Full Throttle/Slowboy Turbos Compared to......

bboy · 08-28-2004, 02:41 AM

What AVO and APS offer. Yeah, I'm a total geek. This is how I spend my Friday nights. Many of these values are derived from the various web site's data. Other data is estimated not by me but by the manufacturers. In one case I took a stab at the value based on comparison to turbos so similar it only made sense. It's all color coded for ease of reading. I can add other turbos if you PM me a link to the data.

http://forums.nasioc.com/forums/atta...chmentid=58631

I'm interested in what the Guru's think of the exhaust housing A/R (a rather murky number since it does not really get at the "volume" of the turbine housing).

* Registered users of the site do not see these ads.

wrex03 · 08-28-2004, 07:11 PM

I'm having a gt30 -10 tuned on the ninth.

totoherbs · 08-28-2004, 07:18 PM

Quote:

Originally Posted by wrex03

I'm having a gt30 -10 tuned on the ninth.

Nice. Let us know how it turs out.

Red Rocket · 08-29-2004, 12:44 AM

Quote:

Originally Posted by bboy

What AVO and APS offer. Yeah, I'm a total geek. This is how I spend my Friday nights. Many of these values are derived from the various web site's data. Other data is estimated not by me but by the manufacturers. In one case I took a stab at the value based on comparison to turbos so similar it only made sense. It's all color coded for ease of reading. I can add other turbos if you PM me a link to the data.

http://forums.nasioc.com/forums/atta...chmentid=58631

I'm interested in what the Guru's think of the exhaust housing A/R (a rather murky number since it does not really get at the "volume" of the turbine housing).

Wow, that's a fantastic table, thanks!!!

I've measured the .82 AR AVO housing at around 10 cm^2 nozzle area....it's big.

Kevin

Red Rocket · 08-29-2004, 12:48 AM

Your avo data is wrong on the CHRA's, by the way - the avo450 is a gt30r-10, same as the FT-10, and the AVO500 = FT-12. Don't know about the 400.

Kevin

Red Rocket · 08-29-2004, 12:51 AM

Actually, a bunch of stuff is wrong....I'm too tired to help figure it all out right now, but I can tomorrow. It would be nice get a comprehensive table together, I aplaud your efforts.

Kevin

cronic · 08-29-2004, 12:55 AM

I am more worried about the bigger SBR/FTS turbos.. I had offered my car up to test the -12 have it tuned and dyno'd within 5 days of getting it, but i did not feel the offer was good enough on their end.. I would hate to dump all that money on one and it not perform. I would have loved to see what the -12 would have done over the green.. I held a -10 in my hands last weekend and i think that little guy is going to do well..

I will say Mike at slowboy was awesome to deal with and very fast to respond to my emails, and if they turn out to make power then a -12 will be on my car for sure in the future

bboy · 08-29-2004, 04:09 PM

The AVO data came from the avoturboworld.com site:

http://210.15.220.239/html/garrett/garrett_avo.html

I happy to hear you disagree with the figures since I want these data as accurate as possible. The problem with the above page is that the turbos are not listed as AVO400, AVO450, etc. I have tried to interpret was in the table on that page as best I can.

I'd like "manufacturer" data if possible. I'll dig some more, but I've been to AVO Aurstralia, USA, Swiss', UK, and Japan. No more dat than stuff such as 320 PS or 1.4 kg/cm which I assume is a misprint of 1.4 kg/cm^2 (the least informative means of characterizing a turbo's output imaginable; it's like saying 35 psi turbo).

Back to digging, let me know if anyone has figures to update the table.

Red Rocket · 08-29-2004, 06:24 PM

Quote:

Originally Posted by bboy

The AVO data came from the avoturboworld.com site:

http://210.15.220.239/html/garrett/garrett_avo.html

I happy to hear you disagree with the figures since I want these data as accurate as possible. The problem with the above page is that the turbos are not listed as AVO400, AVO450, etc. I have tried to interpret was in the table on that page as best I can.

I'd like "manufacturer" data if possible. I'll dig some more, but I've been to AVO Aurstralia, USA, Swiss', UK, and Japan. No more dat than stuff such as 320 PS or 1.4 kg/cm which I assume is a misprint of 1.4 kg/cm^2 (the least informative means of characterizing a turbo's output imaginable; it's like saying 35 psi turbo).

