[nyt]

Originally Posted by CHECKERED

I would like to opt-out, I’d rather lose a few degrees of timing instead of the engine.

On another note, the flash with revised knock tables should be available on a case by case basis and not the general public (just like beta firmware).

The changes are subtle and still pull timing when needed. I wouldn't worry, but obviously nobody is going to force you to run the map. There's something up with IJE0S at least where it's oversensitive. I was seeing timing pulls running straight E85 at 8psi previously. With these changes I'm even able to hold timing through gear changes.

[Terry @ BMS]

Originally Posted by CHECKERED

I would like to opt-out, I’d rather lose a few degrees of timing instead of the engine.

On another note, the flash with revised knock tables should be available on a case by case basis and not the general public (just like beta firmware).

The settings we're testing now I expect will be OK for the general public when using E85 and the timing curves provided. But as is always the case I just make my best judgement on these things and customers are welcome to second guess that judgement and adjust things as they see fit. Any changes to the knock system posted for general use will definitely be well documented.

[The Leopard]

Originally Posted by Terry @ BMS

The subtle knock response changes have been working well and I expect they will make their way in to all of our E85 back end flash maps in the coming weeks.

Looking at the changes, the 6 knock factor tables have huge changes (50%), which, if it is a multiplicative table, can be many times less sensitive.

I'm curious to see what we find. The N54 is so sturdy, much of which is likely do to this knock algorithm. My previous project was a Subaru which has pistons made of glass.

[Terry @ BMS]

There are other tables you guys haven't seen yet that limit the threshold changes. The master setting in the firmware is in the 3.8 range so it seems large changes are required to produce a minimal effect. We still see plenty of timing drops with the changes made so far. Enough to convince me the system is still very active.

[Deen@StreetKingImports]

Originally Posted by musc

I'm still on flash only ILA0S via Cobb. I am going to wait and see how things unfold using INA0S. Interestingly a lot of timing pull has cleared up with the warmer weather. Previously my Cyl 5 would take a big dive even using E65-70.



I definitely have faith in him but I think even he would agree that these tables should be edited with caution. I would venture to say he will leave these tables stock on the back end flashes and leave it up to the end user to experiment and make the changes on their own.

Ya I hear ya maybe, flash versions with updated knock tables and non updated tables could please both of the audience?

[nyt]

I'm still curious how it will respond on a car running straight 91/93 with higher boost, in a condition that should pull timing, but I don't want to be that guinea pig

[lcvette]

Terry, just a guess, but could "pre" mean a high probability condition occurring? Eg.. Ect, iat, load, rpm showing optimal knock conditions and the table is setup to pre-emptively avoid knock. When tuning other platforms avoiding knock was always better than having to remove it. Attack rates were large... 5-6 degrees and required slow decay rates to recover. Avoiding true knock was obviously the best case scenario (to save the motor) but also because of how aggressive the protection was and how bad it hurt power.

I agree with you, with proper octane and the cooling effects of alcohol based fuels these tables are overly sensitive. When I was learning the knock module on DFI, I set the attack rates very high when building tunes. I also found that ethanol had a very broad (comparatively) safe Afr. We found we made best power with a richer mixture 11.0-11.3 and higher timing and had safer egt's and lower ect's. However it also did fine at 12.9 and lower timing but had higher ect's and egt's. I was shocked when I saw how lean these motors are tuned but I am so new to DI I don't know what to think about it. However with the introduction of port injection, i do wonder if these leaner afr's are still a great idea given you now have fuel in the cylinder heating up and available for knock earlier than the DI fuel. Just thinking out loud... Like I said I'm so green to DI, it's hard to tell up from down yet... Lol

Chris

[The Leopard]

Originally Posted by Terry @ BMS

There are other tables you guys haven't seen yet that limit the threshold changes. The master setting in the firmware is in the 3.8 range so

Please complete your thought T. I don't want to speculate what you imply the 3.8 to mean.

Thanks

[nyt]

Originally Posted by lcvette

Terry, just a guess, but could "pre" mean a high probability condition occurring? Eg.. Ect, iat, load, rpm showing optimal knock conditions and the table is setup to pre-emptively avoid knock. When tuning other platforms avoiding knock was always better than having to remove it. Attack rates were large... 5-6 degrees and required slow decay rates to recover. Avoiding true knock was obviously the best case scenario (to save the motor) but also because of how aggressive the protection was and how bad it hurt power.

