Timing issues? 16197427 93 TBI

[Fast355]

OKay Fast 355.

I am better understanding the spark timing scheme now and your comment about raising the tables in that area are jiving now.

If I am wrong here, please correct me.

The main spark Bias is always subtracted in the timing calculation. I was not understanding that initially. I was thinking it is only applied under certain circumstances. In fact, it is always used and the effect shifts depending on the spark table it is looking at. This way it is not required to have many spark bias' set up to handle the varying conditions.

Anyway, question now is, WHEN does the Altitude Spark Bias come into play? I put the Main Spark Bias back into the program without negative effect but I have not put the Altitude Spark bias in yet. Is there a specific table it is associated with in $OD?

I am glad I pulled the Bias' out because it quantified a problem and even gladder I didn't melt anything. Maybe not the best plan I ever had.

Thanks,
Jeremy

A bias value is used so that you can have a negative value in the table. I would have to dig up the whole hac but I am certain from memory there is a baro correction of the spark advance in the calibration. That would be where the altitude spark bias comes up. The higher in altitude you get the lower your baro kpa. Also keep in mind that reported spark advance from the datastream will be innacurate if the initial or base timing is any value other than zero. On my definition I had to alter the equation to add the 10° initial back into the calculation. Otherwise it reported the timing 10° lower than actual. Add the altitude bias back in and ignore changing any bias values. They are all used, otherwise they would not have a valie other than zero. When I added IAT compensation in, I had to add a 10° bias to allow the IAT table to retard the timing in hot weather.

[TBI Tinker]

A bias value is used so that you can have a negative value in the table. I would have to dig up the whole hac but I am certain from memory there is a baro correction of the spark advance in the calibration. That would be where the altitude spark bias comes up. The higher in altitude you get the lower your baro kpa. Also keep in mind that reported spark advance from the datastream will be innacurate if the initial or base timing is any value other than zero. On my definition I had to alter the equation to add the 10° initial back into the calculation. Otherwise it reported the timing 10° lower than actual. Add the altitude bias back in and ignore changing any bias values. They are all used, otherwise they would not have a valie other than zero. When I added IAT compensation in, I had to add a 10° bias to allow the IAT table to retard the timing in hot weather.

Thank you for the reply Fast355

I recognize you are a VAN guy and this is a half-VAN [pick-up]. JI recall the times my friend and I frequented the street races in Dallas with his Dodge conversion van. We didn't win EVERY race but it was a great feeling when he would smoke a Camaro in the big orange early 70's 'hippy mobile' with the Sierra Nevada mural on the side. Surprised a lot of people for sure. I can appreciate why you enjoy hot rodding your vans.

Copy that. I understand a 'bias' value is substracted from the Main Spark Table and allows for negative timing. I am trying to get my head wrapped around when the engine transitions from one spark table to the next and how the program prioritizes which table to read from. Unfortunately, my brain doesn't comprehend hex well enough follow the path of operations in the background. I still have a lot to learn it so I appreciate the input.

I am dialed in at the distributor [base] 10 degrees and have set the Initial Timing for 10 degrees in the XDF. I have also changed the ADX to accommodate the 10degree initial timing. I did notice the data stream appears to be reading total applied timing rather than commanded timing. Does the Main Spark Table then show the total Commanded timing prior to any bias being applied?? I think this is correct?

Anyhow. I can add/take away the altitude bias and it will do pretty good when it's removed but dies on tip-in when I put it back in. I feel like it is just masking the issue I need to adjust as you have eluded to as well.

Fuel. It seems like a fuel issue. Still. Ie. TPS increases suddenly, Map [kPa] increases and the timing nose-dives below zero . If it had fuel, the MAP wouldn't transition [ramp rate] as quickly allowing the timing to keep-up by not calling for the BIAS to prevent knock or whatever. Am I off my rocker here?

That being said, I went in and added 5 degrees to all the areas shown as 'negative' with no real performance gains with the Altitude bias @ 9.8. I also increased AE vs MAP by 60%. No real gains while Alt Bias was @ 9.8 but it did run a bit better on tip-in when the bias was zeroed.

I found the Altitude Spark Advance Correction vs BARO vs Vacuum I think you are referring to. Also Baro Correction Factor vs RPM vs TPS. This is where I start to get befuddled.

First, there is box to choose BARO and MAP being used together or NOT. Currently this box is NOT checked. Does it need to be? Seems like it should be but it is listed as a transmission parameter.

Second, do the two tables listed above add or subtract timing? Looks like the Alt Spk Adv Corr vs BARO vs Vacuum subtracts and the other adds?

Third, I am confused about this equation: .351563 * (X-Y). If X is Main Spark Table [commanded] timing and Y is the Bias how does this make negative timing?

Seems like it would be (.351563 * X)-Y ??? I must be missing something.

As I continue to research, I appreciate any help as always.

Jeremy

[TBI Tinker]

Update

I found the issue with the timing. Burst knock retard. Apparently it anticipates knock even when there is none. Similar effect to the bias values I think. Looks like a proper VE table / fueling scheme to keep things semi-rich at low rpm/high MAP would do away with the need for it possibly. I only have a NB02 but it appears I am plenty rich currently so no worries at the moment. [After screwing with everything having to do with AE] Staying around .900 under acceleration. Ordering a WB02 so I can adjust the fuel trims in open loop. I was looking at the 30-4110 from AEM. Is there any real reason to move up to the 30-3110? Curious if there are better options this day and age.

