Friday, May 24, 2013
Roush Boost Gauge Install and Boost
Roush Boost Mechanical LED Boost Gauge
The instructions from Roush are extremely poor. The gauge comes with a "T" shaped plastic fitting and the instructions only say "plug it to boost/vacuum line". No one has any idea at all which line is that.
The best way to install the boost gauge is to tap into the opening on the manifold's passenger side, just above cylinder 8. There is a hexagonal bolt that Roush plugs in at factory to seal the port.
Buy a 1/8 NPT to 1/8" Barbed fitting for $2. Remove the factory bolt, wrap the NPT fitting with a line of Teflon sealant and plug it into that hole. Attach the boost gauge hose to the barbed fitting. Done.
I finally managed to get my Roush gauge to work yesterday. To my big surprise, it was only reading about 7-7.5psi boost and -16/-18 in.Hg at idle. Vacuum reading was perfect but I was surprised with the boost. 85mm pulley normally is rated at 8.5psi.
VMP tuning wrote me and explained that the manifold boost eventually depends on the tune and how the TVCT settings are calibrated. Normally a Ford Racing or Roush tune would show up more boost in the manifold whereby VMP's tune actually reduced it by close to 1 psi, yet still making 661hp at crank!
The boost kicks in at around 5psi from low rpm and immediately climbs to 7psi as of 1500rpm. Stays steady all the way to red line.
I will install the 82mm pulley today which is supposed to increase the boost by 1 psi. I hope to read about 8-8.5psi at the gauge.
Tuesday, May 21, 2013
Final Dyno Figures with 85mm Pulley
Dyno tested the car on 18.05 at Oscar Halter's dyno shop in Frauenfeld.
Roush OEM calibration is designed to deliver approximately 489 hp / 590nm at wheels ( 575hp/ 680nm at crank). I did not even bother with it and went straight ahead with VMP Tuning's genious Justin Starkey's performance tune.
The car put down 562hp / 648 nm at wheels ( 661hp/762nm at crank), about 86hp and 82nm more power than Roush OEM calibration. All this with only using the 85mm stock pulley that makes about 8-8.5psi boost. The 2013 GT500 has 661hp / 849nm torque. The car has same hp level as the 2013 GT500 despite the missing torque that GT500 has due to it's bigger displacement. The difference is 87nm but it'd be interesting to see how both cars would perform at a drag race.
We spent significant timin in datalogging, AFR, ignition timing, battery/alternator voltage. As in the graph below, the AFR measured at the exhausts were dead spot on 11-11.2 until 7200rpm. Although it climbed from 11.0 to 11.4 between 6800-7100, it was still in safe zone. Assuming that tail pipe sniffer reads about 2% leaner than actual AFR, these numbers would be 10.8-11.2 before the CATs which is absolutely safe. The alternator voltage was also steady at 14.2-14.4 volts meaning fuel pumps were getting steady supply of power with no indication of running out of duty cycle.
I have the green light from Justin to try the smaller 82mm pulley which is supposed to increase the boost by 1psi to about 9-9.5, using closed airbox.
The pulley is expected to safely add another 20-30 hp and 30-35nm torque at the wheels which should bring the final numbers to 690hp/800nm at crank.
The next step is the 79mm pulley which adds another 0.5psi to make it around 10 psi in total. Now i am a little skeptical about going that low but it is always the best to go step by step with proper datalogging. The 79mm will put the injectors at their limit and I assume the car wont be safe to run above 7000rpm with stock injectors. The installed fuel pump voltage booster might definitely help to keep AFR steady but I dont think 79mm is a good option without larger injectors. Upgrade to 79mm should definitely put the power levels to around 705hp / 815nm. Which is way more than any street car can use without proper, sticky, drag radials.
Roush OEM calibration is designed to deliver approximately 489 hp / 590nm at wheels ( 575hp/ 680nm at crank). I did not even bother with it and went straight ahead with VMP Tuning's genious Justin Starkey's performance tune.
The car put down 562hp / 648 nm at wheels ( 661hp/762nm at crank), about 86hp and 82nm more power than Roush OEM calibration. All this with only using the 85mm stock pulley that makes about 8-8.5psi boost. The 2013 GT500 has 661hp / 849nm torque. The car has same hp level as the 2013 GT500 despite the missing torque that GT500 has due to it's bigger displacement. The difference is 87nm but it'd be interesting to see how both cars would perform at a drag race.
