When you click on links to various merchants on this site and make a purchase, this can result in this site earning a commission. Affiliate programs and affiliations include, but are not limited to, the eBay Partner Network.
Hi Jim! Welcome to the party, and thanks for posting up that data from way back in 2013. It is still very relevant. It's reassuring to see both GTSs have similar issues with under-fueling and knock in the similar RPM/Load scenarios. And we both had similar outcomes if less knock and less retard by adding more fuel.
I am curious about the interplay of the flappy since the RPM range of many of the knocks (for both cars) is around the flappy activation zone. I wonder if the flappy changes the airflow in the runners enough to shift the AFR needs in each cylinder.
What are your thoughts on further knock mitigation actions for my particular car? Is it good enough; or try to minimize more? If so please share your thoughts on next steps
The flappy does change airflow, but more importantly the VE of the engine at that point.
It wants to knock more, because the cylinder pressures are higher.
We talked about this months ago..but I never heard back to help ya out.
What I do in my fuel maps, is I increase the requested A/F in the moderate load areas around highway RPMs to help tip-in on cruise control to give the engine a good torque friendly burn to not fall so far behind the power/speed curve on the hills. Depending on VR, peak torque will be between 12.5 and 13.7..I have found. Richer for higher RPMs and leaner for lower RPMs.
It's not a flat left-to-right A/F target across the cells. Having free casual access to an inductive load dyno for a # of hours every month does help. Set the CC at 70 with a super light dyno load, then bump up the load (a hill) and see how the engine reacts. Then pick city speeds, rinse wash repeat.
Timing...if I get the fuel where I want it, I let Ken's chips find the highest knock-limited cylinder pressure it can find for me.
For my engine in my condition (new heads, new cams with slight magic, etc)..I've gotten good results. The car around town and on the highway feels a little lighter, and the CC doesnt fight as hard on hills.
Ken's chips saved the hassle of finding "good to great" timing values.
So..for you, get the fuel right, then chase timing.
For snappiness around town, the places I find where tip-in throttle happens for me, I look for ~13.3-5 for best low RPM torque.
Maybe it is the same patch Ken wrote for me which was very helpful when I was trying to assess the impact of timing cylinder to cylinder as it were. As a test vehicle it was excellent but it is not a long term solution in that it advances or retards every cell for the given cylinder being tested. I found that by retarding 2 and 6 a couple of degrees I could advance all the others more across the board. - doubtless Ken will fill you in on this.
The patch should allow the map to be advanced generally and in detail without regard to individual cylinder issues. The whole map can be advanced by using the last cell in the warmup table. This general advance will be removed later as the overall advance level and per-cylinder retard at different rpms will be eventually handled by the EZK-S chip after the custom map is baked into it.
The patch should allow the map to be advanced generally and in detail without regard to individual cylinder issues. The whole map can be advanced by using the last cell in the warmup table. This general advance will be removed later as the overall advance level and per-cylinder retard at different rpms will be eventually handled by the EZK-S chip after the custom map is baked into it.
I asked in the other thread, what chip do you use? I have a BOX of simlar chips that Ive "collected" from illegal BMWs at the racetrack, and theyre not the same..what model do you use?
I have a handful to send back like you asked a while back
The flappy does change the airflow, but more importantly, the VE of the engine at that point. It wants to knock more because the cylinder pressures are higher. We talked about this months ago...but I never heard back to help you out. In my fuel maps, I increase the requested A/F in the moderate load areas around highway RPMs to help tip-in on cruise control to give the engine a good torque-friendly burn to not fall so far behind the power/speed curve on the hills. Depending on VR, peak torque will be between 12.5 and 13.7, I have found. Richer for higher RPMs and leaner for lower RPMs.
It's not a flat left-to-right A/F target across the cells. Having free casual access to an inductive load dyno for a # of hours every month does help. Set the CC at 70 with a super light dyno load, then bump up the load (a hill) and see how the engine reacts. Then pick city speeds, rinse, wash repeat.
Timing... if I get the fuel where I want it, I let Ken's chips find the highest knock-limited cylinder pressure it can find for me. For my engine in my condition (new heads, new cams with slight magic, etc), I've gotten good results. The car around town and on the highway feels a little lighter, and the CC doesn't fight as hard on hills. Ken's chips saved the hassle of finding "good to great" timing values. So..for you, get the fuel right, then chase timing. For snappiness around town, the places I find where tip-in throttle happens for me, I look for ~13.3-5 for best low RPM torque.
Hi Jeff, thanks for the response. Yes, I remember talking with you early on in my ST learning curve, and most of what you said was over my head at the time. I just forgot to follow up with you. My apologies. But thanks for the valuable input here. I am meeting up with Ken tonight to discuss possible changes to the base ignition map before adding the EZKs timing on top of it.
Question for you @Speedtoys , when you have the PEM chip in the LH and Ken's EZKs chip in the EZK, are you able to datalog knocks and Cyl retard?
Hi Jeff, thanks for the response. Yes, I remember talking with you early on in my ST learning curve, and most of what you said was over my head at the time. I just forgot to follow up with you. My apologies. But thanks for the valuable input here. I am meeting up with Ken tonight to discuss possible changes to the base ignition map before adding the EZKs timing on top of it.
Question for you @Speedtoys , when you have the PEM chip in the LH and Ken's EZKs chip in the EZK, are you able to datalog knocks and Cyl retard?
Nope. I let Ken's logic manage it...because it's not in my control if there are any.
...
In reference to my original post, this is where some benchmark data would be helpful. Is the number of these knocks at high load still within the totally normal zone? This seems comparable with what I would see in my S4/5spd.
