Kevin Doe
April 17th, 2008, 11:00 PM
I've been having a discussion over PMs the last day or so, and I figured I'd post it for everyone to see....
In the warmer weather is seems as if the car is bucking more at light throttle and also with no throttle. I've read your idle tuning thread several times and really tried to wrap my head around what is going on, and how all those things tie together. It would be really great if I could see the logic diagrams, but thats a pipe dream.
I see you talk about increasing the TF multiplier and and also increasing the decay rate. I hate to arbitrarily be chaning the table w/o some data to drive the change. That leads to me think that logging data while I'm getting bucking will tell me what I need to know. Things like how much to increase the mulitplier, how much to increase the decay rate, and where to change it.
My real question here is what are the right parameters to log for this. Do you have a fairly comprehensive list of parameters that I could log that will tell me the whole story. Also, can you explain how i look at the data and what I'm really looking for while its bucking. I assume I look in the region at the onset of bucking and go from there.
I get bucking both in the light throttle and no throttle regions. Usually only able to get out of it by giving it a lot of throttle for a minute, or pushing the clutch in. I have an MS4 cam for reference, so I'm sure thats dirving a lot of the bucking (not even sure if I'll be able to remove all the bucking). I also have poly motor/trans mounts, and poly diff mounts. My car is a RX-7 with an LS1/T56 and a Ford Cobra independent diff. So everything is mounted pretty rigidly and there is no forgiveness to any bucking.
Any help is appreciated.
Thanks,
Kevin
In my experience, bucking at low RPMs comes from timing jumping around, incorrect timing, and incorrect IAC airflow. You can see the effects of the TC/TF by logging their pids along with IACDES_B. Ideally, you have the base desired airflow values that reference temp. Then, for off throttle, non-idle control, the throttle cracker is responsible for getting the airflow right. Throttle follower builds off the cracker when in non-idle and throttle is applied. Idle and non-idle is determined by B0107 and B0108.
Same goes for timing. If you're not in idle mode, the spark adjusters are suspended. Therefore, you're relying on your high/low and base spark tables. The table you're in (base or high/low) is determined by B5916 and B5917. Personally, the approach I took was to say - look at the base spark table for all off-throttle timing regardless of speed. That meant setting B5916 & B5917 to 1.19% and 255mph. Then, I could set decel and idle timing in the base tables and adjust the on-throttle timing in the high/low tables. Remember, you want to minimize the timing from jumping around too much when off the throttle. Therefore, getting it to idle smoothly with as little timing as possible is helpful. Then, between idle and ~2000rpm, increase it very slowly. You shouldn't need more than 35* of timing anywhere above 2K in the base table. For the on-throttle (high/low tables), try setting the timing lower than you think would work for the low rpm, low airflow cells. Better yet, experiment in an empty parking lot with the bi-di controls. Try something crazy like 15 degrees. Trial and error is about the only way you're going to find what the car likes - there no magic formula.
Hope this helps some...
Jeff
Thanks for taking the time to reply. After reading your reply and looking at my tune and various logs I think I have my timing under control. I'm thinking the issue is from the airflow being off.
I have done the RAFIG process recently up a cold startup to operating temp and updated the values in the desired airflow table. They look nice and smooth. During the bucking I see the timing movin around maybe 3 degrees during the swings, but that looks to be driven due to the rpms moving aruond. I think it seems like an effect, not the cause.
Now a few questions. When I log IACDES_B, should that be the value in the desired airflow table at the current operating temp? If not, how is that calculated?
So in the non-idle region the TC/TF are responsible for getting airflow right. From what I understand, I would want the sum of TC and TF airflow to equal the desired IAC airflow (IACDES_B).
Forgive me if these seem like dumb questions.
IACDES_B will show the final value of airflow desired from the IAC (like the name says). So, base (desired) airflow, learned airflow trims (LTIT/STIT), TC, TF, etc...
A 3 degree swing in timing is more than just a side effect. It's contributing to the problem. I noticed some light surging even when timing changed from 23.5 to 24 degrees on my car. The problem was, that little 1/2 degree change created enough of a surge to cause the next rpm jump to create a 1 degree of timing swing...which then amplified higher. It's like ripples in a pond. Once the process starts, it can grow rather quickly if you have steep changes in your timing maps. That cam should take any more than 25 degrees to idle in an M6. I'd be willing to bet you could get it done with ~23 degrees assuming the desired airflow is set up for that timing value and the fueling isn't too rich.
If you don't mind taking a look at my tune I'd appreciate it. It doesn't look like I can attach a .tun file here on a PM, so I'll make a thread out of this. I'm sure there are some others that may be interested in this information as well. I guess it can't hurt. I'll also attach a log. Its a log from a track day I did at a road course. As I was pulling off the track I was getting some throttle on surging. I didn't have the important stuff logged, but I do have timing and DYNCYLAIR_DMA, and various other parameters. The surging occurs from frame 10130 to 10170. There also appears to be some light surging from frame 65 to frame 110. I'll try and get out this weekend and log some more important PIDs during the surging.
The log file is from the april 7th BBL update, so I'm not sure you can see it if you don't have the beta software or not. Well here they are.