We spent significant timin in datalogging, AFR, ignition timing, battery/alternator voltage. As in the graph below, the AFR measured at the exhausts were dead spot on 11-11.2 until 7200rpm. Although it climbed from 11.0 to 11.4 between 6800-7100, it was still in safe zone. Assuming that tail pipe sniffer reads about 2% leaner than actual AFR, these numbers would be 10.8-11.2 before the CATs which is absolutely safe. The alternator voltage was also steady at 14.2-14.4 volts meaning fuel pumps were getting steady supply of power with no indication of running out of duty cycle.
I have the green light from Justin to try the smaller 82mm pulley which is supposed to increase the boost by 1psi to about 9-9.5, using closed airbox.
The pulley is expected to safely add another 20-30 hp and 30-35nm torque at the wheels which should bring the final numbers to 690hp/800nm at crank.
The next step is the 79mm pulley which adds another 0.5psi to make it around 10 psi in total. Now i am a little skeptical about going that low but it is always the best to go step by step with proper datalogging. The 79mm will put the injectors at their limit and I assume the car wont be safe to run above 7000rpm with stock injectors. The installed fuel pump voltage booster might definitely help to keep AFR steady but I dont think 79mm is a good option without larger injectors. Upgrade to 79mm should definitely put the power levels to around 705hp / 815nm. Which is way more than any street car can use without proper, sticky, drag radials.
Friday, May 17, 2013
The Drive and Tuning
First Start with Roush OEM Calibration
The car fired up immediately. The install was so perfectly done there was not a single hick up, surge or any computer error code. I drove around with the OEM calibration and to be honest I was a little disappointed. It drove almost like stock at part throttle and Roush tune really kicks the power in when you go WOT and reach about 3-4k rpm. Roush tune lived only for 30 mins or so, until I drove home from Garage Dutli to upload the VMP tune from the legendary tuner Justin Starke.
VMP Tune (http://www.vmptuning.com/)
I haver been reading about VMP and their owner, master tuner, Justin Starke for a long time. He is one of the most reputable tuners who specializes in TVS style blowers. He has tuned 100s of Roush cars and swapped 11-13 GT500 factory twin screw with his own VMP TVS supercharger since years.
From his website you can buy a "remote tuning" for $350 which saves you a ton of money from doing actual dyno tuning in Europe which will easily cost you close to 700-1000 Euros! The way Justin works is; he sends you a tune that is made for your specifications and asks you to datalog the car using SCT's LiveLink logging software. All you need to do is datalog idle, rpm range in neutral gear and do either a 2nd, 3rd or 4th gear street WOT pull. The higher the gear, the better.
Now, the difference between Roush OEM and VMP performance tune is day and night. The car was a complete different beast and so much more responsive with VMP's tune. Not to mention, Roush is extremely conservative tune that hides away an easy 60-80hp. VMP releases these hidden horsepower without a pulley change. It is estimated that a 575hp Roush OEM package is easily upgraded to 630-640hp without any pulley swap.
The car has absolutely zero driveability issues. Idle is smooth, part throttle has no surge or bucking issues. Throttle response was a little snappy at first but I rapidly got used to it. The power level is insane. Burning tires is almost unavoidable in 2nd gear and even sometimes in the 3rd.
I will soon dyno the car to proove the results but here take a look my datalog after I uploaded the VMP tune:
AFR Log
The pull was made on 3rd gear from 1250rpm to 6950rpm.
The AFR on a supercharged mustang is usually set to low 11s, typically 11.0-11.2 range to be on the safe side with the boost while maximizing power. My AFR was measured starting at 11.7 at 1500rpm, dropped to 11.2 as of 2500 and remained constant almost all the way to the top of the rpm range where it dropped to 11-11.1 above 5500rpm. Just as expected and just as how it should be.
Spark Advance
Here is the graph of my spark advance timing. The thick blue line is the timing where the thin yellow line is the knock sensor, starting at zero right at 48th second.
In the recent days, I learned so much about how knock sensors in 11-13 Mustang GT and Spark timing work together. I always thought that the parameter in SCT livelink software that measures knock sensors, was actually indicating whether the car had a knock or not. Which is completely false. The Knock sensor, whether being zero, a + or - number, indicates if the PCM is commanded to advance or retard timing. If the knock sensor reads a positive value like +1, it means it is commanding the PCM to retard the spark timing by 1 degree because it actually measures less than ideal conditions to make extra power. Retarding the timing causes cylinder pressure to drop and helps avoiding a possible detonation. If the knock sensor reads a value like "-2" it means, the ideal conditions exist and spark timing can be advanced up to 2 degrees to produce more power.