Back to digging, let me know if anyone has figures to update the table.

We went over this with AVO Steve last year some time. You are looking at the wrong page, as we were - those are genral use turbos. There is a subaru specific page you need to look at. I suggest PM'ing AVO steve, his user name on here is "avoturbo".

Kevin

Mancini · 08-29-2004, 07:41 PM

I'll just throw this in, but Slowboy is a local shop around here and it's a good shop. A friend of mine has spent a rediculous amount of money there for his 1g. I had my heads for my 98 RS-T o-ringed there and they were great with customer service.

So I'm hoping that you get some good numbers as that was a turbo I was looking into (albeit in DSM trim) when I was running my RS-T. It'll be a nice addition to the turbo's we have available. And from what I calculated a long time ago it seemed great for a 2.5L running on pump gas. Keep us posted!

Brian

bboy · 08-30-2004, 01:29 AM

I'm hoping for good numbers too from the FT/SB turbos. I'm not hunting for slams here, rather trying to figure out which one to pick and how they compare to the other Garrett based turbos on which we have limited, but some, data.

Red Rocket- the CHRA's connect the hot side to the cold. It's my understanding that the -10, -11, -12 denotes the impeller size and that the CHRA remains the same for all these turbos with a nearest descriptor being "GT30R center section".

As for the data on the AVO site, bummer, that's the best table I've seen. I just spent a few hours hunting for more AVO info and came up empty handed for the most part. The only new number: 53 mm inlet diameter for exhaust housing. Work the numbers with a 0.86 A/R and that makes the radius 3.95 inches (in close agreement with Garrett data--see below)

Some things I'd like to set straight for those reading the pdf.

"Trim" while good descriptor for other things, it is not for a turbo. The "trim" is nothing but a ratio or the inner and outer dimensions of the impeller/expeller squared and multipled by 100. "Trim" can give you an idea of how much air an impeller can compress/express if the outer diameter (larger) is held constant. A wheel with 2.6/3.0 inch diameters is going to compress/express more air than one with 2.1/3.0. Think of it as a lever action, larger "trims" offer more leverage on the air.

A/R is also a ratio, or the area of the inlet (nose) of the turbine divided by the radius of from the center of the impeller shaft to the center point of the housing. The A/R is usually constant as you move radially around the housing. As the radius diminishes, the area does too, and the the rate of change is constant for both the cross-section and the radius. A/R, like Trim, describes a level of compressability/expressability of a gas within the housing.

A major mis-understanding by many is that A/R (especially of the turbine) says nothing about how the "volume" of the housing. Take your krispy creme donut and the inner tube that I used to ride on behind a boat in my youth, the A (cross sectional area) and the R (radius) of these two donut-shaped objects are widely different, but their A/R is very much the same. A/R says something about rate of compression, but nothing about volume of exhaust per unit time that "can" move through an exhaust housing.

In general for the same size housing (the R being the same) a higher A/R ratio will flow more exhaust, but if you change both dimensions all bets are off.

This brings me to the P18, P20 or T04E descriptions of compressor housings, without any correlate description of the exhaust housing. A P20 housing means nothing to me. How it compares to at TO4E I have now idea, or for that matter to a 8 cm FP housing. There are folks out there with tons of turbos and they can compare them qualitatitively, but the "descriptor" has no dimensions to it at all. If the radius, R, of the housing were specified it would save us all a lot of grief. From the R we could derive the A or A/R.

Reading the Garrett catalog very closely the R, radius, for GT25, GT30, GT35, GT40 would all seem to be 3.93 inches on the compressor side. We might assume that the radius is the same on the turbine side for Garrett brand turbos, but we don't know and we know even less about what AVO, APS, FT have done in modifying the turbine side to fit the Suby flanges.

Let's say AVO maintains the Garrett radius, R=3.93, and the inlet is 2 inch diameter, that's pi r squared for A. r =1.0, 3.14/3.93 = .8 A/R. What if APS decides lets make the R=3.5, barely perceptable by eye, now we have 3.14/3.5, A/R = 0.9. A/R went up, looks like more flow, but it it won't not becuase the housing is actually smaller. Now let's say FT changes both: the inlet is now 1.75 inchs (apeture of a GT Spec uppipe) and the radius is 4.13 , neither huge changes. A/R is now 2.40/4.13 = 0.58.

Which one flows more? Which one spools the fastest?

