Bolted up a 20g-9.....untuned !!
i hope buschur wont mind that i am posting this. its on his forum. Ted B was exactly right
Wow, what a two day testing frenzy we just went through.
Robert from Forced Performance contacted me last week and said he had a bunch of testing he would like to have done. I thought he wanted me to do the testing. Ended up he wanted to fly in, bring some testing equipment and be involved in the testing to make sure things were conducted on the up and up and very scientifically.
I didn't know it until this weekend but Robert isn't just a dude that opened a turbo shop. Robert is a very smart guy. He has a degree in engineering and looks at things very differently than I do. Many of my things are trial and error but the error is honestly not very often after 17 years of doing 4G63's
The mission was to test the 20G 6 bladed wheel that we were involved so heavily with turdochargerscom to make, test the current turbo we call a 20G which actually uses a 5 bladed compressor wheel and to test a new compressor wheel that will be the next generation wheel to come.
I had to do wiring and such on my RS for this test. Just making dyno pulls was not enough for Robert. He wanted to know "why" things happen more than just, "yep this one makes more power".
An additional air temp sensor was installed (2 total) and an additional 5.2 bar map sensor (2 total). I wired them all into a patch harness between the AEM and factory wiring and also wired in the Innovate LM1 system while I was at it.
First up to test was the turbo that was already on the car. This was our EVO9 20G-9-5. For those of you not up to date on this what that turbo is is this. A EVO9 turbo that has had a 5 bladed compressor wheel installed, the turbine wheel is also modified and the turbo uses a 10.5 cm turbine housing. The turbine housing and manifold for the testing were our standard ported units and the 02 housing was the SS unit we sell.
First test was typical. The car made a little more power than last time it was on the dyno and I was posting about it but still in line with what it has been making. The numbers were around 420 whp and 475 ft lbs of torque. Boost peaked at 28 psi and fell to 25 psi. I used our HD actuator during all testing and the turbine housing was never removed for any tests. Just the turbo cartidge was slid out and new one installed each time.
What we were curious to see what the tempurature rise during the pulls and the back pressure reading in the turbine housing. I know as many of you do that generally on a fast spooling street turbo a 2:1 ratio is common. In other words if you have 20 psi of boost in the intake manifold you will have 40 psi in the exhaust manifold. Not great for top end power but makes the turbo spool up very fast.
To our suprise, or my surprise atleast, the ratio was nowhere near 2:1. With 28 psi of boost the backpressure only got as high as 32 psi. Intake air temps were barely changing, a maximum of a 2 degree increase on any pull. This is all actually quite amazing.
To take a step back for a second. Around 10 years ago I performed this same test on my 1991 Eagle Talon. Using the same turbo, a 20G with the TD05H turbine, big clip on the turbine wheel and in the 7cm housing. I can't at this point in time remember the exact numbers but I do know I never ran over 27 psi of boost in that car, so we can assume the boost was at its highest 27 psi. The back pressure I could not read with a 30 psi gauge. I had a very hard time back then finding a gauge that went higher. I don't remember what I used but the back pressure in that turbo was over 50 psi in the manifold.
At this point in the testing of the EVO I knew these numbers were really good and told Robert the story about the 1991 Talon. He agreed that he had seen the same thing. I asked why the huge difference.
One of the big differences is the much improved twin scroll 10.5 cm housing the EVO has on it, the 02 housing is another large improvement. Then Robert's answer added to these two was the improvement of the aerodynamics of the 5 bladed wheel he makes and we use in our "20G's". IF the compressor wheel is too large for the turbine wheel to drive it will increase the back pressure drastically. I also learned this weekend that a mix match of the wrong sized compressor vs. turbine will cause overheating, excessive back pressure AND compressor surging. Very interesting indeed.
Not completely convinced we made a turbo change. I installed a real 20G turbo. This is the copy of the Mitsubishi wheel, 6 bladed that I worked so hard on with turdochargerscom and then had them back stab us on.