I agree with you, with proper octane and the cooling effects of alcohol based fuels these tables are overly sensitive. When I was learning the knock module on DFI, I set the attack rates very high when building tunes. I also found that ethanol had a very broad (comparatively) safe Afr. We found we made best power with a richer mixture 11.0-11.3 and higher timing and had safer egt's and lower ect's. However it also did fine at 12.9 and lower timing but had higher ect's and egt's. I was shocked when I saw how lean these motors are tuned but I am so new to DI I don't know what to think about it. However with the introduction of port injection, i do wonder if these leaner afr's are still a great idea given you now have fuel in the cylinder heating up and available for knock earlier than the DI fuel. Just thinking out loud... Like I said I'm so green to DI, it's hard to tell up from down yet... Lol

Chris

I'd think "pre-window" would be referring to a time period

[jyamona]

Originally Posted by nyt

I'd think "pre-window" would be referring to a time period

It is most definitely referring the the time "window" where it is looking for knock. There are beginning and ending window tables for each cylinder. I don't remember off hand, but I think the values were "degrees crank" (*crk), which would make sense.

Aren't there only two physical knock sensors? Then it would apply the time window to determine each cylinder.

[The Leopard]

Yes, there are exactly two physical knock sensors.

[Terry @ BMS]

Originally Posted by jyamona

It is most definitely referring the the time "window" where it is looking for knock. There are beginning and ending window tables for each cylinder. I don't remember off hand, but I think the values were "degrees crank" (*crk), which would make sense.

Aren't there only two physical knock sensors? Then it would apply the time window to determine each cylinder.

Right but then what does prewindow mean?

[PR3CI5N335i]

When will these changes be applied to the BMS BEF's? My car is AT IJEOS with E60 and 100% Meth and still have bad post shift timing drops. I'll be running stage 2 later this month and it would be nice if these timing drops were sorted out by then

Edit: just saw that Terry said within the next few weeks. Can't wait!

[Terry @ BMS]

Just email me if you want to test out the new maps.

[jyamona]

Originally Posted by Terry @ BMS

Right but then what does prewindow mean?

Not sure....trying to think what it could be for.

[knissel]

Originally Posted by jyamona

Not sure....trying to think what it could be for.

Maybe it means pre-ignition?

[jyamona]

Originally Posted by knissel

Maybe it means pre-ignition?

That is possible. I know it states "pre-window", but it could be ignition window, and not the knock sensor window. I will look at the nomenclature for the rest of the related tables.

[knissel]

I don't know much about this stuff but I thought this was a good diagram to explain how knock works.

[3000GT MR]

Originally Posted by jyamona

That is possible. I know it states "pre-window", but it could be ignition window, and not the knock sensor window. I will look at the nomenclature for the rest of the related tables.

How does historic knock, octane, and lambda's play a roll in all this? When clearing adaptations what exactly are you clearing? How and why do they make a difference? Any type of window sounds like a specific area that the DME is looking for to keep track of outside the action-reaction that most of the tables are doing. IDK just a though

[Terry @ BMS]

Originally Posted by jyamona

That is possible. I know it states "pre-window", but it could be ignition window, and not the knock sensor window. I will look at the nomenclature for the rest of the related tables.

It might be to allow you to set the system to be more or less sensitive when each cylinder is first entering its measurement range. Like say the first 5%. Shrug.

[PR3CI5N335i]

Loaded up a new MHD E85 IJEOS file last night from Terry and my timing drops are finally gone. I've had these issues for over a year now and I feel like my car has been running at 85% of its power potential due to these timing drops but not anymore. Thanks for the hard work guys!

Map 3 75add, E60, 100% Meth, AT, stock turbo's (for now)

[Bullitt1841]

Originally Posted by knissel

I don't know much about this stuff but I thought this was a good diagram to explain how knock works.

Wow, great find!

Terry, why don't you strap one of these on your test car to verify your DME changes. Dinan and Kenne Bell supposedly have

http://www.jandssafeguard.com/monitors.html

[Terry @ BMS]

I used to run them in the 90s. Our DME does a much better job of it lol.

[Bullitt1841]

No way! Counting LED's is high tech!

[Spxxx]

Lol funny article from 1991 they quote.