GSLENK
10+ Year Contributor
- 1,416
- 54
- May 25, 2011
-
DC,
Maryland
Sure we have a few threads on what a BLT should result in, but we don't have much real world data.
If anyone can post up a log of a very good boost leak test I would greatly appreciate it. If you want to clip it for me, I want a few seconds from the max MAP pressure, all the way down to "0" pressure.
I took a log of my latest boost leak test. I started at ~19 psi. My compressor couldn't keep up with the leak that happened at the inlet of the turbo (my BLT tester sucks, so overall this data is skewed terribly).
In a nutshell, I exported my data to excel, and sampled it by interval (to remove extraneous computation time). Then I plotted MAP vs time, and did some analysis.
My whole system with a crappy tester (that audibly leaks itself at ~14psi probably sooner) took a little over 2 minutes to leak down from 19 psi to 0 psi. I was curious to see what effect this had at various boost levels, and how much performance is really hurting.
By doing an 8th degree polynomial regression , I was able to mimic my test results almost perfectly with an equation, from about 1-100 seconds of the total 0-180 seconds logged. Then taking the derivative at select points I can estimate my psi lost per second.
Here's the end result of my calculations (how much psi lost per second @ different pressure levels. the rate is decrease in psi per second)
@18 psi, lost .919 psi/s
@16psi, .78psi/s
@14psi, .63psi/s
@10psi, .347
@5, .141
@3, .09
@1, .03
Again this is heavily skewed, I am just trying to find out what a real acceptable boost leak is, and find how much an acceptable level of leak can affect performance.
Anyone want to hook it up with some good boost leak down logs?
If anyone can post up a log of a very good boost leak test I would greatly appreciate it. If you want to clip it for me, I want a few seconds from the max MAP pressure, all the way down to "0" pressure.
I took a log of my latest boost leak test. I started at ~19 psi. My compressor couldn't keep up with the leak that happened at the inlet of the turbo (my BLT tester sucks, so overall this data is skewed terribly).
In a nutshell, I exported my data to excel, and sampled it by interval (to remove extraneous computation time). Then I plotted MAP vs time, and did some analysis.
My whole system with a crappy tester (that audibly leaks itself at ~14psi probably sooner) took a little over 2 minutes to leak down from 19 psi to 0 psi. I was curious to see what effect this had at various boost levels, and how much performance is really hurting.
By doing an 8th degree polynomial regression , I was able to mimic my test results almost perfectly with an equation, from about 1-100 seconds of the total 0-180 seconds logged. Then taking the derivative at select points I can estimate my psi lost per second.
Here's the end result of my calculations (how much psi lost per second @ different pressure levels. the rate is decrease in psi per second)
@18 psi, lost .919 psi/s
@16psi, .78psi/s
@14psi, .63psi/s
@10psi, .347
@5, .141
@3, .09
@1, .03
Again this is heavily skewed, I am just trying to find out what a real acceptable boost leak is, and find how much an acceptable level of leak can affect performance.
Anyone want to hook it up with some good boost leak down logs?