The results were very different. The 5 bladed wheel is quite a bit smaller than the 20G wheel and in my mind a larger wheel makes more power, period. This is not necessarily the case. The last time the 20g 6 bladed wheel was on my car I was still on the HKS 280 cams and the car was making around 380 whp-390 ft lbs of torque. Now the car has had a cam change to the Revolvers and more tuning and is making 420-475 ft lbs. Well it won't make that power with the 6 bladed wheel. First dyno pull the car only made 385 whp-397 ft lbs. The boost was double checked and then the datalog. The larger compressor wheel was in fact hurting the back pressure as Robert said it would. Not the 50+ psi of back pressure I saw in the DSM days but a large raise over the 5 bladed wheel. 28 psi of boost pressure was causing 48 psi of back pressure in the turbine housing. Intake temps were climbing 7 degrees. Now as I was learning more about turbos and how they do and don't work the intake air temps climb I noticed before made more and more sense. AFR's were also about .5 of an air fuel point richer with this turbo, showing the air flow in CFM was not there.
We did further testing than what I am just writing about. I don't want to write a novel. Tests were performed from 2,000 rpm to 7,000 in 3rd and 4th gears. We also did testing from 2,000 rpm to 4500 rpm in 5th. These were done on both turbos.
As we continued with all this testing I questioned how a larger compressor wheel makes the AFR go richer. It was explained to me that it doesn't matter how big the compressor wheel is if the exhaust wheel can't drive it efficiently and the exhaust can't get out of the engine past the turbine wheel. O.K. it makes sense to me know.
Last on our test schedule was a brand new compressor wheel. The results were very good but still not perfect so it is back to the drawing board for Robert. In a few more weeks there will be another unit for us to test and gather this same data on. Depending on what future testing produces there could be another turbo coming out.
It never ends, modifying that is, if you are as passionate about all this as we are here and Robert is at his shop, parts will continue to evolve and new better things will always be developed and the horsepower limits will continue to grow and be pushed.
I love this stuff. I love to learn, love to test and love the fact that no matter how good we think something is that we can continually push ourselves to do even better.
Thanks for reading!!
David Buschur
Robert from Forced Performance contacted me last week and said he had a bunch of testing he would like to have done. I thought he wanted me to do the testing. Ended up he wanted to fly in, bring some testing equipment and be involved in the testing to make sure things were conducted on the up and up and very scientifically.
I didn't know it until this weekend but Robert isn't just a dude that opened a turbo shop. Robert is a very smart guy. He has a degree in engineering and looks at things very differently than I do. Many of my things are trial and error but the error is honestly not very often after 17 years of doing 4G63's
The mission was to test the 20G 6 bladed wheel that we were involved so heavily with turdochargerscom to make, test the current turbo we call a 20G which actually uses a 5 bladed compressor wheel and to test a new compressor wheel that will be the next generation wheel to come.
I had to do wiring and such on my RS for this test. Just making dyno pulls was not enough for Robert. He wanted to know "why" things happen more than just, "yep this one makes more power".
An additional air temp sensor was installed (2 total) and an additional 5.2 bar map sensor (2 total). I wired them all into a patch harness between the AEM and factory wiring and also wired in the Innovate LM1 system while I was at it.
First up to test was the turbo that was already on the car. This was our EVO9 20G-9-5. For those of you not up to date on this what that turbo is is this. A EVO9 turbo that has had a 5 bladed compressor wheel installed, the turbine wheel is also modified and the turbo uses a 10.5 cm turbine housing. The turbine housing and manifold for the testing were our standard ported units and the 02 housing was the SS unit we sell.
First test was typical. The car made a little more power than last time it was on the dyno and I was posting about it but still in line with what it has been making. The numbers were around 420 whp and 475 ft lbs of torque. Boost peaked at 28 psi and fell to 25 psi. I used our HD actuator during all testing and the turbine housing was never removed for any tests. Just the turbo cartidge was slid out and new one installed each time.
