MITYAZ GSR Build Thread

[mityaz]

Ok, since i need to resolve boost levels ABOVE 30psi and since the last lonely column in the 3xMAP tables ends at 29 psi, i wanted to try this idea. Please tell me if you see something inherently wrong with this thinking.

I removed a 9psi column and added a new one at 37.5psi. My 3D graph is still very linear bu I now have a "defined" zone in the high PSI range. Why would/wouldn't this work? Have anyone tried it? @razorlab @UT_EvoX

The top table is before and the bottom is after. I realize the values are crazy high, but who gives a freak, it's just a number. As long as I can define that load level with proper fueling/timing value i should be good (i think).

[UT_EvoX]

I might come off as rude but you need professional tuning. This is 101 stuff

Sent from my SM-N950U using Tapatalk

[UT_EvoX]

This is what you HAVE to do lol. The ECU does not interpolate beyond the rightmost column. Once you hit that, you're done.

Sent from my SM-N950U using Tapatalk

[mityaz]

I might come off as rude but you need professional tuning. This is 101 stuff

Sent from my SM-N950U using Tapatalk

Well, i dont take things personal. I am, however, pretty excited that i came to it myself! Thank you for confirming.

That's why i am on this forum guys. We all have to start somewhere.

I was tuned professionally by Chet once haha. None of these tables were changed like that even tho i was running 33+ psi.

I would imagine that this would be common knowledge by now.

Sent from my SM-G950U using Tapatalk

[UT_EvoX]

I might come off as rude but you need professional tuning. This is 101 stuff

Sent from my SM-N950U using Tapatalk

Well, i dont take things personal. I am, however, pretty excited that i came to it myself! Thank you for confirming.

That's why i am on this forum guys. We all have to start somewhere.

I was tuned professionally by Chet once haha. None of these tables were changed like that even tho i was running 33+ psi.

I would imagine that this would be common knowledge by now.

Sent from my SM-G950U using Tapatalk

Again, this isn't personal, but that's terrible that you were tuned for that much boost but the SD/MAP load tables weren't rescaled...

I have to imagine that was just oversight. Running on MAF or not, it's good to scale those all the way out to the max your particular MAP can read.

Sent from my SM-N950U using Tapatalk

[mityaz]

I might come off as rude but you need professional tuning. This is 101 stuff

Sent from my SM-N950U using Tapatalk

Well, i dont take things personal. I am, however, pretty excited that i came to it myself! Thank you for confirming.

That's why i am on this forum guys. We all have to start somewhere.

I was tuned professionally by Chet once haha. None of these tables were changed like that even tho i was running 33+ psi.

I would imagine that this would be common knowledge by now.

Sent from my SM-G950U using Tapatalk

Again, this isn't personal, but that's terrible that you were tuned for that much boost but the SD/MAP load tables weren't rescaled...

I have to imagine that was just oversight. Running on MAF or not, it's good to scale those all the way out to the max your particular MAP can read.

Sent from my SM-N950U using Tapatalk

Let me ask you this. As far as extrapolating goes... for example, my high octane fuel map, not fuel cal map, ends with a certain number at 320 load. Once the ECU exits the boundaries of 320 load does it just maintain the last recorded number OR does it predict that number based on where the slope of the 3D table would project it to be? The fuel map in my case has all the same numbers so the slope of the 3D table would be very much exactly horizontal, thus the number woul be the same as the last recorded one (at load 320). But now that we look at a timing map, for example, there is now a non-horizontal slope present. What happens once i exit the boundaries here? Last recorded number or projected number according to the slope of the line graph at that point?
Thank you

Sent from my SM-G950U using Tapatalk

[razorlab]

As an aside, you should try smoothing out your MAP tables a bit. Since you are basically running your car off MAP full time, they could use some loving.

[mityaz]

As an aside, you should try smoothing out your MAP tables a bit. Since you are basically running your car off MAP full time, they could use some loving.

I know, i did. I just wanted to post a "proof of concept" type of picture.

Thank you

Sent from my SM-G950U using Tapatalk

[UT_EvoX]

I might come off as rude but you need professional tuning. This is 101 stuff

Sent from my SM-N950U using Tapatalk

Well, i dont take things personal. I am, however, pretty excited that i came to it myself! Thank you for confirming.

That's why i am on this forum guys. We all have to start somewhere.

