I am building an engine currently for one of my customers who drives in the open rally class, The engine is a 4g63 Mitsu with a big 16G turbo. The class allows any engine mods to the stock block etc, as long as the compressor inlet on the turbo is reduced to 34MM. Since I have to make a new reducer for him, I am curious if there is a specific way to design it to have it effect the incoming air as little as possible. It must be within 50 MM of the compressor wheel and a minimum of 3MM thick. The engine doesn't have to be limited to that specific turbo, if there is an advantage to run a different sized turbo based on the restrictor, he will do it. So any ideas or theory is appreciated.
Eng-Tips is the largest engineering community on the Internet
Intelligent Work Forums for Engineering Professionals
-
Congratulations waross on being selected by the Tek-Tips community for having the most helpful posts in the forums last week. Way to Go!
Getting around a turbo restrictor requirement 1
- Thread starter MNRaptor
- Start date
MikeHalloran
Mechanical
Do the rules say what part of the restrictor must be within a given distance from the wheel? It makes a difference whether it says the 'closest part' of the restrictor or the 'smallest diameter'. I'm thinking of a really long conical passage from the actual turbo inlet to a short aperture that's 34mm diameter. It would work best if the double angle of the cone could be less than 7 degrees. Upstream of the restrictor, I'd put a trumpet bell shape.
Yeah, just like the venturi of a carburetor.
Mike Halloran
NOT speaking for
DeAngelo Marine Exhaust Inc.
Ft. Lauderdale, FL, USA
Yeah, just like the venturi of a carburetor.
Mike Halloran
NOT speaking for
DeAngelo Marine Exhaust Inc.
Ft. Lauderdale, FL, USA
There was an excellent article in Race Engine Technology Magazine about just such a situation. I'm sorry, but I borrowed the magazine and don't have the issue number. It was quite a technical model that used CFD modeling as well as real world dyno info to validate the findings. Essentially the conclusions were that the larger the airbox the better and that a diffuser of between 3 and 7 degree taper somewhat over 100mm long gave the best results. This was for a normally aspirated engine, and the rules may preclude any sort of airbox before the turbo, but it is worth looking into. Some current spec LeMans and ALMS cars run 2.0 4cylinder turbocharged engines with restrictors- the AER MG engine comes to mind. If your customer is willing to perhaps pay a consulting fee, you may be able to get some useful information from them.
Lou Scannon
Automotive
Looking at the turbo side of it, the restrictor is going to reduce inlet pressure as you try to increase air flow for more power. So what you want is a compressor that can efficiently compress air at a high enough pressure ratio to make the quest for increased air flow a winning proposition. I would expect the ideal turbo for this situation to be quite a bit different from the big 16G. Take a look at some diesel engine turbos, they are efficient at pressure ratios up to 4:1 in order to meet their manifold pressure requirements and altitude ratings.
Good intercooling will also become important, due to the high temperatures resulting from the high pressure ratio.
Good intercooling will also become important, due to the high temperatures resulting from the high pressure ratio.
incandescent
Chemical
Hemi, does the restrictor not limit massflow at a given pressure drop across it? I might be wrong but by my logic the pressure ratio of the turbo has little to do with power production because the restrictor is the limiting factor.
I can however see how one would want a real quick spooling turbo to draw the maximum mass that the restrictor will allow. Area under the curve so to speak. I believe in general, turbos that are efficient at high pressure ratios take a long time to spool up, whereas turbos that are efficient at low pressure ratios spool up quickly. Would you not want a turbo as smaller turbo in a restrictor application vs unrestricted?
I am curious about all this and willing to be corrected and learn.
Best Regards
I can however see how one would want a real quick spooling turbo to draw the maximum mass that the restrictor will allow. Area under the curve so to speak. I believe in general, turbos that are efficient at high pressure ratios take a long time to spool up, whereas turbos that are efficient at low pressure ratios spool up quickly. Would you not want a turbo as smaller turbo in a restrictor application vs unrestricted?