It will be interesting to see if we can work on further reducing these remaining knocks through ignition manipulation.
My personal opinion is that what you are seeing is totally within the normal zone. You can fine-tune it, but in my experience that is just chasing ghosts. There are other things that affect ignition timing to a much greater degree-- temperatures (engine and ambient), altitude, fuel-- that chasing a degree or two in a real-life environment only leads to insanity. Nothing wrong with that, of course.
Today I got a terrific how to session from Jim C. and Michael B. It was a bit like drinking from a fire hose.
1993 928 GTS MT 22k miles
30# Ford injectors
S3 Cams
Stock Cat
Hard to put in words the difference. But, off idle stutter/jerking is gone. Tip in is perfect. Night and day difference.
I would like to make some more changes. But, this is very nice for now.
Before:
How on earth did you get into a postion where you were trying to run like that before the ST2 session?
The only thing I can think of would be if you told ST2 you were running injectors of greater capacity than those actually installed. I am surprised the poor thing would run at all yet alone run "poorly".
How on earth did you get into a postion where you were trying to run like that before the ST2 session?
The only thing I can think of would be if you told ST2 you were running injectors of greater capacity than those actually installed. I am surprised the poor thing would run at all yet alone run "poorly".
I have not driven it at since I put the PEMs in there - I did set the injector size and latency. But, that was it.
I was in a holding pattern until I got someone to help me understand the ST. That came in the form of Michael.
He and Jim C. guided me through the proper setup for my GTS. It started with increasing the latency. Jim was a great help in understanding the ST. Michael was instrumental in educating me on the SharkPlotter.
How on earth did you get into a postion where you were trying to run like that before the ST2 session?
The only thing I can think of would be if you told ST2 you were running injectors of greater capacity than those actually installed. I am surprised the poor thing would run at all yet alone run "poorly".
Fred, that plot was done closed-loop (factory O2-sensor enabled), O2-adjust was a solid +20%. The actual AFR (with the 20% adjustment) was around 15:1, but when running closed-loop SP plots each point as what the AFR would be if it were running open-loop-- without the adjustment-- which would have been around 18:1. That approach is appropriate for adjusting the map, which really was around 25% too lean.
Normally, to optimize the map, you would log data with the O2-sensor (NBO2) disabled (disconnect it or set Sharktuner to force-disable the O2-sensor). In that case it would have run like crap (if at all), until the map were adjusted. Sharkplotter is happy either way, and uses the O2-adjust values (if enabled) as part of the suggested map adjustments.
The root problem here was the opening-time (latency) setting for the 30# injectors, Ford Motorsports M9593-B302. Ford specs a very quick 0.71ms opening time (latency) for our fuel pressure, at least that's what I came up with from their nearly incomprehensible spec sheet. It turns out that number is nonsense, reality is more like 1.0ms which I think is where we wound up. (Stock injectors are 0.94 ms). The B302 injectors are the fat metal-body versions with a reddish top, our GT runs the BB302 which are skinny bodies with the same reddish color, Opening time for those is spec'ed at 1.20ms which is correct per my testing. This has a big effect at idle and light load where fuel pulses are short, much less so under heavy load.
@Kevin in Atlanta , please post your SP file once you have some high load and high rpm runs and optimize the fuel maps for those loads. It will be useful to see the knock data from your GTS with a fresh rebuild, GT-ish cams, and stock exhaust.
Fred, that plot was done closed-loop (factory O2-sensor enabled), O2-adjust was a solid +20%. The actual AFR (with the 20% adjustment) was around 15:1, but when running closed-loop SP plots each point as what the AFR would be if it were running open-loop-- without the adjustment-- which would have been around 18:1. That approach is appropriate for adjusting the map, which really was around 25% too lean.
Normally, to optimize the map, you would log data with the O2-sensor (NBO2) disabled (disconnect it or set Sharktuner to force-disable the O2-sensor). In that case it would have run like crap (if at all), until the map were adjusted. Sharkplotter is happy either way, and uses the O2-adjust values (if enabled) as part of the suggested map adjustments.
The root problem here was the opening-time (latency) setting for the 30# injectors, Ford Motorsports M9593-B302. Ford specs a very quick 0.71ms opening time (latency) for our fuel pressure, at least that's what I came up with from their nearly incomprehensible spec sheet. It turns out that number is nonsense, reality is more like 1.0ms which I think is where we wound up. (Stock injectors are 0.94 ms). The B302 injectors are the fat metal-body versions with a reddish top, our GT runs the BB302 which are skinny bodies with the same reddish color, Opening time for those is spec'ed at 1.20ms which is correct per my testing. This has a big effect at idle and light load where fuel pulses are short, much less so under heavy load.
Jim,
That makes sense. Finding latency times for injectors is nigh on impossible or so I found when I purchased my Design 2 [fat body] 4 hole 30lb injectors. I simply installed mine, told ST2 I now had 30 lb injectors and data logged across the range. My AFR numbers looked remarkably similar to that of the stock injectors which as you say is around 1.0 ms so I just ;eft the latency as per stock items. If the latency is out one would expect to see AFR's consistently overfuelled or underfuelled across the board or so I concluded. Of course I never had to dick around with NBO2 because I do not have one on a non cat motor.
@Kevin in Atlanta , please post your SP file once you have some high load and high rpm runs and optimize the fuel maps for those loads. It will be useful to see the knock data from your GTS with a fresh rebuild, GT-ish cams, and stock exhaust.
10-05-2023 | 03:01 PM