I was tuned professionally by Chet once haha. None of these tables were changed like that even tho i was running 33+ psi.

I would imagine that this would be common knowledge by now.

Sent from my SM-G950U using Tapatalk

Again, this isn't personal, but that's terrible that you were tuned for that much boost but the SD/MAP load tables weren't rescaled...

I have to imagine that was just oversight. Running on MAF or not, it's good to scale those all the way out to the max your particular MAP can read.

Sent from my SM-N950U using Tapatalk

Let me ask you this. As far as extrapolating goes... for example, my high octane fuel map, not fuel cal map, ends with a certain number at 320 load. Once the ECU exits the boundaries of 320 load does it just maintain the last recorded number OR does it predict that number based on where the slope of the 3D table would project it to be? The fuel map in my case has all the same numbers so the slope of the 3D table would be very much exactly horizontal, thus the number woul be the same as the last recorded one (at load 320). But now that we look at a timing map, for example, there is now a non-horizontal slope present. What happens once i exit the boundaries here? Last recorded number or projected number according to the slope of the line graph at that point?
Thank you

Sent from my SM-G950U using Tapatalk

Without a higher load cell to reference, the ECM ends at the 320 row. As such, it's really important for your fuel and timing tables to encompass your desired load range so you have some safety net beyond normal operating parameters.

[razorlab]

As an aside, you should try smoothing out your MAP tables a bit. Since you are basically running your car off MAP full time, they could use some loving.

I know, i did. I just wanted to post a "proof of concept" type of picture.

Thank you

Sent from my SM-G950U using Tapatalk

as an example:

[UT_EvoX]

His example isn't that bad... looks like an actual VE/cylinder fill map to me. (just one of the factory maps IIRC)

If your VE map is smooth, you're probably not doing it correctly. Transitions on the X-axis should be pretty smooth, but you'll get big variances as you travel along the Y-axis due to the engine's natural VE characteristics. If the Y-axis resolution in the table was higher, you'd have smoother transitions from RPM breakpoint to RPM breakpoint; but big deal, the ECM smooths final load already.

[UT_EvoX]

Here's an example of how my ever-evolving standard cylinder fill map looks like on stock cam, stock exhaust manifold cars. As you add breathing mods and more efficient turbines, the top end does change a little. If you add cams and/or more displacement, that's where things start to change A LOT on the whole surface.



This matches damn close to MAF calcs (floats a little bit above it as I intend it to) under boost and elsewhere. If you took this table and multiplied by about 0.95 you'd be pretty close to an actual SD/cylinder fill map for stock cams/stock ex manifold/stock frame turbo.

If I were to flatten or smooth things out, it wouldn't be close at all to the more accurate MAF calcs in RPM regions where you have higher VE.

[mityaz]

Without a higher load cell to reference, the ECM ends at the 320 row. As such, it's really important for your fuel and timing tables to encompass your desired load range so you have some safety net beyond normal operating parameters.

Right. So this much i understand.
But for example, previously my highest column in the MAP table was at 29 PSI and the load values in it were peaking at 311. However, my load reporting did not stop at 311. it went all the way to 350+. Clearly there is some extrapolation going on. I just wanted to see if you knew of how exactly it happens as far as "going out of bounds"

[razorlab]

I have countless hours of large amounts of logs (some single logs are 100mb+) that I have analyzed into a scatter plots to tune my MAP + MAF tables.

I haven't heard one mention of people using scatter plots to tune cars so... hey what do I know.

[UT_EvoX]

Without a higher load cell to reference, the ECM ends at the 320 row. As such, it's really important for your fuel and timing tables to encompass your desired load range so you have some safety net beyond normal operating parameters.

Right. So this much i understand.
But for example, previously my highest column in the MAP table was at 29 PSI and the load values in it were peaking at 311. However, my load reporting did not stop at 311. it went all the way to 350+. Clearly there is some extrapolation going on. I just wanted to see if you knew of how exactly it happens as far as "going out of bounds"

Temperature modifier on MAP calcs is part of it - probably 5-7% worth of it. Can't say otherwise as I do see the same sometimes. It might extrapolate on the assumed slope for 1 more cell on the X-axis. But don't trust it beyond that as I know for a fact when mapped to only 29.xx PSI, if you touch the 33 PSI boost ceiling of the stock MAP sensor, you will flatten out on the MAP calcs before that 33 PSI ceiling. Not a big deal if your MAP is set 5-10% higher the MAF calcs and you're relying on the MAF.

