msm z28

Can someone please spell out step by step on how to data trace the MAT Compensation Counts vs Mat.. I'm using tuner pro RT and a 730 ECM $8D

Not sure if I'm missing something.. when I look at the DA association I don't see the MAT comp counts listed as an item to data trace.

My car is running much better without the INV Table checked.. so now I'm doing minor adjustments to the MAT Comp table for hot vs cold days

A step by step would greatly help, since I'm kinda doing trial and error.. I'm getting there, but need to data trace this.. so I can see where it is live on a cold day vs hot day

Thanks in advance,
Sal

ULTM8Z

yeah, if you have a relocated MAT sensor, unchecking the Inv MAT table makes a huge difference, taking the cooltant temperature out of the calculation.

If you are logging WB O2 data with S_AUJP (which I highly recommend) you may be able to make use of this...

https://www.thirdgen.org/forums/diy-...ysis-tool.html

I've used it to dial in the MAT Compensation Counts table such that my AFR's are extremely consistent across all the MAT temperature ranges.

84Elky

Hopefully this will provide some insight into what's going on under the covers. I believe this is what you're referring to as "without the INV Table checked".


AUJP $8D Switch 0x018 b4
This is a very important Switch (Flag) because it significantly affects the calculation of the BPW Scaling Factor (L006D), which significantly affects the calculation of BPW.

When b4 is set, values from the following Tables are used to compute the BPW Scaling Factor in conjunction with air flow and a Coolant Temp/MAT differential:

• at 0x80E=COOLANT/MAT DELTA MULT .vs. AIR FLOW
• at 0x7FD=Air Flow .vs. MAT or Coolant/MAT Differential

But when b4=0, only MAT is used to compute the BPW Scaling Factor.

The table below provides code-simulated BPW values with the Switch set and not set:

Effect of Switch 0x018 b4 on BPW (Coolant=200°F, Air Flow = 32 Gm/Sec)


Observations:

• BPW is considerably richer when only MAT is used to compute the BPW Correction Factor (when b4 =0); versus using air flow and a coolant/MAT temperature differential (when b4 =1).
• The above example assumed 200°F coolant temperature. But the observed BPW changes would be “virtually identical” at any other coolant temperature because coolant temperature is not used in the calculation of BPW when 0x018 b4=0. The use of “virtually identical” is necessary because coolant temperature is used to compute an INT update delay factor that could have a minor effect on later BPW correction.

The above results show why it’s important to minimize temperature fluctuations and tune in a relatively constant MAT and coolant temp range where normal engine operation occurs. Otherwise there will be unwanted changes in INT and BLM.

It also shows that a more consistent tune can be achieved with a relocated MAT sensor that minimizes temperature fluctuations. The factory GM MAT sensor is actually a solid core coolant temperature sensor which is subject to heat-soaking and more prone to reporting the temperature of the TPI plenum rather than actual incoming air temperature.This is why replacing the factory MAT sensor with a sensor which measures true air temperature will provide a more accurate MAT (for example, AC Delco 213-243 [GM 12160244]). The relocation is generally best in the air intake before the throttle body, and as far from the engine as possible to avoid unnecessary heat. Note that MAT relocation will require some reduction in VE Table values because the lower MAT will naturally provide more fuel as shown above, even if b4 =1.

It could also be argued that if the MAT sensor is replaced and relocated, Switch 0x018 b4 should be set =0. The theory being that coolant temp will have virtually no impact on BPW, and the relatively constant MAT will ensure values from VE tables will produce a consistent BPW unaffected by coolant temp.

ULTM8Z

Right, disabling the Inv MAT table by setting b4=0.

Dominic Sorresso

But what about when u are sitting there at idle? My MAT is relocated and acts as an IAT now. I watch my BLMs change as the Intake temps rise due to lack of air stream. I have modified the MAT v CTS comp table in order to maintain a more consistent BLM across even varying intake air temps. For the most part I have done that. I see some variation when intake air temps reach into the 45C range, which does happen on a hotter day sitting in traffic. I don’t worry too much about that since its unlikely I will encounter that temps would be that high when the car is moving. Not to mention that driving the car in different seasons also has different “consistent” air temps.