Answer: they flow basically the same because the uppipe, not the turbo, is the restriction here, but the FT spools faster. Without real numbers, no one knows.

I'm happy to call Steve at AVO if he'll give me some mo' info, but he's not going to like the $500 price drop from the competition--and there is the rub. We may be better off with member measurements.

duncangrant · 08-30-2004, 06:35 AM

Good luck on your quest.

Some thoughts (I'm happy to be corrected):
Pxx turbine housings: the xx is the A/R in mm.

A/R ratio constant along the volute -
I think this is a myth. Maybe it was once true back in the early days of turbochargers but it doesn't square with any housing I've seen. The minimum value of the R is constrained by the major radius of the wheel plus a bit. Consider the area of the small end of the scroll relative to the large end. The ratio is probably around 1:10, maybe larger. Now for A/R to be constant along the scroll the R at the thick end is going to be 10x the wheel radius. Imagine a 3" wheel having a 30" housing - nonsense! Actually in practice the R stays fairly constant along the scroll. Some Garrett compressor housings clearly have a constant R throughout the volute - you can see the spacing from the inside edge of the scroll increasing towards the small end, with the centre-line staying at constant R.

Is A/R really a good indicator of turbine performance?
Consider two turbines of the same nozzle area (A), #1 with the scroll wrapped tightly around the wheel (the normal one) and #2 with the scroll wrapped round at twice the radius so it has 1/2 the A/R of #1. Conventional thinking on A/R is that #2 with the small A/R will spool earlier and choke earlier than #1. But why, they have the same nozzle area? I don't buy it. It seems to me that the important parameter is the nozzle area and A/R is only useful as an indirect indicator of A if you assume that the scroll is tightly wrapped around the wheel - which in fact seems to be normal practice. I'd suggest that the area that's really important is not the volute entry area or 'throat area' (as the standard definition of A/R) but the area of the exit nozzle - that is the circumfrential slot that focuses the exhaust gas onto the turbine wheel. I guess the area of the entry nozzle should be at least as large as the exit nozzle.

To convert A/R to throat area given wheel size -
Rt = throat radius assuming a circular cross-section throat
Ri = 'scroll inside radius' or wheel major radius + a bit
A = throat cross-sectional area
R = throat centroid radius
=>
Rt = (A/R + sqrt((A/R)^2 + 4*pi*Ri*A/R)) / (2*pi)
A = pi*Rt^2

AZScoobie · 08-30-2004, 01:59 PM

Good work man.

C

happasaiyan · 08-30-2004, 02:31 PM

where does the SR50 stand? i would think everything the same as the SR55, but with a smaller wheel.

AZScoobie · 08-30-2004, 03:28 PM

Quote:

Originally Posted by happasaiyan

where does the SR50 stand? .

In a league of its own... WAY below the others.

C

bboy · 09-02-2004, 06:32 PM

I love how Clark state's things emphatically, leaving no doubt about what he thinks.

Duncangrant, I could not agree more. I just posted in another thread my agreement with you. I looked the damn physic's law up again. It's Poiseuille's Law:

Resistance R = (8hL)/pr4 or Volumetric Flow Q = (pDPr4)/(8hL)

where R = resistance, P = pressure, n = mean flow velocity, Q = volumetric flow rate, L = tube length, h = fluid viscosity, and r = tube radius, D=?

If you change the r of the nozzle even the tiniest bit it's magnified to the fourth power. I don't know the units in the equation, but if r changes by 2, that change has a 2*2*2*2 effect, or 16 fold. So, say you are flowing 500 cfm through a 2.5 inch nozzle--> little change in nozzle diameter, 0.5 inches--> the change in the nozzle area is 4.9 in^2 to 3.14 in2 or 3.14/4.9= 0.61 = 61%-->you expect 39% less flow based on area, but the actual flow drops 96% according to Poisseille--> you go from 500 cfm to 21 cfm. (That is not a typo, 21 cfm, I did make a bunch of reasonable assumptions about the other values).

It's not an accurate model for exhaust, and there is a lot more going on, but that one apeture at the can make a huge difference in the capacity for turbine flow.