A good performance tune must leave the knock sensors on and must ensure that spark timing is almost instantly adjusted as knock sensors command. As you can see in my graph above, the "yellow" knock sensor drops to -2 and commands increasing the timing which is shown in thick blue line. The timing is advanced up to 17-18 degrees as commanded by the knock sensor to make use of the most power.
One of the most feared reasons why a supercharged engine dies is when the PCM is not able to retard the timing almost instantly when Knock sensor orders to. The reasons can be; a) the knock sensors are left off by the tuner b) the tune and equipment i.e. injectors, spark plugs etc. used are not properly installed so that even though when knock sensor commands a retard, it takes long time for spark to be actually retarded. It is matter of miliseconds a detonation can occur and destroy the engine.
Now take a look at the zoomed section of my Spark advance graph below.
The car fired up immediately. The install was so perfectly done there was not a single hick up, surge or any computer error code. I drove around with the OEM calibration and to be honest I was a little disappointed. It drove almost like stock at part throttle and Roush tune really kicks the power in when you go WOT and reach about 3-4k rpm. Roush tune lived only for 30 mins or so, until I drove home from Garage Dutli to upload the VMP tune from the legendary tuner Justin Starke.
VMP Tune (http://www.vmptuning.com/)
I haver been reading about VMP and their owner, master tuner, Justin Starke for a long time. He is one of the most reputable tuners who specializes in TVS style blowers. He has tuned 100s of Roush cars and swapped 11-13 GT500 factory twin screw with his own VMP TVS supercharger since years.
From his website you can buy a "remote tuning" for $350 which saves you a ton of money from doing actual dyno tuning in Europe which will easily cost you close to 700-1000 Euros! The way Justin works is; he sends you a tune that is made for your specifications and asks you to datalog the car using SCT's LiveLink logging software. All you need to do is datalog idle, rpm range in neutral gear and do either a 2nd, 3rd or 4th gear street WOT pull. The higher the gear, the better.
Now, the difference between Roush OEM and VMP performance tune is day and night. The car was a complete different beast and so much more responsive with VMP's tune. Not to mention, Roush is extremely conservative tune that hides away an easy 60-80hp. VMP releases these hidden horsepower without a pulley change. It is estimated that a 575hp Roush OEM package is easily upgraded to 630-640hp without any pulley swap.
The car has absolutely zero driveability issues. Idle is smooth, part throttle has no surge or bucking issues. Throttle response was a little snappy at first but I rapidly got used to it. The power level is insane. Burning tires is almost unavoidable in 2nd gear and even sometimes in the 3rd.
I will soon dyno the car to proove the results but here take a look my datalog after I uploaded the VMP tune:
AFR Log
The pull was made on 3rd gear from 1250rpm to 6950rpm.
The AFR on a supercharged mustang is usually set to low 11s, typically 11.0-11.2 range to be on the safe side with the boost while maximizing power. My AFR was measured starting at 11.7 at 1500rpm, dropped to 11.2 as of 2500 and remained constant almost all the way to the top of the rpm range where it dropped to 11-11.1 above 5500rpm. Just as expected and just as how it should be.
Spark Advance
Here is the graph of my spark advance timing. The thick blue line is the timing where the thin yellow line is the knock sensor, starting at zero right at 48th second.
In the recent days, I learned so much about how knock sensors in 11-13 Mustang GT and Spark timing work together. I always thought that the parameter in SCT livelink software that measures knock sensors, was actually indicating whether the car had a knock or not. Which is completely false. The Knock sensor, whether being zero, a + or - number, indicates if the PCM is commanded to advance or retard timing. If the knock sensor reads a positive value like +1, it means it is commanding the PCM to retard the spark timing by 1 degree because it actually measures less than ideal conditions to make extra power. Retarding the timing causes cylinder pressure to drop and helps avoiding a possible detonation. If the knock sensor reads a value like "-2" it means, the ideal conditions exist and spark timing can be advanced up to 2 degrees to produce more power.