Elky,

Is that table “universal” or does it pertain only to SAUJP?

84Elky

I experience the same fluctuations using the IAT described in the above post, and it's in an intake tube outside the engine compartment. Not much that can be done. But I look at it this way: want the best tune at "normal" coolant and intake air temps, realizing it will be a little rich before things warm up, and when intake air temp rises slightly while sitting still, things will lean out a bit but no adverse effects at low RPM.

Yes, above only addresses AUJP $8D (S_AUJP or otherwise), but likely similarly applies to any other GM mask that has the bit option to restrict the calculation to only use MAT/IAT. Not sure about your $D0. Maybe RBob knows.

As an aside, I run with the switch set (use air flow, etc.) because have everything dialed in. But on a lark the other day, I cleared the switch to use only MAT/IAT and across the board BLMs averaged approx. 10% richer. Not unexpected since the above table is confined to only one scenario at various MATs (coolant 200*F and 32 gm/sec air flow).

ULTM8Z

Basically what I did was tune one IAT temperature range to the right PE/OL AFR and BLMs using all the regular fueling tables (WB O2 obvioulsy required for PE/OL). After that I used the MAT Compensation counts table to tune the other IAT temp ranges using that link in my earlier post.

What I found was that the $8D code has the compensation counts spaced too far apart (12 counts per division). I ended up at around 3-5 counts per division.

My air tube has the filter completely outside the engine compartment with heat shielding around the filter. The IAT sensor is located directly in the filter. It's a great set up to the extent that the IAT temps are never more than about 15 degF over ambient temp.

Given the air tube is 100% non-metallic, I tune per the assumption that the air stream picks negligible heat before going into the engine. My BLM and AFR consistenc seems to bear that out.

msm z28

I guess my question wasn't super clear when I asked HOW to data trace a parameter that wasn't listed in my DA association ...

Sorry guys... looks like the conversation turned into something else...

I gave my reason WHY I wanted to trace it, but with all due respect that wasn't my question.. I just needed help on HOW to do it...

I literally asked this question 4 years ago and didn't get the answer then either... see this post
https://www.thirdgen.org/forums/diy-...trace-mat.html

I have learned a tremendous amount of information here and I'm grateful for that.. just like everyone else here we all learn in different ways and at different speeds. I just needed help with something very specific.. here's where it led me...

I have LITERALLY figured out this DAMN relocated IAT sensor tuning so my tune is on the money at various temps at WOT (which is critical!) The O2 sensor will adjust to some degree for part throttle and idle BLM numbers, but WOT is not so forgiving with hot and cold outside temp variations.. It's been frustrating over the last several years to experience WOT AFR ratios that varied by over .5 in AFR when the outside temps changed from 65 to 90 degrees F. I'm talking like 12.2 vs 12.8 WOT AFR

Here's a tip... it wasn't solved in one post... I have gathered little bits of info from several posts and applied theory along with trial and error to get my answer. There's a right way to get this IAT sensor functioning for OUR cars with the AUJP or applicable 7730 $8D variance.

I will finally be able to give MY contribution to this board very shortly on the findings that worked for me and I think it will work for many others here... My post will follow in a few more days.. I waiting for an early morning temp of around 50 degrees to finalize my conclusion.. so far 65 or 90 degrees F outside temp and same AFR at WOT

Elky.. I think this will work for your car based on what you just posted in the above post

msm z28

While I'm waiting for that cold morning to confirm my final results...

Any guesses what my final MAT Comp Counts table or MAT Inverse Lookup table looks like??

I'm curious to the feedback I may get here...

ULTM8Z

I'm actually curious to see what you ended up with compared to mine... lol

msm z28

You were one of the people that got me on the right track along with Elky and RBob.. lol

ULTM8Z

That table was like black magic to me for the longest time until I dove into and started playing around and seeing the results... now it's one of the most important and impactful tables in my tune.

RBob

iTrader: (1)

Since you aren't using the intake tract heating model, can grab the MAT from the data stream and use it. I'm not sure how to set up the data tracing and such in TP, so I'm not much help there.

Note that if the intake tract heating model was in use, it becomes more difficult. That final value isn't saved to RAM. So would need to change the code to do that first. Then change the ALDL stream to add it in.