What we were curious to see what the tempurature rise during the pulls and the back pressure reading in the turbine housing. I know as many of you do that generally on a fast spooling street turbo a 2:1 ratio is common. In other words if you have 20 psi of boost in the intake manifold you will have 40 psi in the exhaust manifold. Not great for top end power but makes the turbo spool up very fast.
To our suprise, or my surprise atleast, the ratio was nowhere near 2:1. With 28 psi of boost the backpressure only got as high as 32 psi. Intake air temps were barely changing, a maximum of a 2 degree increase on any pull. This is all actually quite amazing.
To take a step back for a second. Around 10 years ago I performed this same test on my 1991 Eagle Talon. Using the same turbo, a 20G with the TD05H turbine, big clip on the turbine wheel and in the 7cm housing. I can't at this point in time remember the exact numbers but I do know I never ran over 27 psi of boost in that car, so we can assume the boost was at its highest 27 psi. The back pressure I could not read with a 30 psi gauge. I had a very hard time back then finding a gauge that went higher. I don't remember what I used but the back pressure in that turbo was over 50 psi in the manifold.
At this point in the testing of the EVO I knew these numbers were really good and told Robert the story about the 1991 Talon. He agreed that he had seen the same thing. I asked why the huge difference.
One of the big differences is the much improved twin scroll 10.5 cm housing the EVO has on it, the 02 housing is another large improvement. Then Robert's answer added to these two was the improvement of the aerodynamics of the 5 bladed wheel he makes and we use in our "20G's". IF the compressor wheel is too large for the turbine wheel to drive it will increase the back pressure drastically. I also learned this weekend that a mix match of the wrong sized compressor vs. turbine will cause overheating, excessive back pressure AND compressor surging. Very interesting indeed.
Not completely convinced we made a turbo change. I installed a real 20G turbo. This is the copy of the Mitsubishi wheel, 6 bladed that I worked so hard on with turdochargerscom and then had them back stab us on.
The results were very different. The 5 bladed wheel is quite a bit smaller than the 20G wheel and in my mind a larger wheel makes more power, period. This is not necessarily the case. The last time the 20g 6 bladed wheel was on my car I was still on the HKS 280 cams and the car was making around 380 whp-390 ft lbs of torque. Now the car has had a cam change to the Revolvers and more tuning and is making 420-475 ft lbs. Well it won't make that power with the 6 bladed wheel. First dyno pull the car only made 385 whp-397 ft lbs. The boost was double checked and then the datalog. The larger compressor wheel was in fact hurting the back pressure as Robert said it would. Not the 50+ psi of back pressure I saw in the DSM days but a large raise over the 5 bladed wheel. 28 psi of boost pressure was causing 48 psi of back pressure in the turbine housing. Intake temps were climbing 7 degrees. Now as I was learning more about turbos and how they do and don't work the intake air temps climb I noticed before made more and more sense. AFR's were also about .5 of an air fuel point richer with this turbo, showing the air flow in CFM was not there.
We did further testing than what I am just writing about. I don't want to write a novel. Tests were performed from 2,000 rpm to 7,000 in 3rd and 4th gears. We also did testing from 2,000 rpm to 4500 rpm in 5th. These were done on both turbos.
As we continued with all this testing I questioned how a larger compressor wheel makes the AFR go richer. It was explained to me that it doesn't matter how big the compressor wheel is if the exhaust wheel can't drive it efficiently and the exhaust can't get out of the engine past the turbine wheel. O.K. it makes sense to me know.
Last on our test schedule was a brand new compressor wheel. The results were very good but still not perfect so it is back to the drawing board for Robert. In a few more weeks there will be another unit for us to test and gather this same data on. Depending on what future testing produces there could be another turbo coming out.
It never ends, modifying that is, if you are as passionate about all this as we are here and Robert is at his shop, parts will continue to evolve and new better things will always be developed and the horsepower limits will continue to grow and be pushed.
I love this stuff. I love to learn, love to test and love the fact that no matter how good we think something is that we can continually push ourselves to do even better.