I am curious about all this and willing to be corrected and learn.
Best Regards
Lou Scannon
Automotive
The restrictor will become a limiting factor when the flow through it becomes sonic, at a pressure ratio of around 2:1.
You are right, a turbo will not be able to overcome this absolute flow limit but a high pressure ratio capable turbo will be able to function well up against this limit whereas a less capable turbo will be in surge and/or less efficient.
Depending on the nature of the racing, turbo spool time may also be an important factor. It may be that a faster spooling turbo will work better on some courses, and a higher PR capable turbo will work better on others.
You are right, a turbo will not be able to overcome this absolute flow limit but a high pressure ratio capable turbo will be able to function well up against this limit whereas a less capable turbo will be in surge and/or less efficient.
Depending on the nature of the racing, turbo spool time may also be an important factor. It may be that a faster spooling turbo will work better on some courses, and a higher PR capable turbo will work better on others.
Look back a few years(7 or 8 maybe) to the time when Toyota got a death sentence (banned) from the World Rally Championship for "sucessfully" creating a device to bypass the restrictor plate. The only reason that they were caught is that an ex-employee involved with the project provided the details to the F.I.A. and the rest is history.
I don't remember the details of how it worked, but I believe there was a description of how it functioned in Racecar Engineering magazine.
I don't remember the details of how it worked, but I believe there was a description of how it functioned in Racecar Engineering magazine.
red300zx99x
Automotive
From what I remember Toyota created a device that was attached to an intake part, that when not bolted onto the engine unfolded, probably some linkage mechanism. But when off the car looking at it youd think nothing off it. When this device opened up it moved the valve that regulated boost. No valve in this application, but I would love to one day actually see a pic of the device.
incandescent
Chemical
Hemi, I believe I made a mistake. Appreciate if you can the rest could comment on the accuracy of the paragraph I type below.
A restrictor doesn't place restrictions on mass flowing across it, but volume. Change the density of the air, and massflow will increase. The only way to change density is to change pressure (or temperature and hence pressure), this is how forced induction can make large amounts of power from a given port. At NA pressure differentials, air volume for a given mass is large (in other words un-dense), so achieving the same power without creating a higher pressure differential, is impossible as flow goes sonic.
Hemi, could you explain a little more about how flow would go sonic around PR 2? I cannot see how it would. If it were so, why are their forced induction systems that operate in the PR 2 and beyond range? Drag race, tractor pull, F1 turbo engines of old.. they all exceed PR 2.
Best Regards
A restrictor doesn't place restrictions on mass flowing across it, but volume. Change the density of the air, and massflow will increase. The only way to change density is to change pressure (or temperature and hence pressure), this is how forced induction can make large amounts of power from a given port. At NA pressure differentials, air volume for a given mass is large (in other words un-dense), so achieving the same power without creating a higher pressure differential, is impossible as flow goes sonic.
Hemi, could you explain a little more about how flow would go sonic around PR 2? I cannot see how it would. If it were so, why are their forced induction systems that operate in the PR 2 and beyond range? Drag race, tractor pull, F1 turbo engines of old.. they all exceed PR 2.
Best Regards
GregLocock
Automotive
"Hemi, could you explain a little more about how flow would go sonic around PR 2?"
It's called sonic choking in orifice plates. You'll need a reasonable understanding of fluid mechanics to get to the bottom of it.
Cheers
Greg Locock
It's called sonic choking in orifice plates. You'll need a reasonable understanding of fluid mechanics to get to the bottom of it.
Cheers
Greg Locock
incandescent
Chemical
I understand that flow stops increasing once flow goes sonic.
I think I know where I may have gone wrong. PR 2 is post vs pre restrictor (atmospheric on pre, and half atmospheric on post), and not PR of the turbo. Am I right?
I think I know where I may have gone wrong. PR 2 is post vs pre restrictor (atmospheric on pre, and half atmospheric on post), and not PR of the turbo. Am I right?