RBob.

84Elky

Difficult to determine what you want to trace from item description because XDFs vary. Assume you want to see the values coming from the table at 0x4DE. If not that table, highlight the item in TPro, press F2 and provide hex address. I'll post how to see values if using S_AUJP.

Apologies - If need 0x7FD, no can do because as RBob stated, a value from that table is not stored in a RAM variable. If need 0x4DE, can do.

msm z28

DAMN.. NOT PERFECT YET

UPDATE.. I finally got the car out in 58 degree weather last night and at WOT it was on the rich side by about a half point (12.2 vs 12.8)... Originally the thing that made a huge difference was putting the IAT temp sensor right in front of the throttle body instead of the coldest spot way in front by the air filter... Logically I want my engine to respond to the air temp just as it enters the throttle body... not outside coldest temp.. I had more difficulty getting the tune right that way.. so its gonna stay right in front of the throttle body like a stock LT1 Vette.. The tuning has been much better like this.

Car runs WAY better without the INV table checked

I used the MAT Comp Counts table from BAFL bin for a LT1 1992 Corvette. Since I'm using a similar sensor and location to a LT1 Vette, I thought I'd be close and I am... my motor has an intake very similar to a miniram

BUT.. not perfect.. so, I'm thinking that the lower values of the table below 56c need to be increased so I'm a little leaner in the cold weather.. I think Dominic was right when he only needed to change a couple of values in the MAT comps table..

This leads me to HAVE to data trace the table at 0x7FD so I can know which value I'm adjusting.. With the cooler nights.. it's play season for me.. I want to get this right..

Ultm8Z.. I'd be curious if you fine tuned your MAT comps table a little more since the release of your miniram bin.. If so PM me and I'll give you my email

Elky84.. I'm gonna need to figure some way to tune 0x7FD without a blind fold on... lol

Rbob.. I currently datalog my traditional MAT value for the incoming air, but I read its not exactly the same temp that is adjusted in the MAT comps table, so I looking for some direction on tuning the MAT comps table.. I'm not gonna use the INV table anymore.. car runs way better unchecked

Thanks everyone.. still committed to getting this right... There's a lot of Mustangs waiting for me to play.. lol

ULTM8Z

yeah I dialed in the MAT tables a little more since then. only thing is mine is built around having the MAT sensor right at the filter location under the fender.

you'll notice that the values are a lot closer together than what GM originally programmed in.

but the fact is now the car drives the same in any weather condition and at any operating temperature.

ultm8z@yahoo.com

ps... not sure if you've seen my MAT tuning tool... its focused on PE, but I used it to dial in my PE AFRs across all MAT temps using only the MAT comp counts table after dialing them in at one MAT. temperature. I'll send that to you as well.

RBob

iTrader: (1)

You can either use a lookup table to convert the AD MAT into an actual value. Or change the ALDL stream to use the linear MAT value. This one is easy to convert:

C° = N * .75 - 40

RBob.

84Elky

Here's my $.02. Trying to be objective. Comments address WOT and 0x018 b4=0 to use only MAT because that appears the focus of OPs posts.

0x7fd Table Values: It's not possible to report values retrieved from this table without a code hack to add a variable. Not easy, so sorry . But are they really needed? The value returned is labelled "AirFlow Counts in gm/sec". Note "Counts". This is not true density-air-flow in gm/sec because every value returned from the table is significantly less than that which would result from the ideal gas law gm/sec at WOT (220+ gm/sec). The 0x7fd value appears merely to be what the GM engineers determined necessary to compute 0x006D (the BPW Temperature Scaling factor), which is later used for calculation of initial BPW before any adjustment or compensation.