Thanks for reading!!
David Buschur
Originally Posted by nrcooled
The simple fact is that if your air/fuel readings are the same or lower that is a good indication that you aren't pushing as much air or less air with the 20G then with your previous turbo.
I would take your car to get tuned on a dyno to figure out exactly where you're at.
I would take your car to get tuned on a dyno to figure out exactly where you're at.
It would seem that way, but from what I've found from logging different boost levels, and even AFR's after certain mods, AFR tends to richen up if airflow is increased. Remember, we're not driving around with a carburetor for fuel management the car has a very competent ECU and when it see airflow exceed a certain value it will richen things up as a safeguard. That is one of the reasons these cars need to be tuned to get the most power from mods. When I turn the boost up past 21psi(where the car was tuned, 11.5afr) I notice that AFR's go in to the mid 10s. I even noticed that after my header was installed, with boost at 21psi, the AFR's went into low10's and even high nines. I then went back and leaned it back out and the car felt much stronger.
Originally Posted by Drifto
It would seem that way, but from what I've found from logging different boost levels, and even AFR's after certain mods, AFR tends to richen up if airflow is increased. Remember, we're not driving around with a carburetor for fuel management the car has a very competent ECU and when it see airflow exceed a certain value it will richen things up as a safeguard. That is one of the reasons these cars need to be tuned to get the most power from mods. When I turn the boost up past 21psi(where the car was tuned, 11.5afr) I notice that AFR's go in to the mid 10s. I even noticed that after my header was installed, with boost at 21psi, the AFR's went into low10's and even high nines. I then went back and leaned it back out and the car felt much stronger.
I thought the ECU will only compensate when it's in open loop.
Closed loop is what the flash controls correct? This would be a constant thing.
Not sure if I got closed loop and open loop confused, but hopefully you see what I'm saying.
Closed loop is what the flash controls correct? This would be a constant thing.
Not sure if I got closed loop and open loop confused, but hopefully you see what I'm saying.
Closed loop systems utilize feedback...i.e., feedback from the sensors. Open loop doesn't. The question then really is WOT open loop or closed, or a partial mix with feedback only from the knock sensor for instance. In all likelihood it is the latter.
Evolving Member
Joined: Jun 2005
Posts: 135
Likes: 0
From: World's largest Archipelago
Originally Posted by Ted B
My results prove differently.
The 6-blade 20G apparently caused excessive backpressure in the exhaust, which is why raising the boost beyond ~23psi in the midrange resulted in NO additional torque, and despite holding 3 psi more at the power peak, created NO more hp than the TME did at less boost pressure. This is indicative of a pressure ratio problem due to poor match between the 6-bladed 20G compressor and the turbine wheel.
The 6-blade 20G apparently caused excessive backpressure in the exhaust, which is why raising the boost beyond ~23psi in the midrange resulted in NO additional torque, and despite holding 3 psi more at the power peak, created NO more hp than the TME did at less boost pressure. This is indicative of a pressure ratio problem due to poor match between the 6-bladed 20G compressor and the turbine wheel.
1) DB tested the 6 bladed 20G to be superior to the 03' turbo
2) Ted tested the TME to be superior to the 6 bladed 20G
3) DB & FP tested the 5 bladed 20G to be superior to the 6 bladed 20G
So we don't know if the 5 bladed 20G is superior to the TME but we do know that it is superior to the 6 bladed 20G and the 03' turbo. So the logical thing for you guys holding the 6 bladed 20G is either to send it in for an upgrade or to post test results of your own contesting Ted's findings. Insinuating that DB fixed the results because of his fallout with turbochargers.com (a good thing to me
) is unfair to the man. We have all seen him made U-turns on his own accord before so why assume he's not doing the same? Arguing that the 6 bladed 20G is equal to the 5 bladed 20G when the available test results show otherwise is a discredit to this community because you are encouraging people to buy the wrong product. Its worse if you profit from such an action. Either prove it or shut up. If you don't believe in DB's testing, you shouldn't be buying the 6 bladed 20G in the first place cause I haven't seen any test results from turbochargers.com and even if there are, I will take a huge pinch of salt with it
Seriously, I wanted to buy the 6 bladed 20G when DB first released it but I didn't simply because I know who built it. I'm glad things turned out better for me this way. I have respect for FP and I will buy what they built without hesitation 
So there....for anyone sitting on the fence right now, either buy the 5 bladed 20G cause its the least risky choice or don't buy one at all. If you cheap out to save $300 and don't get the result, I think you deserve wasting your $1000
I think it's safe to say that the 20G-9-5 makes more power than the TME based on the dyno numbers and track times. The stock IX is also better than the TME, but the TME is still a good upgrade for 03/04s.