08-30-2004, 06:35 AM	#12
duncangrant Scooby Newbie Member#: 30009 Join Date: Nov 2002 Location: Scotland Vehicle: 98 Legacy BG5B Black twin-turbo	geek alert Good luck on your quest. Some thoughts (I'm happy to be corrected): Pxx turbine housings: the xx is the A/R in mm. A/R ratio constant along the volute - I think this is a myth. Maybe it was once true back in the early days of turbochargers but it doesn't square with any housing I've seen. The minimum value of the R is constrained by the major radius of the wheel plus a bit. Consider the area of the small end of the scroll relative to the large end. The ratio is probably around 1:10, maybe larger. Now for A/R to be constant along the scroll the R at the thick end is going to be 10x the wheel radius. Imagine a 3" wheel having a 30" housing - nonsense! Actually in practice the R stays fairly constant along the scroll. Some Garrett compressor housings clearly have a constant R throughout the volute - you can see the spacing from the inside edge of the scroll increasing towards the small end, with the centre-line staying at constant R. Is A/R really a good indicator of turbine performance? Consider two turbines of the same nozzle area (A), #1 with the scroll wrapped tightly around the wheel (the normal one) and #2 with the scroll wrapped round at twice the radius so it has 1/2 the A/R of #1. Conventional thinking on A/R is that #2 with the small A/R will spool earlier and choke earlier than #1. But why, they have the same nozzle area? I don't buy it. It seems to me that the important parameter is the nozzle area and A/R is only useful as an indirect indicator of A if you assume that the scroll is tightly wrapped around the wheel - which in fact seems to be normal practice. I'd suggest that the area that's really important is not the volute entry area or 'throat area' (as the standard definition of A/R) but the area of the exit nozzle - that is the circumfrential slot that focuses the exhaust gas onto the turbine wheel. I guess the area of the entry nozzle should be at least as large as the exit nozzle. To convert A/R to throat area given wheel size - Rt = throat radius assuming a circular cross-section throat Ri = 'scroll inside radius' or wheel major radius + a bit A = throat cross-sectional area R = throat centroid radius => Rt = (A/R + sqrt((A/R)^2 + 4piRiA/R)) / (2pi) A = pi*Rt^2

Thread Tools
Show Printable Version Email this Page
Display Modes
Linear Mode Switch to Hybrid Mode Switch to Threaded Mode

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Will 20G hit full boost later on 91octane compare to 93octane?	Liam	Newbies & FAQs	16	10-06-2007 10:49 PM
06 turbo compared to 04-05s	RedlineWRX04	Factory 2.0L Turbo Powertrain (EJ Series Factory 2.0L Turbo)	2	06-28-2006 12:11 AM
Lurch when going from full throttle to part throttle	Macabre	Factory 2.0L Turbo Powertrain (EJ Series Factory 2.0L Turbo)	9	09-06-2003 06:02 PM
stock turbo compared to VF series?	nhluhr	Factory 2.0L Turbo Powertrain (EJ Series Factory 2.0L Turbo)	2	12-04-2002 11:13 PM
What would be the benefit of ball bearing turbo compared to other ??	setiawani	Factory 2.0L Turbo Powertrain (EJ Series Factory 2.0L Turbo)	4	09-28-2002 12:02 AM

08-28-2004, 02:41 AM	#1
bboy Scooby Specialist Member#: 56468 Join Date: Mar 2004 Chapter/Region: NWIC Location: Seattle, WA Vehicle: 04 Improved STI Dirty White	Full Throttle/Slowboy Turbos Compared to...... What AVO and APS offer. Yeah, I'm a total geek. This is how I spend my Friday nights. Many of these values are derived from the various web site's data. Other data is estimated not by me but by the manufacturers. In one case I took a stab at the value based on comparison to turbos so similar it only made sense. It's all color coded for ease of reading. I can add other turbos if you PM me a link to the data. http://forums.nasioc.com/forums/atta...chmentid=58631 I'm interested in what the Guru's think of the exhaust housing A/R (a rather murky number since it does not really get at the "volume" of the turbine housing). * Registered users of the site do not see these ads.

08-28-2004, 07:11 PM	#2
wrex03 Scooby Specialist Member#: 29963 Join Date: Nov 2002 Location: chicago Vehicle: 04 sti silver	I'm having a gt30 -10 tuned on the ninth.