This is one of the reasons other safety nets like OBTR, over load/boost protection/cut, etc. are important. If you exceed your 3xMAP mappings or fuel/timing etc tables' values, there's nothing to reel things in. I've recently started bringing my overboost/overload protection timer down to 250 ms as you can ABSOLUTELY destroy a stock engine in one hit just from a 3-port hose coming off or an end customer plumbing the thing incorrectly.

[UT_EvoX]

I have countless hours of large amounts of logs (some single logs are 100mb+) that I have analyzed into a scatter plots to tune my MAP + MAF tables.

I haven't heard one mention of people using scatter plots to tune cars so... hey what do I know.

All of my MAP tables are derived from hundreds of hours of logging and running through pivot tables, then manual tweaks based upon observances thereafter on many other cars and conditions (mostly on the top end).

If yours are flat and smooth, they're not right. No estimated VE or cylinder fill table that matches actual airflow is ever flat and smooth. We can agree to disagree.

Just look at the 15 HDFX VE slope tables on the EcoBoost platform... on a stock car, they get you within 3% STFT+LTFT under all conditions. None of them are flat/smooth along the RPM axis.

[mityaz]

Without a higher load cell to reference, the ECM ends at the 320 row. As such, it's really important for your fuel and timing tables to encompass your desired load range so you have some safety net beyond normal operating parameters.

Right. So this much i understand.
But for example, previously my highest column in the MAP table was at 29 PSI and the load values in it were peaking at 311. However, my load reporting did not stop at 311. it went all the way to 350+. Clearly there is some extrapolation going on. I just wanted to see if you knew of how exactly it happens as far as "going out of bounds"

Temperature modifier on MAP calcs is part of it - probably 5-7% worth of it. Can't say otherwise as I do see the same sometimes. It might extrapolate on the assumed slope for 1 more cell on the X-axis. But don't trust it beyond that as I know for a fact when mapped to only 29.xx PSI, if you touch the 33 PSI boost ceiling of the stock MAP sensor, you will flatten out on the MAP calcs. Not a big deal if your MAP is set 5-10% higher the MAF calcs and you're relying on the MAF.

This is one of the reasons other safety nets like OBTR, over load/boost protection/cut, etc. are important. If you exceed your 3xMAP mappings or fuel/timing etc tables' values, there's nothing to reel things in. I've recently started bringing my overboost/overload protection timer down to 250 ms as you can ABSOLUTELY destroy a stock engine in one hit just from a 3-port hose coming off or an end customer plumbing the thing incorrectly.

Got it. Thanks a lot. this makes sense. This is kind of what i was speculating too.

I dont have the equivalently small margin of error as the stock block would, as i am not terribly concerned about overboost, but your logic makes perfect sense. thanks for sharing

[razorlab]

I highly recommend processing your logs through scatter plots to help tune your tables, especially anything airflow, pressure, fueling related.

Once you understand how to read them, they are immensely helpful to tuning. Key is to get enough data in your logs.

I use EFI Analytics software for this.

[UT_EvoX]

Without a higher load cell to reference, the ECM ends at the 320 row. As such, it's really important for your fuel and timing tables to encompass your desired load range so you have some safety net beyond normal operating parameters.

Right. So this much i understand.
But for example, previously my highest column in the MAP table was at 29 PSI and the load values in it were peaking at 311. However, my load reporting did not stop at 311. it went all the way to 350+. Clearly there is some extrapolation going on. I just wanted to see if you knew of how exactly it happens as far as "going out of bounds"

Temperature modifier on MAP calcs is part of it - probably 5-7% worth of it. Can't say otherwise as I do see the same sometimes. It might extrapolate on the assumed slope for 1 more cell on the X-axis. But don't trust it beyond that as I know for a fact when mapped to only 29.xx PSI, if you touch the 33 PSI boost ceiling of the stock MAP sensor, you will flatten out on the MAP calcs. Not a big deal if your MAP is set 5-10% higher the MAF calcs and you're relying on the MAF.