GregLocock
Automotive
It's the pressure ratio across the orifice itself that is important. I'm not sure the number is exactly 2, but it is of that order.
Cheers
Greg Locock
Cheers
Greg Locock
mikeyboy19
Automotive
I'm just wondering.. Is it too far off the point to be looking at the volume of the plenum engine side of the restrictor plate?
(With the logic that the restrictor plate becomes critical at a given mass flow rate and hence engine speed.)
Any thoughts?
(With the logic that the restrictor plate becomes critical at a given mass flow rate and hence engine speed.)
Any thoughts?
-
1
- #15
Typically, the best you can do is a well designed venturi with a smooth surface finish, rounded inlet, and a 7-10 degree diffuser. The diffuser would be ideal if it opened to the full diameter of the turbo inlet, but typically the distance from compressor wheel rule makes this a bit difficult.
The venturi (for air STP) will choke when the throat to upstream pressure ratio is about 0.528 (the 2:1 rule mentioned earlier). For a venturi with a rockin' diffuser and no downstream disturbances (such as a spinning compressor wheel) this will occur when the diffuser exit pressure is approximately 15-20% less than the inlet static pressure ~(0.85*14.7psia in this case). The value of a well designed venturi over the plate becomes apparent when you realize that the choking pressure for the plate is the pressure at the plate exit (0.528*14.7 psia). Actually in practice the orifice plate is more like 0.65-0.75 *14.7 because the fluid physically creates its own inefficient diffuser with a stagnation region on the back side of the plate. The net result is your engine has to pull more vacuum to generate the same flowrate.
lets look at your restrictor's mass flow rate...
mdot = cd * rho * A * v (at the smallest cross section)
vcritical = sqrt(k*r*t) ~= 1100 fps for air@STP (choked, limiting flow case)
A = pi*d^2*0.25 = 907.89mm^2 = 1.407in^2 = 0.00977ft^2
cd = 0.95-0.99 for a good venturi
rho = 0.074915 lbm/ft^3 (air at STP)
rhothroat = 0.528*0.074915 lbm/ft^3 = 0.0395 lbm/ft^3
remember that our pressure is different at the throat of the venturi
Qdot = cd*A*vcritical = 10.75 ft^3/sec = 612 ft^3/min (CFM)
Mdot = rhothroat * Qdot
Mdot = 24.19 lbm/min
Hope this helps, also when I worked for a flowmeter company I designed and built a restrictor for an Eclipse Open class car, it may still be floating around if you are interested.
Buddy Cochran
The venturi (for air STP) will choke when the throat to upstream pressure ratio is about 0.528 (the 2:1 rule mentioned earlier). For a venturi with a rockin' diffuser and no downstream disturbances (such as a spinning compressor wheel) this will occur when the diffuser exit pressure is approximately 15-20% less than the inlet static pressure ~(0.85*14.7psia in this case). The value of a well designed venturi over the plate becomes apparent when you realize that the choking pressure for the plate is the pressure at the plate exit (0.528*14.7 psia). Actually in practice the orifice plate is more like 0.65-0.75 *14.7 because the fluid physically creates its own inefficient diffuser with a stagnation region on the back side of the plate. The net result is your engine has to pull more vacuum to generate the same flowrate.
lets look at your restrictor's mass flow rate...
mdot = cd * rho * A * v (at the smallest cross section)
vcritical = sqrt(k*r*t) ~= 1100 fps for air@STP (choked, limiting flow case)
A = pi*d^2*0.25 = 907.89mm^2 = 1.407in^2 = 0.00977ft^2
cd = 0.95-0.99 for a good venturi
rho = 0.074915 lbm/ft^3 (air at STP)
rhothroat = 0.528*0.074915 lbm/ft^3 = 0.0395 lbm/ft^3
remember that our pressure is different at the throat of the venturi
Qdot = cd*A*vcritical = 10.75 ft^3/sec = 612 ft^3/min (CFM)
Mdot = rhothroat * Qdot
Mdot = 24.19 lbm/min
Hope this helps, also when I worked for a flowmeter company I designed and built a restrictor for an Eclipse Open class car, it may still be floating around if you are interested.