08-29-2004, 12:48 AM	#5
Red Rocket Scooby Specialist Member#: 10507 Join Date: Sep 2001 Vehicle: '04 Ex STi Owner '97 4Runner	Your avo data is wrong on the CHRA's, by the way - the avo450 is a gt30r-10, same as the FT-10, and the AVO500 = FT-12. Don't know about the 400. Kevin

08-29-2004, 12:51 AM	#6
Red Rocket Scooby Specialist Member#: 10507 Join Date: Sep 2001 Vehicle: '04 Ex STi Owner '97 4Runner	Actually, a bunch of stuff is wrong....I'm too tired to help figure it all out right now, but I can tomorrow. It would be nice get a comprehensive table together, I aplaud your efforts. Kevin

08-29-2004, 12:55 AM	#7
cronic Scooby Guru Member#: 9749 Join Date: Aug 2001 Chapter/Region: SWIC Vehicle: -	I am more worried about the bigger SBR/FTS turbos.. I had offered my car up to test the -12 have it tuned and dyno'd within 5 days of getting it, but i did not feel the offer was good enough on their end.. I would hate to dump all that money on one and it not perform. I would have loved to see what the -12 would have done over the green.. I held a -10 in my hands last weekend and i think that little guy is going to do well.. I will say Mike at slowboy was awesome to deal with and very fast to respond to my emails, and if they turn out to make power then a -12 will be on my car for sure in the future

08-29-2004, 04:09 PM	#8
bboy Scooby Specialist Member#: 56468 Join Date: Mar 2004 Chapter/Region: NWIC Location: Seattle, WA Vehicle: 04 Improved STI Dirty White	The AVO data came from the avoturboworld.com site: http://210.15.220.239/html/garrett/garrett_avo.html I happy to hear you disagree with the figures since I want these data as accurate as possible. The problem with the above page is that the turbos are not listed as AVO400, AVO450, etc. I have tried to interpret was in the table on that page as best I can. I'd like "manufacturer" data if possible. I'll dig some more, but I've been to AVO Aurstralia, USA, Swiss', UK, and Japan. No more dat than stuff such as 320 PS or 1.4 kg/cm which I assume is a misprint of 1.4 kg/cm^2 (the least informative means of characterizing a turbo's output imaginable; it's like saying 35 psi turbo). Back to digging, let me know if anyone has figures to update the table.

08-29-2004, 07:41 PM	#10
Mancini Scooby Specialist Member#: 20726 Join Date: Jul 2002 Chapter/Region: MAIC	I'll just throw this in, but Slowboy is a local shop around here and it's a good shop. A friend of mine has spent a rediculous amount of money there for his 1g. I had my heads for my 98 RS-T o-ringed there and they were great with customer service. So I'm hoping that you get some good numbers as that was a turbo I was looking into (albeit in DSM trim) when I was running my RS-T. It'll be a nice addition to the turbo's we have available. And from what I calculated a long time ago it seemed great for a 2.5L running on pump gas. Keep us posted! Brian