But what is known is that changing values in the table will linearly affect the 0x006D scaling factor, which in turn will proportionately change initial BPW. Also note that more "counts" are returned as temp increases. Hotter = leaner, so need more fuel. So while you can't see 0x7FD directly, you can log the following to see the effect of 0x7fd. Just place the following hex values in an ALDL scalar provided in the S_AUJP XDF:
80 times a second
...006F=Calculated grams air per cylinder (a product of the scaling factor, kPa, EGR partial pressure and kPa-adjusted VE (16 bit; just report raw for relative comparisons)
...006B=Calculated Air flow from ideal gas law (gm/sec) (a product of gm air per cylinder and RPM (8 bit; no conversion)
...00E2=BPW (this will not be the initial PW, but is after all compensations) (16 bit; Conversion: X * 0.015259)

10 times a second
...006D=BPW temperature scaling factor (a product of 0x7fd, airflow and grams air per cylinder (16 bit; just report raw for relative comparisons)

Regarding the closer grouping of values in 0x7fd (as opposed to the factory spacing of 12 counts per entry): This merely causes a smaller range of returned counts across a range of temperatures. It's similar to entering the same value, for example, for 46, 68 and 90 *F. Important: The effect of those identical entries on the resulting 0x006B scaling factor and thus on BPW with all other things being equal, will be identical in the temperature range of 46-90 *F -- always resulting in the same BPW.

IAT Sensor Location: (Addressing "Logically I want my engine to respond to the air temp just as it enters the throttle body... not outside coldest temp." At WOT, air is being passed into the engine at 220+ gm/sec (see reporting "0066B" above). Even without detailed calculation of true air velocity from gm/sec, it's hard to imagine that air sits in the intake tube at WOT velocity, whatever it is, long enough to attain the temp at the TB. Additionally, an IAT sensor will pick up the temp of it's surroundings -- even outside the engine compartment, but significantly less than if inside. Don't understand how temp inside the compartment can be less variable that that outside. Variable temp is not what is desired.

Bottom line for 0x7fd table at WOT: Must first determine a range of temps in which WOT will be invoked. Then set 0x7fd table values in that temp range to the singular value that provides the desired BPW (ie - WBo2 AFR). If temperature changes, BPW will always change proportionately with 0x7fd, so it's desirable that temperature fluctuations are at a minimum.

What affects BPW calculation, which is what ultimately determines the WBo2-read WOT AFR (from code examination):
...kPa adjusted VE
...BPW Temperature Scaling factor above (controlled by 0x7fd)
...kPa
...EGR partial pressure if installed
...Grams air per cylinder (computed from above items)
...Limited air flow from ideal gas law (from grams air per cylinder and RPM)
...Injector flow rate
...Cylinder volume
...Calculated theoretical AFR while in PE

As can be seen, the only variable item that can be adjusted to affect BPW aside from VE is the scaling factor which results from a value from 0x7fd.

Sorry for the long post.

84Elky

Your question prompted me to look at this because it always been puzzling. Yes, there are 2 different values, but they serve the same purpose. Here's how things work. When MAT/IAT is read from the sensor, it's provided as an inverted value. But it is next inverted by the code. So an inverted value from the sensor is code-inverted so that it is no longer inverted. What??? True!

Using an example:

• Assume a MAT/IAT of 88*F at the sensor. The sensor is read and returns 0xDF (223dec). We don't really know this, but because of the next statement, we do.
• Next, the sensor value is inverted: 255dec - 223dec = 32dec (0x50) (True, NOT-inverted raw MAT/IAT). We now know 0xDF (223dec) was returned by the sensor to the ECM.
  • This 0x50 value cannot be used for any table lookups. It needs to be scaled or linearized for that purpose. That's where a table internal to the code comes in (LF133, see tables attachment).
• The 0x50 true raw MAT/IAT is passed to this internal table and a scaled/linearized value of 0x5F is returned. This is the value required for various Table look-ups such as 0x7fd MAT Counts (This scaled/linearized value can be logged by setting an ALDL Scalar to "0063"). Note 0x5F differs from 0x50 (the true raw MAT/IAT), which hopefully explains what you read about 2 different values.
• The 0x5F from LF133 is next used to query the 0x7fd table to get the "MAT Counts", which results in 0x57 (87dec Counts) being returned (see code table).

A lot of detail but necessary.
Next, look at the MAT table in the ALDL data stream at 0x013c (see code table).This is the same table used to report MAT in TPro. Using the true raw MAT/IAT input value of 0x50 (80dec) results in an extrapolated output value of 88*F.

