PDA

View Full Version : New Binary Editor


86GT
01-16-2006, 10:23 PM
For those that use CalEdit and dis like it, there is a new binary editor program availiable . This program is similar to PCMX Explorer. It graphs the tables and functions. It can even compare 4 binaries and show all of the differences. It highlights changes and so on. The best news is that it is free! Yes you read that right FREEBIE!!!!

I would suggest making a donation and getting the pro version to keep his fine work up. I personally have the A9L and A9P files.

http://tunerpro.markmansur.com/ (http://tunerpro.markmansur.com/)

93Cobra#2771
01-17-2006, 08:02 AM
Clint - I assume the tweecer code hasn't been busted to allow burning to it? Would you save the file as a binary, and then open the file with the tweecer and burn to tweecer? Or what?

QWKSNKE
01-17-2006, 08:03 AM
about to check it out :nice:

Dale McPeters
01-17-2006, 01:43 PM
:)
Should have more xdf files available shortly.

fireguy50
01-17-2006, 01:48 PM
interesting

86GT
01-18-2006, 10:09 PM
Yes, you can simpley take a base binary and then make the changes that you want with TunerPro and then save it back to the same or new bin file. You then only need CalEdit to open the bin file and build your datalog payload and download to the tweecer.

Cal Edit now is only a download tool and a payload builder.

93Cobra#2771
01-19-2006, 01:48 PM
HOT DAMN! A solution for the quirky tweecer software, and no longer a need to rely on software updates that NEVER come. Now, if he could just get it to datalog... :D

86GT
01-19-2006, 04:37 PM
That would be the ultimate. I'm sure some one could use a USB sniffer to find out what is being tansfered to reverse it. I was able to reverse the payloads. I know the parameter address. I just don't know how to communicate to it.

gt90stang
01-21-2006, 12:27 AM
Clint,

I checked out the A9P and many of the scalars have wrong values, like injector slopes and others.

Don

86GT
01-21-2006, 10:10 AM
Clint,

I checked out the A9P and many of the scalars have wrong values, like injector slopes and others.

Don

Opps that is correct. I spent alot of time getting the Functions and tables correct and started on the scalars but did not finish. The scalars are shifed X amount of registers in the A9P and I just did not get it all completed.

I will have to look at that sorry.

Cougar5.0
01-22-2006, 10:10 AM
How come when I change a single value in the MAF transfer in TunerPro, CalEdit says "File is not a valid EEC Binary"?

I can change tables galore in TunerPro no problem when opening binary in CalEdit.

Also, why does TunerPro change values in MAF just like CalEdit does?

This sucks - MAF's are a friggin' NIGHTMARE! I'm going to have to use the "copy" function in CalEdit to get the MAF into my new binary - assuming everything else created in TunerPro works??

Cougar5.0
01-22-2006, 10:43 AM
OK, I figured it out already.

The A9L.xdf definition file has the Y axis "output type" set up as "integer" - it should be "floating point". If the drop-down is changed to "floating point" the bin will load into CalEdit fine with just the usual round-off/conversion errors (that weren't so bad).

86GT
01-22-2006, 08:35 PM
Below is a link to my xdf for the A9L and A9P.

Don I sent some time last night updating the A9P to have the same paramerters as the A9L. They should be correct now.


http://webpages.charter.net/eecanalyzer/Doc/

Dale McPeters
01-23-2006, 09:29 AM
You guys need to try out the moates emulator......

gt90stang
01-23-2006, 02:38 PM
Where do we find the Moates emulator?

Don

Cougar5.0
01-23-2006, 02:43 PM
Below is a link to my xdf for the A9L and A9P.

Don I sent some time last night updating the A9P to have the same paramerters as the A9L. They should be correct now.


http://webpages.charter.net/eecanalyzer/Doc/

Those definition files look great. I added the function/constant name to the title so it's easy to search for a particular function. I suppose the Item Finder could be improved to make this unnecessary. I don't suppose you would know how to find the constant TTNOV? :D

Cougar5.0
01-23-2006, 08:16 PM
You guys need to try out the moates emulator......

Well, I don't know what that is, but I was reading a post on Tweecer forum where you were talking about using torque truncation to pull timing to save your tranny. Did you ever attempt this? Did you ever locate the scalar TTNOV? Are the values in the table degrees of timing pulled?

86GT
01-23-2006, 10:19 PM
Those definition files look great. I added the function/constant name to the title so it's easy to search for a particular function. I suppose the Item Finder could be improved to make this unnecessary. I don't suppose you would know how to find the constant TTNOV? :D

Don found an error in the A9P so I fixed it and reposted them. I have asked Mark if he could display the function number or FORD ID and he said it can not be displayed or sorted upon. This is one feature that I really like about the PCMX. I can sort by description, Ford ID, or address. Those are very helpfull.

You could do it the way you described but you are still limited to one type of sort.

86GT
01-23-2006, 10:19 PM
You guys need to try out the moates emulator......

Tell me more!!!

