Re: Any C2 BB owners have there cars on a dyno? Did mine today.
That's a pretty aggressive curve, which should provide a stronger SOTP pull in the low end of the rev range. I'm assuming your total WOT advance is 37 degrees as you stated before. Given your unusually warm recent temperatures I believe the transient detonation will go away when the weather becomes more seasonable. If necessary you can always substitude a medium spring for the light spring to slow the centrifugal, or back the total down a degree or two, but I would not recommend less than 36 total WOT advance. If there is still some ping at that point, slow down the centrifugal.
It sounds to me like you're very close to optimum, and another session in cool temperatures would be a good idea to get some final numbers on the most aggressive map you can run and also a conservative map for use in very hot weather if the engine gets into detonation.
I reviewed your test files with the WinPEP7 software and have some comments.
1. Is it possible that the RFI shielding on the big block wires were blocking the signal to the mag pickup.(?) This should be solveable for the next session.
2. As is typical the pulls start at about 3000. Since we spend a lot of time below this level, start the pulls at 1000. You want the entire HP/torque curve throughout the useable range. Many dyno operators will resist this. You can test that the engine will accept WOT at 1000 by simply punching it at 1000 in third and fourth gear on the road. The engine should pull smoothly. If not, there's something wrong with the tune. Pulling from 1000 will nail the bottom ends of the 80 and 90 percent torque bandwidths.
3. I could not make any sense out of the A/F data. It was nearly constant across the range, so I don't think it was providing an accurate reading.
4. You can add notes to the test files. When you are looking at the graph, all the files displayed are listed above the graph. Right click on a listed file and you get a menu. Click "Run info" and you'll see a "run notes" text box. You can type in data relating to the configuration, which should correspond to your test plan - details of the baseline config. for the first couple of reference pulls, and then any changes you make for subsequent pulls. This provides a permanent record of the what you did during the session and the performance curves for each config.
If you do another session this fall in cooler weather, my suggestion is to see if the engine will tolerate two light springs in road testing. If so, then do pulls at 36, 38, and 40 degrees WOT advance to find what is optimum. Then you can test this optimum setting with a light-medium and medium-medium spring combinations to determine the effect of slowing the centrifugal.
It's too bad you can't widen the LSA of the cam. My Cosworth Vega has about 220 degrees .050" lifter rise (above the tops of the clearance ramps) duration on both cams, but the IPOML is 102 deg. ATDC, and the LSA is 106 deg. In order to get a stable idle and steady manifold vacuum system for the EFI it has to idle at at least 1500. The narrow LSA was set up to provide a lot of exhaust gas dilution for NOx control, but it really screws up the idle and kills the low end torque. Fifth gear was not useable below 45 MPH, and low end torque was abyssmal.
Because it's a twin cam I was able to reindex the cams by drilling new holes for the index pins. Computer simulations indicated retarding the inlet cam 8 degrees and advancing the exhaust cam by the same amount produced the broadest torque bandwidth. This yielded an inlet POML of 110 degrees and a 114 degree LSA, which, incidently, yields valve timing almost exactly equal to the L-79 cam.
With this valve timing setup it will idle as low as 700, but I have to use 900 to keep the alternator spinning fast enough to carry the load. The 80 percent torque bandwidth is now 2000-6700, which is the broadest I've ever seen, and it will now pull from about 30 MPH (1000 revs) in fifth gear with no problem.
Road engines with manifolds and mufflers don't like a lot of overlap. Even the L-79 has too much. The modern LS7 only has half the effective overlap of the L-79. Most of the aftermarket cams available for vintage Corvette engines were designed back in the sixties and seventies, when high overlap "racing cams" were the rage. They worked well with headers and open exhaust, but are a poor choice for mufflers, even with headers.
Duke
P. S. in the "run info" window there is an "altitude" listing, which you can't change. I believe this is the computed density altitude based on the temperature and pressure at the time of the run. For your runs it was 4500 feet! The best time to do a dyno test is when the outside temperature is in the fifties or sixties, which will usually result in a correction factor near unity if your actual altitude is no more than about 1500 feet. The corrected performance data is most accurate when the CF in near unity.