A good performance tune must leave the knock sensors on and must ensure that spark timing is almost instantly adjusted as knock sensors command. As you can see in my graph above, the "yellow" knock sensor drops to -2 and commands increasing the timing which is shown in thick blue line. The timing is advanced up to 17-18 degrees as commanded by the knock sensor to make use of the most power.
One of the most feared reasons why a supercharged engine dies is when the PCM is not able to retard the timing almost instantly when Knock sensor orders to. The reasons can be; a) the knock sensors are left off by the tuner b) the tune and equipment i.e. injectors, spark plugs etc. used are not properly installed so that even though when knock sensor commands a retard, it takes long time for spark to be actually retarded. It is matter of miliseconds a detonation can occur and destroy the engine.
Now take a look at the zoomed section of my Spark advance graph below.
At 49th second, the yellow line goes positive to +0.5. Which is the knock sensor commanding the spark to retard by 0.5 degrees as the ideal conditions for making extra power at this second do not exist. The command for retard does not mean anything bad. It only shows that at that moment, all the factors that effect horspower generation like fuel condition, supercharger intake temperature, air intake temp, fuel pressure, injectors, cam timing etc. are in an overall system condition that requires timing to be retarded to protect the engine. This happens in almost less than 1/3 of a second.
Now, look at the spark advance line in thick blue. You can see that, immediately, how the timing is retarded by 0.5 degree as knock sensor commanded at the same time. VMPs tune is designed to retard timing up to 6 degrees when needed and only increase by 2 degrees if possible for performance. You can also see in the first spark timing graph that how timing is advanced to max 18 and kept there so that the pcm does not continue advancing timing until it causes detonation. This for me is a good proof that the car is properly tuned and all the sensors in it are working in perfect order to keep the engine safe.
I have heard several stories of destroyed engines and most of them were because either knock sensors were off or the spark was not retarded when needed even though knock sensors were commanding to.
Now I am counting days to take the car to a dyno shop and see how much are the final hp/torque figures.
Roush TVS2300 Installation
Roush TVS2300 Installation
First of all, big thanks to the guys at Dutli Garage in Zurich and especially Nico Seiler for doing an extremely clean and professional installation. These guys are perfectionists and they never rush with their work. Check them out at www.dutli.ch
I will not go through the entire installation details but here are a few important interesting bits that especially helped a lot with the installation.
Machined Timing Cover
The most scary part of the install is machining the front timing cover on the engine. Ford does not ship these timing covers as compatible either with their Ford Racing or the Roush superchargers. There are aluminium ribs and bosses that need to be trimmed down and cut in half, which otherwise make it impossible to install the new belt and idler pulleys of the supercharger. it is a terribly difficult 5-6 hour job to do and extremely scary as not many people are naturally feeling comfortable cutting pieces of the engine covers!
From www.bulecollarperformance.com you can order a premachined timing cover for $150 and just simply choose to replace the timing cover instead. Here is a beautiful picture of how the naked coyote looks like when you remove the timing and valve covers.
Electrical Wiring
I dont know about you guys but the only thing I hate most about any installation is working on wires. It drives me crazy reading through the manual over and over to make sure I cut the right wires and get the repinning correctly. It is a messy job. If you miss a pin or do a wrong connection you might easily damage your electrical system and even if it does not, it will be a huge mess trying to deal with computer error codes after the install.
Remember those VMP Plug&Play wiring harness I mentioned before in the "parts list" page? BUY THEM! They prooved to be extremely efficient and easy to install. Simple plug and play and you wont have to cut, repin any wire. This job would normally take 2 hours alone. With these cable extensions, it is a 10 min task.
Rest of the Installation
With the timing cover cutting and wiring out of the way, the rest of the installation is quite straight forward. Here are some pictures.
I love these gauges. The Aeroforce Interceptor (bottom) is amazing. Monitors and displays two parameters at the same time and it allows the user to set annunciators for alarming when certain parameters exceed limits. For my setup I will set the following alarms
- IAT2: Intake Charge Temperature: 150 F upper limit. Unfortunately it only works in Fahrenheits. 150F is considered as intake charge temperature max safe limit. Above this knock sensors usually fight really hard to pull a lot of timing to protect the engine. Ideally, if you ever see your supercharger setup running above 150F you should seriously consider a heat exchanger upgrade.
Like the AFCO unit VMP sells AFCO Dual Fan Heat Exchanger
- Engine Coolant Temperature: Ideally should stay below 220 F. If it is higher than you will run the risk of detonation if knock sensors can't for some reason pull timing fast enough.