This is one of the reasons other safety nets like OBTR, over load/boost protection/cut, etc. are important. If you exceed your 3xMAP mappings or fuel/timing etc tables' values, there's nothing to reel things in. I've recently started bringing my overboost/overload protection timer down to 250 ms as you can ABSOLUTELY destroy a stock engine in one hit just from a 3-port hose coming off or an end customer plumbing the thing incorrectly.

Got it. Thanks a lot. this makes sense. This is kind of what i was speculating too.

I dont have the equivalently small margin of error as the stock block would, as i am not terribly concerned about overboost, but your logic makes perfect sense. thanks for sharing

I would still guard against conditions that would push you beyond your 4-bar MAP's limits. I do on every high boost car I tune. In fact, it's even more important on such a high specific-output setup, in my opinion. You're pushing the limits as-is, you ought to put in some additional safety nets.

40 PSI on a built block vs. 25 PSI on a stock block. I'd say you're pushing the envelope a bit further as the stock block is far from "weak." LOL

[mityaz]

Without a higher load cell to reference, the ECM ends at the 320 row. As such, it's really important for your fuel and timing tables to encompass your desired load range so you have some safety net beyond normal operating parameters.

Right. So this much i understand.
But for example, previously my highest column in the MAP table was at 29 PSI and the load values in it were peaking at 311. However, my load reporting did not stop at 311. it went all the way to 350+. Clearly there is some extrapolation going on. I just wanted to see if you knew of how exactly it happens as far as "going out of bounds"

Temperature modifier on MAP calcs is part of it - probably 5-7% worth of it. Can't say otherwise as I do see the same sometimes. It might extrapolate on the assumed slope for 1 more cell on the X-axis. But don't trust it beyond that as I know for a fact when mapped to only 29.xx PSI, if you touch the 33 PSI boost ceiling of the stock MAP sensor, you will flatten out on the MAP calcs. Not a big deal if your MAP is set 5-10% higher the MAF calcs and you're relying on the MAF.

This is one of the reasons other safety nets like OBTR, over load/boost protection/cut, etc. are important. If you exceed your 3xMAP mappings or fuel/timing etc tables' values, there's nothing to reel things in. I've recently started bringing my overboost/overload protection timer down to 250 ms as you can ABSOLUTELY destroy a stock engine in one hit just from a 3-port hose coming off or an end customer plumbing the thing incorrectly.

Got it. Thanks a lot. this makes sense. This is kind of what i was speculating too.

I dont have the equivalently small margin of error as the stock block would, as i am not terribly concerned about overboost, but your logic makes perfect sense. thanks for sharing

I would still guard against conditions that would push you beyond your 4-bar MAP's limits. I do on every high boost car I tune. In fact, it's even more important on such a high specific-output setup, in my opinion. You're pushing the limits as-is, you ought to put in some additional safety nets.

40 PSI on a built block vs. 25 PSI on a stock block. I'd say you're pushing the envelope a bit further as the stock block is far from "weak." LOL

Point taken. I completely agree.

Sent from my SM-G950U using Tapatalk

[UT_EvoX]

I highly recommend processing your logs through scatter plots to help tune your tables, especially anything airflow, pressure, fueling related.

Once you understand how to read them, they are immensely helpful to tuning. Key is to get enough data in your logs.

I use EFI Analytics software for this.

I'll have to check it out.

Wouldn't you want a 3D plot for approximating a SD/VE/cylinder fill table though? MAP vs. MAFCalcs only works if RPM is fixed.

[razorlab]

It is 3D.

[mityaz]

@UT_EvoX

Hey, i tried doing some logging with the new MAP scaling. What ive found is that any load over 400 gets logged as zero. How do you manage to log loads over 400? Im assuming is has something to do with the logged bits and the formula. Thanks

Sent from my SM-G950U using Tapatalk

[UT_EvoX]

@UT_EvoX

Hey, i tried doing some logging with the new MAP scaling. What ive found is that any load over 400 gets logged as zero. How do you manage to log loads over 400? Im assuming is has something to do with the logged bits and the formula. Thanks

Sent from my SM-G950U using Tapatalk

It is just the logging. Tops out at 400 load.

The underlying calculation should be seeing greater than 400 load.

But IMO I think you're looking at higher load than you're actually making unless you're over 600 lb-ft.

Sent from my SM-N950U using Tapatalk

[athanasiosnoir]

@UT_EvoX

I am also getting 0 load readings when i hit the 400 sealing at 2.15 bar boost and only making 460 lb-ft.