That's a pretty aggressive curve, which should provide a stronger SOTP pull in the low end of the rev range. I'm assuming your total WOT advance is 37 degrees as you stated before. Given your unusually warm recent temperatures I believe the transient detonation will go away when the weather becomes more seasonable. If necessary you can always substitude a medium spring for the light spring to slow the centrifugal, or back the total down a degree or two, but I would not recommend less than 36 total WOT advance. If there is still some ping at that point, slow down the centrifugal.
It sounds to me like you're very close to optimum, and another session in cool temperatures would be a good idea to get some final numbers on the most aggressive map you can run and also a conservative map for use in very hot weather if the engine gets into detonation.
I reviewed your test files with the WinPEP7 software and have some comments.
1. Is it possible that the RFI shielding on the big block wires were blocking the signal to the mag pickup.(?) This should be solveable for the next session.
2. As is typical the pulls start at about 3000. Since we spend a lot of time below this level, start the pulls at 1000. You want the entire HP/torque curve throughout the useable range. Many dyno operators will resist this. You can test that the engine will accept WOT at 1000 by simply punching it at 1000 in third and fourth gear on the road. The engine should pull smoothly. If not, there's something wrong with the tune. Pulling from 1000 will nail the bottom ends of the 80 and 90 percent torque bandwidths.
3. I could not make any sense out of the A/F data. It was nearly constant across the range, so I don't think it was providing an accurate reading.
4. You can add notes to the test files. When you are looking at the graph, all the files displayed are listed above the graph. Right click on a listed file and you get a menu. Click "Run info" and you'll see a "run notes" text box. You can type in data relating to the configuration, which should correspond to your test plan - details of the baseline config. for the first couple of reference pulls, and then any changes you make for subsequent pulls. This provides a permanent record of the what you did during the session and the performance curves for each config.
If you do another session this fall in cooler weather, my suggestion is to see if the engine will tolerate two light springs in road testing. If so, then do pulls at 36, 38, and 40 degrees WOT advance to find what is optimum. Then you can test this optimum setting with a light-medium and medium-medium spring combinations to determine the effect of slowing the centrifugal.
It's too bad you can't widen the LSA of the cam. My Cosworth Vega has about 220 degrees .050" lifter rise (above the tops of the clearance ramps) duration on both cams, but the IPOML is 102 deg. ATDC, and the LSA is 106 deg. In order to get a stable idle and steady manifold vacuum system for the EFI it has to idle at at least 1500. The narrow LSA was set up to provide a lot of exhaust gas dilution for NOx control, but it really screws up the idle and kills the low end torque. Fifth gear was not useable below 45 MPH, and low end torque was abyssmal.
Because it's a twin cam I was able to reindex the cams by drilling new holes for the index pins. Computer simulations indicated retarding the inlet cam 8 degrees and advancing the exhaust cam by the same amount produced the broadest torque bandwidth. This yielded an inlet POML of 110 degrees and a 114 degree LSA, which, incidently, yields valve timing almost exactly equal to the L-79 cam.
With this valve timing setup it will idle as low as 700, but I have to use 900 to keep the alternator spinning fast enough to carry the load. The 80 percent torque bandwidth is now 2000-6700, which is the broadest I've ever seen, and it will now pull from about 30 MPH (1000 revs) in fifth gear with no problem.
Road engines with manifolds and mufflers don't like a lot of overlap. Even the L-79 has too much. The modern LS7 only has half the effective overlap of the L-79. Most of the aftermarket cams available for vintage Corvette engines were designed back in the sixties and seventies, when high overlap "racing cams" were the rage. They worked well with headers and open exhaust, but are a poor choice for mufflers, even with headers.
Duke
P. S. in the "run info" window there is an "altitude" listing, which you can't change. I believe this is the computed density altitude based on the temperature and pressure at the time of the run. For your runs it was 4500 feet! The best time to do a dyno test is when the outside temperature is in the fifties or sixties, which will usually result in a correction factor near unity if your actual altitude is no more than about 1500 feet. The corrected performance data is most accurate when the CF in near unity.
Comment