- AFR: Since I can not set an alarm for this because Annunciator functionality works all the time. For example, you can not set a limit to it only at WOT. However I choose to display the measured AFR at Bank1 on the first display.
- Spark Advance: Same story here as AFR. You cant set alarm for spark in certain conditions. So best is, display spark advance on the second display to make sure it is not some value like 20 at upper rpm doing WOT pulls. Normally for 85-82mm pulleys, 17-18 max spark advance is considered safe.
- Boost Gauge: I am still waiting for a special plug that will allow me to connect boost hose to the manifold. The gauge is installed but the hose is just measuring outside air pressure :) Hopefully next week it will work.
First of all, big thanks to the guys at Dutli Garage in Zurich and especially Nico Seiler for doing an extremely clean and professional installation. These guys are perfectionists and they never rush with their work. Check them out at www.dutli.ch
I will not go through the entire installation details but here are a few important interesting bits that especially helped a lot with the installation.
Machined Timing Cover
The most scary part of the install is machining the front timing cover on the engine. Ford does not ship these timing covers as compatible either with their Ford Racing or the Roush superchargers. There are aluminium ribs and bosses that need to be trimmed down and cut in half, which otherwise make it impossible to install the new belt and idler pulleys of the supercharger. it is a terribly difficult 5-6 hour job to do and extremely scary as not many people are naturally feeling comfortable cutting pieces of the engine covers!
From www.bulecollarperformance.com you can order a premachined timing cover for $150 and just simply choose to replace the timing cover instead. Here is a beautiful picture of how the naked coyote looks like when you remove the timing and valve covers.
I would highly recommend this job to be still done by professionals as it is not easy to remove all the pulleys and harmonic balancer.
I dont know about you guys but the only thing I hate most about any installation is working on wires. It drives me crazy reading through the manual over and over to make sure I cut the right wires and get the repinning correctly. It is a messy job. If you miss a pin or do a wrong connection you might easily damage your electrical system and even if it does not, it will be a huge mess trying to deal with computer error codes after the install.
Remember those VMP Plug&Play wiring harness I mentioned before in the "parts list" page? BUY THEM! They prooved to be extremely efficient and easy to install. Simple plug and play and you wont have to cut, repin any wire. This job would normally take 2 hours alone. With these cable extensions, it is a 10 min task.
Rest of the Installation
With the timing cover cutting and wiring out of the way, the rest of the installation is quite straight forward. Here are some pictures.
Drivers side view before Throttle Body installation. It just looks so bad ass!
Front bumper removed for installing the supercharger heat exchanger, intercooler pump and hoses.
Roush TVS2300 pulley and idler system. Those two black idlers are 75mm. With usage of smaller supercharger pulley, these idlers will be upgraded to 90mm lower and 100mm upper idler to keep the belt tension. I will consider this upgrade when I install the 82mm pulley.
I love these gauges. The Aeroforce Interceptor (bottom) is amazing. Monitors and displays two parameters at the same time and it allows the user to set annunciators for alarming when certain parameters exceed limits. For my setup I will set the following alarms
- IAT2: Intake Charge Temperature: 150 F upper limit. Unfortunately it only works in Fahrenheits. 150F is considered as intake charge temperature max safe limit. Above this knock sensors usually fight really hard to pull a lot of timing to protect the engine. Ideally, if you ever see your supercharger setup running above 150F you should seriously consider a heat exchanger upgrade.
Like the AFCO unit VMP sells AFCO Dual Fan Heat Exchanger
- Engine Coolant Temperature: Ideally should stay below 220 F. If it is higher than you will run the risk of detonation if knock sensors can't for some reason pull timing fast enough.
- AFR: Since I can not set an alarm for this because Annunciator functionality works all the time. For example, you can not set a limit to it only at WOT. However I choose to display the measured AFR at Bank1 on the first display.
- Spark Advance: Same story here as AFR. You cant set alarm for spark in certain conditions. So best is, display spark advance on the second display to make sure it is not some value like 20 at upper rpm doing WOT pulls. Normally for 85-82mm pulleys, 17-18 max spark advance is considered safe.
- Boost Gauge: I am still waiting for a special plug that will allow me to connect boost hose to the manifold. The gauge is installed but the hose is just measuring outside air pressure :) Hopefully next week it will work.
Subscribe to:
Posts (Atom)