08-30-2004, 01:29 AM	#11
bboy Scooby Specialist Member#: 56468 Join Date: Mar 2004 Chapter/Region: NWIC Location: Seattle, WA Vehicle: 04 Improved STI Dirty White	I'm hoping for good numbers too from the FT/SB turbos. I'm not hunting for slams here, rather trying to figure out which one to pick and how they compare to the other Garrett based turbos on which we have limited, but some, data. Red Rocket- the CHRA's connect the hot side to the cold. It's my understanding that the -10, -11, -12 denotes the impeller size and that the CHRA remains the same for all these turbos with a nearest descriptor being "GT30R center section". As for the data on the AVO site, bummer, that's the best table I've seen. I just spent a few hours hunting for more AVO info and came up empty handed for the most part. The only new number: 53 mm inlet diameter for exhaust housing. Work the numbers with a 0.86 A/R and that makes the radius 3.95 inches (in close agreement with Garrett data--see below) Some things I'd like to set straight for those reading the pdf. "Trim" while good descriptor for other things, it is not for a turbo. The "trim" is nothing but a ratio or the inner and outer dimensions of the impeller/expeller squared and multipled by 100. "Trim" can give you an idea of how much air an impeller can compress/express if the outer diameter (larger) is held constant. A wheel with 2.6/3.0 inch diameters is going to compress/express more air than one with 2.1/3.0. Think of it as a lever action, larger "trims" offer more leverage on the air. A/R is also a ratio, or the area of the inlet (nose) of the turbine divided by the radius of from the center of the impeller shaft to the center point of the housing. The A/R is usually constant as you move radially around the housing. As the radius diminishes, the area does too, and the the rate of change is constant for both the cross-section and the radius. A/R, like Trim, describes a level of compressability/expressability of a gas within the housing. A major mis-understanding by many is that A/R (especially of the turbine) says nothing about how the "volume" of the housing. Take your krispy creme donut and the inner tube that I used to ride on behind a boat in my youth, the A (cross sectional area) and the R (radius) of these two donut-shaped objects are widely different, but their A/R is very much the same. A/R says something about rate of compression, but nothing about volume of exhaust per unit time that "can" move through an exhaust housing. In general for the same size housing (the R being the same) a higher A/R ratio will flow more exhaust, but if you change both dimensions all bets are off. This brings me to the P18, P20 or T04E descriptions of compressor housings, without any correlate description of the exhaust housing. A P20 housing means nothing to me. How it compares to at TO4E I have now idea, or for that matter to a 8 cm FP housing. There are folks out there with tons of turbos and they can compare them qualitatitively, but the "descriptor" has no dimensions to it at all. If the radius, R, of the housing were specified it would save us all a lot of grief. From the R we could derive the A or A/R. Reading the Garrett catalog very closely the R, radius, for GT25, GT30, GT35, GT40 would all seem to be 3.93 inches on the compressor side. We might assume that the radius is the same on the turbine side for Garrett brand turbos, but we don't know and we know even less about what AVO, APS, FT have done in modifying the turbine side to fit the Suby flanges. Let's say AVO maintains the Garrett radius, R=3.93, and the inlet is 2 inch diameter, that's pi r squared for A. r =1.0, 3.14/3.93 = .8 A/R. What if APS decides lets make the R=3.5, barely perceptable by eye, now we have 3.14/3.5, A/R = 0.9. A/R went up, looks like more flow, but it it won't not becuase the housing is actually smaller. Now let's say FT changes both: the inlet is now 1.75 inchs (apeture of a GT Spec uppipe) and the radius is 4.13 , neither huge changes. A/R is now 2.40/4.13 = 0.58. Which one flows more? Which one spools the fastest? Answer: they flow basically the same because the uppipe, not the turbo, is the restriction here, but the FT spools faster. Without real numbers, no one knows. I'm happy to call Steve at AVO if he'll give me some mo' info, but he's not going to like the $500 price drop from the competition--and there is the rub. We may be better off with member measurements.

08-30-2004, 01:59 PM	#13
AZScoobie Scooby Guru Member#: 8785 Join Date: Jul 2001 Chapter/Region: SWIC Vehicle: 02 [email protected] netcom.com	Good work man. C

08-30-2004, 02:31 PM	#14
happasaiyan Scooby Specialist Member#: 37731 Join Date: Jun 2003	where does the SR50 stand? i would think everything the same as the SR55, but with a smaller wheel.

09-02-2004, 06:32 PM	#16
bboy Scooby Specialist Member#: 56468 Join Date: Mar 2004 Chapter/Region: NWIC Location: Seattle, WA Vehicle: 04 Improved STI Dirty White	I love how Clark state's things emphatically, leaving no doubt about what he thinks. Duncangrant, I could not agree more. I just posted in another thread my agreement with you. I looked the damn physic's law up again. It's Poiseuille's Law: Resistance R = (8hL)/pr4 or Volumetric Flow Q = (pDPr4)/(8hL) where R = resistance, P = pressure, n = mean flow velocity, Q = volumetric flow rate, L = tube length, h = fluid viscosity, and r = tube radius, D=? If you change the r of the nozzle even the tiniest bit it's magnified to the fourth power. I don't know the units in the equation, but if r changes by 2, that change has a 222*2 effect, or 16 fold. So, say you are flowing 500 cfm through a 2.5 inch nozzle--> little change in nozzle diameter, 0.5 inches--> the change in the nozzle area is 4.9 in^2 to 3.14 in2 or 3.14/4.9= 0.61 = 61%-->you expect 39% less flow based on area, but the actual flow drops 96% according to Poisseille--> you go from 500 cfm to 21 cfm. (That is not a typo, 21 cfm, I did make a bunch of reasonable assumptions about the other values). It's not an accurate model for exhaust, and there is a lot more going on, but that one apeture at the can make a huge difference in the capacity for turbine flow.