FP Black Ball Bearing tested - documentary results inside.
also was this on a 2.3?
Boost response between gears should be the same regardless of weather temperatures.
Yup this was on a 2.3, but i didn't really notice any less "turbo lag" from the larger engine, just lower "boost threshold".
Datalog just RPM, Boost and Speed and evoscan will log fast enough for what people have been asking you for 5 pages now.
Last edited by razorlab; Aug 1, 2011 at 03:25 PM.
To increase your logging frequency... make sure your power supply is connected to your laptop in your car (and plugged into your AC converter)
Reduce the number of channels to RPM, Time, Boost pressure, & TPS.
Go out and get some data points for us. This is the test routine I did when testing dynamic response (we called it T90 times) or the time it takes to hit 90% of rated torque @ that RPM.
Pick a high gear, but not too high that you'll be unsafe...like 3th gear on a flat road.
if you target RPM to test is 3000RPMS... drive your car to 5k rpms and lift off throttle. Coast down to 3k, and go WOT as soon as you hit 3k rpms. Do this 3 times. Do the same for 3.5k, 4k, 4.5k, 5k, 5.5k, 6k, etc. When you get back to your desk measure the amount of time it takes to reach 90% of peak torque from the instant you went WOT. Average the 3 data points and create ONE data point for 3k RPMS. The data point will be in milliseconds...or whatever time unit you're accurate to. Do this for each RPM sample, and generate a curve. You will have created a plot of the transient response time.
You could also do a 1,2,3,4 "race" type log...loggin the same things and measuring the time between achieving target boost pressure. I think there are FAR more external variables in this type of data...and will be for the most part in-conclusive due to differing turbospeeds, shift times, etc.
off topic...If you have a nice EMS you can log the natural frequency of the sensor...which will give you a retarded number of data points, which isn't necessary to really quantify what we're looking for.
Reduce the number of channels to RPM, Time, Boost pressure, & TPS.
Go out and get some data points for us. This is the test routine I did when testing dynamic response (we called it T90 times) or the time it takes to hit 90% of rated torque @ that RPM.
Pick a high gear, but not too high that you'll be unsafe...like 3th gear on a flat road.
if you target RPM to test is 3000RPMS... drive your car to 5k rpms and lift off throttle. Coast down to 3k, and go WOT as soon as you hit 3k rpms. Do this 3 times. Do the same for 3.5k, 4k, 4.5k, 5k, 5.5k, 6k, etc. When you get back to your desk measure the amount of time it takes to reach 90% of peak torque from the instant you went WOT. Average the 3 data points and create ONE data point for 3k RPMS. The data point will be in milliseconds...or whatever time unit you're accurate to. Do this for each RPM sample, and generate a curve. You will have created a plot of the transient response time.
You could also do a 1,2,3,4 "race" type log...loggin the same things and measuring the time between achieving target boost pressure. I think there are FAR more external variables in this type of data...and will be for the most part in-conclusive due to differing turbospeeds, shift times, etc.
off topic...If you have a nice EMS you can log the natural frequency of the sensor...which will give you a retarded number of data points, which isn't necessary to really quantify what we're looking for.
Last edited by R/TErnie; Aug 1, 2011 at 03:35 PM.
keep your bull**** comments out of my thread.
To increase your logging frequency... make sure your power supply is connected to your laptop in your car (and plugged into your AC converter)
Reduce the number of channels to RPM, Time, Boost pressure, & TPS.
Go out and get some data points for us. This is the test routine I did when testing dynamic response (we called it T90 times) or the time it takes to hit 90% of rated torque @ that RPM.
Pick a high gear, but not too high that you'll be unsafe...like 3th gear on a flat road.
if you target RPM to test is 3000RPMS... drive your car to 5k rpms and lift off throttle. Coast down to 3k, and go WOT as soon as you hit 3k rpms. Do this 3 times. Do the same for 3.5k, 4k, 4.5k, 5k, 5.5k, 6k, etc. When you get back to your desk measure the amount of time it takes to reach 90% of peak torque from the instant you went WOT. Average the 3 data points and create ONE data point for 3k RPMS. The data point will be in milliseconds...or whatever time unit you're accurate to. Do this for each RPM sample, and generate a curve. You will have created a plot of the transient response time.
You could also do a 1,2,3,4 "race" type log...loggin the same things and measuring the time between achieving target boost pressure. I think there are FAR more external variables in this type of data...and will be for the most part in-conclusive due to differing turbospeeds, shift times, etc.
off topic...If you have a nice EMS you can log the natural frequency of the sensor...which will give you a retarded number of data points, which isn't necessary to really quantify what we're looking for.
Reduce the number of channels to RPM, Time, Boost pressure, & TPS.
Go out and get some data points for us. This is the test routine I did when testing dynamic response (we called it T90 times) or the time it takes to hit 90% of rated torque @ that RPM.
Pick a high gear, but not too high that you'll be unsafe...like 3th gear on a flat road.
if you target RPM to test is 3000RPMS... drive your car to 5k rpms and lift off throttle. Coast down to 3k, and go WOT as soon as you hit 3k rpms. Do this 3 times. Do the same for 3.5k, 4k, 4.5k, 5k, 5.5k, 6k, etc. When you get back to your desk measure the amount of time it takes to reach 90% of peak torque from the instant you went WOT. Average the 3 data points and create ONE data point for 3k RPMS. The data point will be in milliseconds...or whatever time unit you're accurate to. Do this for each RPM sample, and generate a curve. You will have created a plot of the transient response time.
You could also do a 1,2,3,4 "race" type log...loggin the same things and measuring the time between achieving target boost pressure. I think there are FAR more external variables in this type of data...and will be for the most part in-conclusive due to differing turbospeeds, shift times, etc.
off topic...If you have a nice EMS you can log the natural frequency of the sensor...which will give you a retarded number of data points, which isn't necessary to really quantify what we're looking for.
tps
rpm
boost
and the other stuff evoscan logs by default.. i did order a car charger that connects to the cig lighter that will be sufficient enough?
also.. i can make it so evoscan doesnt display the logged data on screen which helps logging speed also.
ill do everything you said above and report back .. but is the cig lighter sufficient enough for power source? its a sony vaio 2ghz dual core, 4gb mem, etc.. its a decent speed laptop.
we both know its not instant.. i just said instant to express the nature of how it feels compared to the journal. ill do the datalogs and post them up. the smart remarks are not necessary. 200rpm difference on a 2.3 with a black is quite significant difference in feeling of the response.. so i expressed it as "instant". the journal itself is impressive.. add 200rpm and you have amazing based on the cfm the turbos capable of producing.
just bought this awhile also http://www.amazon.com/I-Tec-T2022-Tr.../dp/B000W7F69G found it at autozone so ill do some testing tonight after i fill up on e85 again.
Ok i just did what R/Ternie said and did 3k 4k and 5k.. all 3rd gear, and all on the interstate. back to back so heatsoaking may have occured but who wants to plot the logs for me? i will send them over.
I don't know about you guys, but 200rpm in spool is not worth to cry home to.
This test you just did, tscomp, is worthless without the same data on a JB turbo.
I think a lot of people are getting carried away with all of this ball bearing craze. At best, all it will have is slightly better responce, but much better relaibility. Is that worth over 3 grand?
This test you just did, tscomp, is worthless without the same data on a JB turbo.
I think a lot of people are getting carried away with all of this ball bearing craze. At best, all it will have is slightly better responce, but much better relaibility. Is that worth over 3 grand?