Also, if you notice anything else that looks odd with my tune please don't feel shy to let me know.
Thanks again.
In the warmer weather is seems as if the car is bucking more at light throttle and also with no throttle. I've read your idle tuning thread several times and really tried to wrap my head around what is going on, and how all those things tie together. It would be really great if I could see the logic diagrams, but thats a pipe dream.
I see you talk about increasing the TF multiplier and and also increasing the decay rate. I hate to arbitrarily be chaning the table w/o some data to drive the change. That leads to me think that logging data while I'm getting bucking will tell me what I need to know. Things like how much to increase the mulitplier, how much to increase the decay rate, and where to change it.
My real question here is what are the right parameters to log for this. Do you have a fairly comprehensive list of parameters that I could log that will tell me the whole story. Also, can you explain how i look at the data and what I'm really looking for while its bucking. I assume I look in the region at the onset of bucking and go from there.
I get bucking both in the light throttle and no throttle regions. Usually only able to get out of it by giving it a lot of throttle for a minute, or pushing the clutch in. I have an MS4 cam for reference, so I'm sure thats dirving a lot of the bucking (not even sure if I'll be able to remove all the bucking). I also have poly motor/trans mounts, and poly diff mounts. My car is a RX-7 with an LS1/T56 and a Ford Cobra independent diff. So everything is mounted pretty rigidly and there is no forgiveness to any bucking.
Any help is appreciated.
Thanks,
Kevin
In my experience, bucking at low RPMs comes from timing jumping around, incorrect timing, and incorrect IAC airflow. You can see the effects of the TC/TF by logging their pids along with IACDES_B. Ideally, you have the base desired airflow values that reference temp. Then, for off throttle, non-idle control, the throttle cracker is responsible for getting the airflow right. Throttle follower builds off the cracker when in non-idle and throttle is applied. Idle and non-idle is determined by B0107 and B0108.
Same goes for timing. If you're not in idle mode, the spark adjusters are suspended. Therefore, you're relying on your high/low and base spark tables. The table you're in (base or high/low) is determined by B5916 and B5917. Personally, the approach I took was to say - look at the base spark table for all off-throttle timing regardless of speed. That meant setting B5916 & B5917 to 1.19% and 255mph. Then, I could set decel and idle timing in the base tables and adjust the on-throttle timing in the high/low tables. Remember, you want to minimize the timing from jumping around too much when off the throttle. Therefore, getting it to idle smoothly with as little timing as possible is helpful. Then, between idle and ~2000rpm, increase it very slowly. You shouldn't need more than 35* of timing anywhere above 2K in the base table. For the on-throttle (high/low tables), try setting the timing lower than you think would work for the low rpm, low airflow cells. Better yet, experiment in an empty parking lot with the bi-di controls. Try something crazy like 15 degrees. Trial and error is about the only way you're going to find what the car likes - there no magic formula.
Hope this helps some...
Jeff
Thanks for taking the time to reply. After reading your reply and looking at my tune and various logs I think I have my timing under control. I'm thinking the issue is from the airflow being off.
I have done the RAFIG process recently up a cold startup to operating temp and updated the values in the desired airflow table. They look nice and smooth. During the bucking I see the timing movin around maybe 3 degrees during the swings, but that looks to be driven due to the rpms moving aruond. I think it seems like an effect, not the cause.
Now a few questions. When I log IACDES_B, should that be the value in the desired airflow table at the current operating temp? If not, how is that calculated?
So in the non-idle region the TC/TF are responsible for getting airflow right. From what I understand, I would want the sum of TC and TF airflow to equal the desired IAC airflow (IACDES_B).
Forgive me if these seem like dumb questions.
IACDES_B will show the final value of airflow desired from the IAC (like the name says). So, base (desired) airflow, learned airflow trims (LTIT/STIT), TC, TF, etc...
A 3 degree swing in timing is more than just a side effect. It's contributing to the problem. I noticed some light surging even when timing changed from 23.5 to 24 degrees on my car. The problem was, that little 1/2 degree change created enough of a surge to cause the next rpm jump to create a 1 degree of timing swing...which then amplified higher. It's like ripples in a pond. Once the process starts, it can grow rather quickly if you have steep changes in your timing maps. That cam should take any more than 25 degrees to idle in an M6. I'd be willing to bet you could get it done with ~23 degrees assuming the desired airflow is set up for that timing value and the fueling isn't too rich.
If you don't mind taking a look at my tune I'd appreciate it. It doesn't look like I can attach a .tun file here on a PM, so I'll make a thread out of this. I'm sure there are some others that may be interested in this information as well. I guess it can't hurt. I'll also attach a log. Its a log from a track day I did at a road course. As I was pulling off the track I was getting some throttle on surging. I didn't have the important stuff logged, but I do have timing and DYNCYLAIR_DMA, and various other parameters. The surging occurs from frame 10130 to 10170. There also appears to be some light surging from frame 65 to frame 110. I'll try and get out this weekend and log some more important PIDs during the surging.
The log file is from the april 7th BBL update, so I'm not sure you can see it if you don't have the beta software or not. Well here they are.
Also, if you notice anything else that looks odd with my tune please don't feel shy to let me know.
Thanks again.