Buddy Cochran
crysta1c1ear
Automotive
As I understand it, yes, things choke up at a pressure ratio of 1/0.528, but the mass flowrate at this pressure ratio and beyond is a horrible formula with the product of the input pressure (before the restriction) and the input density (before the restriction), under a square root sign.
(There is lots of other mess in the equation which I'll treat as constants, ha ha.)
Now pressure is proportional to density and temperature so that is like having density² and temperature under the square root sign.
Temperature is the average kinetic energy of particles per unit weight, and kinetic energy is ½mv² so the square root of a temperature is basically an air particle velocity measurement.
End result as I see it, is that the maximum possible mass air flow through a given restriction depends on two things, firstly density, and secondly speed of the molecules (root of temperature).
Imagine air in a space capsule with a pinhole in it. Air molecules moving around at random that 'hit' the hole will go straight through it. If you double the density of the air, twice as many molecules will find their way out. Despite the pressure ratio being infinite, you cannot flow more air molecules than those that randlomly hit the hole.
Now if you double the speed of the air molecules the same molecules come out at twice the speed, its like a film being fast forwarded: the same events happen at twice the speed, and again the mass flow rate can be doubled.
So I'd say you want dense air going in, as dense as possible, and purely from the point of view of maximizing airflow, hotter is better too but increasing temperature is not as advantageous as increasing density, as explained above. (Eg double the density, double the massflow, 4 times the temperature, also double the massflow.)
=
I've seen equations on this bulletin board. I downloaded an equation editor, but it has a free trial date on it. I would have liked to insert the equations.
=
Another note, is that inlet velocity affects the maximum mass flowrate too.
=
I read the article by Motoman about his high velocity ports: interesting reading, I liked it. I think people make too much of the speed of sound. Isn't all the air at the equator travelling faster than the speed of sound due to rotation of the earth? It doesn't stop you talking to the guy next to you! At the right latitude, sound presumably stays still when 'its travelling west' and goes at twice the speed of sound when 'going east' since the whole surrounding air was going that way anyway. I'd best stop, as I'm waffling off-topic.
(There is lots of other mess in the equation which I'll treat as constants, ha ha.)
Now pressure is proportional to density and temperature so that is like having density² and temperature under the square root sign.
Temperature is the average kinetic energy of particles per unit weight, and kinetic energy is ½mv² so the square root of a temperature is basically an air particle velocity measurement.
End result as I see it, is that the maximum possible mass air flow through a given restriction depends on two things, firstly density, and secondly speed of the molecules (root of temperature).
Imagine air in a space capsule with a pinhole in it. Air molecules moving around at random that 'hit' the hole will go straight through it. If you double the density of the air, twice as many molecules will find their way out. Despite the pressure ratio being infinite, you cannot flow more air molecules than those that randlomly hit the hole.
Now if you double the speed of the air molecules the same molecules come out at twice the speed, its like a film being fast forwarded: the same events happen at twice the speed, and again the mass flow rate can be doubled.
So I'd say you want dense air going in, as dense as possible, and purely from the point of view of maximizing airflow, hotter is better too but increasing temperature is not as advantageous as increasing density, as explained above. (Eg double the density, double the massflow, 4 times the temperature, also double the massflow.)
=
I've seen equations on this bulletin board. I downloaded an equation editor, but it has a free trial date on it. I would have liked to insert the equations.
=
Another note, is that inlet velocity affects the maximum mass flowrate too.