More, which is interesting. - The temperatures shown in the XDF item as input to Table 0x7fd are meaningless. Remember from above, the input is not true temperature, but a scaled value. Further, there can be a variety of input values to 0x7fd depending upon the setting of 0x018 b4. If b4=0, The input is scaled MAT/IAT. But if b4=1, the input is a MAT/IAT-Coolant differential that varies all over the place depending upon air flow, coolant temp, and MAT/IAT. Because of this, the correct XDF item should have no input temperatures. This is why it's impossible to correlate output with an input value. Just have to trust what 0x7fd provides based on the scaled input. Again, can determine the scaled value by logging "0063". Correction-Can't log this if b4=1 because no RAM variable.

DISREGARD XDF INFO BELOW -- SEE POST #23.
A more meaningful XDF version is shown below:


Better Description:


Again, sorry for the long post, but this is what happens under the covers and leads to the confusion.

ULTM8Z

damn....where's the smiley or emoticon for "I'm not worthy!" lol... amazing how you're able to decipher this stuff!

Street Lethal

iTrader: (16)

Something doesn't seem right. Do you have a datalog of both instances before and during wide open throttle when AFR is consistent, as well as the one during the 58 degree temperature that came after where it deviated a tad? I just want to verify something...

- Rob

84Elky

Looking at this stuff gives me brain cramps but helps me learn. You are more than worthy my friend!

84Elky

Having thought more about the 0x7fd MAT table matter, what we want to know is the temperature entering the internal code table LF133 that results in a number of "Counts" from table 0x7fd. And those values are known based on code simulation. The temperatures in the 0x7fd *F and *C XDF tables below represent MAT if b4=0, or essentially a Coolant/MAT differential if b4=1. These values enter Table LF133 and provide an output that is input to the 0x7fd XDF table to provide Counts.

The proper XDF information is below. Some of the temps in the middle of the table are those in the original table. Disregard the XDF info in Post #19 above:


Better XDF Description:
Counts from this Table are used to compute a BPW Temperature Scaling Factor (0x006D). Generally, it's not necessary to adjust this table, but any upward/downward change in Counts will result in an increased/decreased PW.

The temperatures shown as Input to this Table are based on the Flag 0x018 b4:
- If b4=0, Input = MAT
- If b4=1, Input = a Coolant/MAT differential
When b4=1 and Coolant >= MAT, the input is MAT plus an adder resulting from a calculation involving air flow and the difference between Coolant Temp and MAT (essentially a "Differential Temperature Factor"). When b4=1 and Coolant < MAT, Input is MAT.

ULTM8Z

Elky, I've read a couple of other threads as well on the MAT Compensation Counts (MCC) table where it was said that "generally" don't have to change that table. I'm not sure what the rationale is behind that though...

Based on the results I got with my WB O2 tuning, focusing exclusively on this MCC table (and I spent months with it, including moving the sensor around), it seems like if you change to a true IAT sensor (rather than that coolant sensor in the plenum "masquerading" as a MAT sensor), and a true cold air tube (not just relocating the filter to the side of the engine compartment, but getting it completely outside of it), you definitely want to play around with this table.

As I was saying in an earlier post, I started the process by dialing in my VE tables and PE/OL AFR's as well as the AE tables at one temperature range of that MCC table. Then as weather permitted (opportunities to mess with this table depend on the local weather), little by little dialed in the fueling at other air temps using only the MCC table (after having disabled the Inv MAT table) in order to get the same WB O2 results as with initial temperature range. The level of operational consistency now is like nothing I've ever had in the ~15 years I've been running this Miniram manifold.

I've moved the sensor from at the air filter to just in front of the throttle body, but didn't detect any significant impacts on the WB results, so it seems to back up the assertion that heat transfer along that length of ABS tube is neglible during engine operation.

84Elky

I've done exactly what you've done except change the MCC Table. No problem with doing that, but the reason for the cautions is that some don't know the effect of making those changes. You do. And if one has a WBo2, it may make sense as you've discovered. But the real reason you're seeing little change in AFR (ie - BPW) with table count spacing smaller than factory =12 in a given temperature range, is because that tells the code to apply a factor to BPW calculation with less consideration of temperature fluctuations, and to keep counts relatively constant within that range.

ULTM8Z

I agree without a WB, yeah it's inadviseable to change things where you can't see the resultant actual AFR...