After reading DB's results on the 20g testing, I would be going with the 5 blade design or staying w/ the stock turbo. If everyone takes the time to read what David wrote, they would understand exactly WHY the 6 blade is no good, and we wouldnt be having this debate.
Originally Posted by davidbuschur
I then pulled the inlet off the turbo to see what exactly I had on the car, found it is the current 20g-9-5 bladed wheel. Interesting. Interesting enough in fact that I put the normal Mitsubishi style 20G wheel'd turbo back on the car. Guess what, AIT's went back up. They climbed every pull by 4-5 degrees on a single 3rd gear run. HP stayed about the same as before, spool up was about the same etc. The turbos are close but the IAT's are pretty amazing. It shows how much more efficient the 20g-9-5 wheel is.
Another perspective in looking at the two turbos is this...
20G EVO9 5 blade $1350
20G EVO9 6 blade $939 shipped
Is having "about the same hp" and "about the same spool" worth $400-$500?
Originally Posted by EFIxMR
Is having "about the same hp" and "about the same spool" worth $400-$500?
As far as I can see things at this point, this is more or less how the entire picture crystallizes . . .
The larger 10.5cm2 turbine housing enables the stock 9.8cm2 16G to give a bit more pop for $250. The TME adds the quicker-spooling TiAl turbine wheel to this, and is the more or less the pinnacle of the EVO VIII turbos. Add the new EVO-IX compressor cover to the TME, and compressor efficiency improves just a little bit further where you need it.
The general consensus agrees that it's difficult to improve upon the 16G within the confines of its shells. FP managed to eek out a little more efficiency from the TME with its WR compressor. Adding the slightly more efficient EVO-IX cover helped this theme a bit more, and the result of this is the 5-blade 20G-9, which is a bit of a misnomer because it's not really a 20G at all. As for what more can be eeked out of this, it appears to range from small to non-existent. If anything can be gained at all, it really wouldn't be worth anything except to a first-time buyer.
As for the real 6-blade 20G compressor, it just seems to create a pressure imbalance when mated to the 16G turbine. It may have initially appeared to be passable when compared to a regular EVO VIII turbo with small hotside, but in reality, it gives up too much for too little as compared to even an EVO VIII 16G with 10.5cm2 hotside as per my testing.
This is all my opinion, called as I see it, formed over 1.5 years of both first and second-hand observations.
Last edited by Ted B; Apr 3, 2006 at 07:11 PM.
It's just funny because I owned a WR before... And its detractors... you know who and them... would say the same story of how it didn't work and the real 20G was great. Now they tell me this new and improved WR is the newest best thing around, and the now 20G is a waste of money because they can't sell it anymore. Are we playing with cars here or are we playing politics?
Pardon me if I take what they now say with a gain of salt, but I've learned consistency is the main ingredient of truth. If they never had a fall out with TC.com, when would the community have been disclosed that this 20G blows hot air?
Pardon me if I take what they now say with a gain of salt, but I've learned consistency is the main ingredient of truth. If they never had a fall out with TC.com, when would the community have been disclosed that this 20G blows hot air?
Last edited by EFIxMR; Apr 3, 2006 at 07:50 PM.
