link). In this test the authors managed X HP at Y boost level with the “initial turbo” combo before the turbo showed signs of diminishing returns (but still within it’s max efficiency island). Since the engine was still intact, they installed a huge turbo to achieve the higher boost pressures necessary to blow the engine and left everything else the same (except AF tuning of course). The interesting part is that they achieved X+40 some HP at THE SAME Y boost level with the larger compressor (obviously with a much higher spool up rpm but that’s not the issue). So the question is what accounts for the extra 40+HP at the same boost? It is reduced backpressure/pumping losses? Is it increased mass flow (at the same boost pressure and temps) and if so how? Is it a combination and if so what’s the distribution? These are the kind of questions I’m interested in being answered.
I’ll give you an off the wall example of what I consider a “smoking gun” kind of answer. Since the turbo is a “non-positive” displacement device it has some amount of flow leaking from the pressure side back to the ambient side (carry back). Possibly, during those brief milliseconds where the mass flow/pressure tends to lag cylinder pressure (initial crankshaft degrees following INT valve open when instantaneous VE is low, I just coined a new term!) this momentary drop in manifold pressure causes some or most of this leakage to cease and actually add to the overall flow. I call this “turbo sprint capacity” (another new term!). This would happen so quickly that a typical manifold pressure gauge wouldn’t even see it, but the result is more flow at the “perceived” same boost pressure. A “blower” could not respond in kind since it has very little leakage (relatively speaking) and therefore no sprint capacity. Again, the momentary drop in pressure is so quick that the boost gauge will still show full boost. The bigger the compressor is on the turbo the larger the “sprint capacity” is with respect to a smaller turbo explaining why a larger flow vs pressure ratio compressor can outflow a smaller compressor under the same dynamic conditions (exact same engine) even if the larger one is not in its highest efficiency band and the smaller is (ie, lower then recommended pressure ratio for the large turbo). Anyway, that’s the kind of explanation I’m looking for if one exists.
BTY, the stock 4 cly Honda engine finally broke a ring land at 18 psi and 420 flywheel hp just in case you were wondering.

 

[Warpspeed]

Mark911 I can see what you are getting at (sort of).

The flow through a centrifugal compressor is not defined by compressor Rpm, but by the total restriction to flow. If throttle position suddenly changes, airflow will too, limited only by the inertia of the air.

But it assumes that the compressor Rpm lines on the flow map are horizontal, and more flow is available at the same boost at the same compressor Rpm. That may sometimes almost be true, or definitely not true, depending where you are on the flow map.

It would only work for reasonably small changes in throttle opening. For large changes boost would suddenly fall off until compressor speed could build back up, the dreaded lag demon.

iolair, throttle response most certainly IS altered adversely whenever the throttle is located further away from the cylinder head. Any serious race engine uses individual throttle bodies very close to the intake valves for that exact reason. A few feet of pipe can make a very big difference.

I can tell you for a fact that snapping closed the throttle for gear changes can cause engine Rpm to be very slow to fall off when there are excessive pipe volumes between throttle and engine. Throttle opening is subjectively not quite so bad as throttle closing.

That is the difference between an engineer and a race car driver. The engineer assumes a few milliseconds mean nothing. The race car driver will tell you in no uncertain terms it feels like crap to drive.

 

[iolair]

Mark,

Maybe the larger compressor also had a larger A/R Turbine housing therefore producing a more favorable delta/p between the compressor outlet and the turbine inlet pressure.....that would easily generate 40HP more at the same pressure ratio.......plus, it would make it much easier to explain.

 

[iolair]

>>Any serious race engine uses individual throttle bodies very close to the intake valves for that exact reason.<<

Warp,

Would you consider a Renault F1 and a BMW F1 serious race engine?