=
I read the article by Motoman about his high velocity ports: interesting reading, I liked it. I think people make too much of the speed of sound. Isn't all the air at the equator travelling faster than the speed of sound due to rotation of the earth? It doesn't stop you talking to the guy next to you! At the right latitude, sound presumably stays still when 'its travelling west' and goes at twice the speed of sound when 'going east' since the whole surrounding air was going that way anyway. I'd best stop, as I'm waffling off-topic.
Crysta1c1ear,
you need to be a little bit careful here, increasing the inlet pressure increases the inlet density. Decreasing the inlet temperature increases the inlet density. rho = P/rT
The compressible flow equation mdot = C*A2*P1*SQRT[(gc/RT1)*k*(2/(k+1))^{(k+1)/(k-1)}]
mdot is directly proportional to inlet pressure
mdot is inversly proportional to the sqrt of inlet temperature
the equation is derived with ABSOLUTE temperature. So doubling the temp takes more than you think...
Buddy
you need to be a little bit careful here, increasing the inlet pressure increases the inlet density. Decreasing the inlet temperature increases the inlet density. rho = P/rT
The compressible flow equation mdot = C*A2*P1*SQRT[(gc/RT1)*k*(2/(k+1))^{(k+1)/(k-1)}]
mdot is directly proportional to inlet pressure
mdot is inversly proportional to the sqrt of inlet temperature
the equation is derived with ABSOLUTE temperature. So doubling the temp takes more than you think...
Buddy
crysta1c1ear
Automotive
bcochran
You are just looking at the same thing as I am from a different angle.
As you point out P = rho*r*T, so T appears on the top of the equation as well as appearing as a square root at the bottom. The end result is
mdot is proportional to the sqrt of inlet temperature.
When air heats up, its pressure rises. You are partly considering temperature as part of this pressure and then considering it as an independent entity at the bottom of the equation.
So you are saying the pressure part of mdot goes up proportional with temperature and another part of mdot comes down inversely proportional to the square root of the temperature.
I have merely combined the two to say what the net effect of temperature is. Furthermore temperature is basically due to kinetic energy of molecules and its square root is thus a measure of the random speed of the molcules.
So it basically boils down to the faster the molecules move, the more get through a given hole, which is what we'd intuitively expect.
=
In the pinholed space capsule analogy, I'm saying if you heat the air so the molecules move 5% faster, they will come out 5% faster.
(1.05*1.05=1.1025 so 5% faster is 10.25% hotter.)
You are saying the temperature rises 10.25%, so the pressure rises 10.25% and the molecules come out 10.25% faster due to the pressure, but that they also come out 5% slower due to temperture and the combined effect of your pressure and temperature calculations is the air comes out 5% faster.
At the end of the day, its the same.
You are just looking at the same thing as I am from a different angle.
As you point out P = rho*r*T, so T appears on the top of the equation as well as appearing as a square root at the bottom. The end result is
mdot is proportional to the sqrt of inlet temperature.
When air heats up, its pressure rises. You are partly considering temperature as part of this pressure and then considering it as an independent entity at the bottom of the equation.
So you are saying the pressure part of mdot goes up proportional with temperature and another part of mdot comes down inversely proportional to the square root of the temperature.
I have merely combined the two to say what the net effect of temperature is. Furthermore temperature is basically due to kinetic energy of molecules and its square root is thus a measure of the random speed of the molcules.
So it basically boils down to the faster the molecules move, the more get through a given hole, which is what we'd intuitively expect.
=
In the pinholed space capsule analogy, I'm saying if you heat the air so the molecules move 5% faster, they will come out 5% faster.
(1.05*1.05=1.1025 so 5% faster is 10.25% hotter.)
You are saying the temperature rises 10.25%, so the pressure rises 10.25% and the molecules come out 10.25% faster due to the pressure, but that they also come out 5% slower due to temperture and the combined effect of your pressure and temperature calculations is the air comes out 5% faster.
At the end of the day, its the same.
- Thread starter
- #20
- Status
Similar threads
- Locked
- Question
- Locked
- Question