For $0D Burst Knock Retard is enabled at 400F-Bit 3. This is NOT the same as Knock Retard. I was confused due to the 'Burst Knock Retard' messing with timing but I was not seeing any 'Knock Retard'. Burst knock retard is a calculated value. Knock retard is a reaction to audible knock translated to the computer by the knock sensor. I disabled it.

On the log trend I was seeing the timing tapering off as soon as the MAP dropped and the TPS increased. Very near simultaneously. I didn't know Burst knock was even a thing until I looked at the .csv file. The .csv file is my friend now. Makes things SO much easier and gives plenty of opportunity to crunch numbers for averaging etc. LOVE IT.

Now that the truck is running well enough to drive I took it for a spin finally. Runs okay. Obviously needs tuning but not too bad. Bog is gone, no lean pop but I did find out the 3-4 clutch is smoked though. I was planning to rebuild it anyway. Was kind of hoping to hold off for a bit but it doesn't surprise me. The rest of the truck was pretty ragged out also. Overall I think it will be a fun ride once I'm done.

1Project2many - You mentioned previously using the torque curve from a Dyno Simulator would simplify shaping the VE tables. I did procure a DTD copy and have the torque curve now. Can you advise on how I need to 'shape' or rough-out the VE table? The simulator gives a VE curve and Torque curve at WOT. I used those numbers and applied them to the 100 and 90 MAP column. Is there anything else to do there or do I just use the curve as a reference with the WB02 to get things dialed in?

Oh. I also agree that .351563 * (X-Y). Not that it matters that I agree with it. But I get it now. A little slow on the uptake for that one.

Hope all is well,

Jeremy

[Fast355]

Update

I found the issue with the timing. Burst knock retard. Apparently it anticipates knock even when there is none. Similar effect to the bias values I think. Looks like a proper VE table / fueling scheme to keep things semi-rich at low rpm/high MAP would do away with the need for it possibly. I only have a NB02 but it appears I am plenty rich currently so no worries at the moment. [After screwing with everything having to do with AE] Staying around .900 under acceleration. Ordering a WB02 so I can adjust the fuel trims in open loop. I was looking at the 30-4110 from AEM. Is there any real reason to move up to the 30-3110? Curious if there are better options this day and age.

For $0D Burst Knock Retard is enabled at 400F-Bit 3. This is NOT the same as Knock Retard. I was confused due to the 'Burst Knock Retard' messing with timing but I was not seeing any 'Knock Retard'. Burst knock retard is a calculated value. Knock retard is a reaction to audible knock translated to the computer by the knock sensor. I disabled it.

On the log trend I was seeing the timing tapering off as soon as the MAP dropped and the TPS increased. Very near simultaneously. I didn't know Burst knock was even a thing until I looked at the .csv file. The .csv file is my friend now. Makes things SO much easier and gives plenty of opportunity to crunch numbers for averaging etc. LOVE IT.

Now that the truck is running well enough to drive I took it for a spin finally. Runs okay. Obviously needs tuning but not too bad. Bog is gone, no lean pop but I did find out the 3-4 clutch is smoked though. I was planning to rebuild it anyway. Was kind of hoping to hold off for a bit but it doesn't surprise me. The rest of the truck was pretty ragged out also. Overall I think it will be a fun ride once I'm done.

1Project2many - You mentioned previously using the torque curve from a Dyno Simulator would simplify shaping the VE tables. I did procure a DTD copy and have the torque curve now. Can you advise on how I need to 'shape' or rough-out the VE table? The simulator gives a VE curve and Torque curve at WOT. I used those numbers and applied them to the 100 and 90 MAP column. Is there anything else to do there or do I just use the curve as a reference with the WB02 to get things dialed in?

Oh. I also agree that .351563 * (X-Y). Not that it matters that I agree with it. But I get it now. A little slow on the uptake for that one.

Hope all is well,

Jeremy

Burst knock is helpful when you are running a fairly aggressive sprak timing map. It will pull some timing on tip-in to prevent the knocking and real knock retard that would otherwise happen. From memory the amount of burst retard as well as the duration of it are both adjustable as well as the load change required to activate it.

[TBI Tinker]

Burst knock is helpful when you are running a fairly aggressive sprak timing map. It will pull some timing on tip-in to prevent the knocking and real knock retard that would otherwise happen. From memory the amount of burst retard as well as the duration of it are both adjustable as well as the load change required to activate it.

Copy that. Once I have some fuel tuning completed I may turn it back on to tinker with it. I suspect when Summer rears it's ugly head I may be looking for ways to calm some actual spark knock. Good to know you have used it to tame spark knock successfully. I wasn't 100% certain but had it in my mind it might be an option. For now, I plan on leaving turned off to simplify the tuning. I'm still just a baby when it comes to tuning with EFI so fighting anomalous spark retard has been frustrating. It was a good learning experience though. The more I come across things like that the bigger and more clear the tuning requirements become. I like how Burst knock can be turned on and off at will with the option to tweak multiple tables for it. Seems like it could lend a lot of flexibility without tinkering with the bias values too much.

This was my first big hurdle. More to come I'm sure. Feeling a bit more confident for now.

Thanks to all for the advice. I'll take all I can get,

Jeremy