Dale McPeters
01-24-2006, 09:29 AM
It is called the Ostrich

http://www.moates.net/product_info.php?cPath=31&products_id=57


Product Description: This is THE standalone USB emulator. It might be the ONLY one. It will emulate 2732A, 27C128, 27C256, 27C512, and more! Up to 4mbit (29F040) 8-bit emulation. Robust, USB-based realtime EPROM emulation in the palm of your hand! Battery backup lasts 75 years (standard AAA), so you can leave it in the car if you want to. With this device, you can upload an entirely new binary without a hiccup while the car is running. Or, you can change one parameter at a time. Fully compatible with TunerPro RT and TunerCat.

It works with Paul Booths latest EEC-Editor software as well.....

QWKSNKE
01-24-2006, 05:27 PM
Ok. I have the TunerPro version downloaded. What is the difference in TunerPro and TunerPro RT?

Also, I am confused on how this works? Do I use this or CalEdit to read from the EEC and if I am still using CalEdit, how do I transfer .CCF files to .XDF files?

ncontt
01-25-2006, 04:26 PM
Ok. I have the TunerPro version downloaded. What is the difference in TunerPro and TunerPro RT?

Also, I am confused on how this works? Do I use this or CalEdit to read from the EEC and if I am still using CalEdit, how do I transfer .CCF files to .XDF files?

:werd: :shrug:

Dale McPeters
01-25-2006, 04:30 PM
Ok. I have the TunerPro version downloaded. What is the difference in TunerPro and TunerPro RT?

Also, I am confused on how this works? Do I use this or CalEdit to read from the EEC and if I am still using CalEdit, how do I transfer .CCF files to .XDF files?

Email me your ccf file. I will convert it to binary for you and explain the xdf file stuff.
I will email you an explaination of how to do it yourself in the future.
I would also suggest that you download the EEC-Editor from Craig Moates website.
Here is the link for it. ( http://www.moates.net/zips/boothware )
I would also suggest that you become a member of this group
http://groups.yahoo.com/group/eeceditor/

86GT
01-25-2006, 07:04 PM
You must be using CalEdit to push the CCF into the tweecer and then using the F2A to read back the tweecer????

This means that you can figure out all of mikes def files. It may take a while but it can be done. By changing a single parameter and downloading it to the tweecer and then reading it back you can determine the changed memory loacation.

Not bad!! Maybe some one can right a utility to do it automatically and create the def files for the other processors.

Dale McPeters
01-27-2006, 09:56 AM
You must be using CalEdit to push the CCF into the tweecer and then using the F2A to read back the tweecer????

Well kind of but not exactly.......


This means that you can figure out all of mikes def files. It may take a while but it can be done. By changing a single parameter and downloading it to the tweecer and then reading it back you can determine the changed memory loacation.

Not bad!! Maybe some one can right a utility to do it automatically and create the def files for the other processors.

Well I really do not have that much time anymore. Doing it that way would be very very time consuming and I just do not have that kind of time anymore. I needed the capability to add a few of my own parameters for a few things that I was working on. And when the '05 release of Caledit never made it to production I had to make a decision as I had quite a few things I had been working on. There are a lot of people that work on the DIY Ford EEC stuff that are a lot smarter than I am, trust me. When I finish up the few minor details that I am currently working on for dual maf's and a few other things my EEC tinkering time will drop pretty much to nothing. The AEM EMS will finally be ordered late this afternoon. Thanks to Bob Kurgan sending me an email to remind me yesterday as I had been out of town/office off and on for the past couple of weeks.

By the way my internet service and home pc were down last night so I did not get to look for the file that you asked about maybe tonight it will be back up and running. If so I will see what I have and drop you an email.

86GT
01-27-2006, 12:39 PM
No problem, busy bee.

05 release, what is that.:tlash:

Dale McPeters
01-27-2006, 12:45 PM
05 release, what is that.:tlash:

That was a year that there was supposed to be some kind of software
release of some kind but for the life of me I can not remember what it
was supposed to be.........:D

By the way what other def files do you need besides the A9S.
I know that I sent you a lot of stuff but can not remember
what all I sent.....:popcorn:

sailorbob
01-27-2006, 02:27 PM
I don't suppose you would know how to find the constant TTNOV? :DI believe it's at 0x9448 in the A9L.

86GT
01-27-2006, 04:24 PM
By the way what other def files do you need besides the A9S.
I know that I sent you a lot of stuff but can not remember
what all I sent.....:popcorn:

I was going to compile a xdf for the A9S. I don't remember which ones you sent me I think it was the A9L only. Send me what you got:chug:

Cougar5.0
01-27-2006, 06:28 PM
I believe it's at 0x9448 in the A9L.

Seems to show up as "Idle Spark Multiplier" (value = 1)?

The TTNOV number if it's in there would be (or a multiple of):

(assuming 25.6" diameter 225-60-15)

AOD w/2.73's - 87
AOD w/3.08's - 95.2

T5 w/2.73's - 117.6
T5 w/3.08's - 133.3

I'm not sure which tranny/rear gear they would use as the default...

fireguy50
01-28-2006, 12:43 AM
Is TTNOV the number of pulses per tire revolution?

Cougar5.0
01-28-2006, 01:13 AM
Is TTNOV the number of pulses per tire revolution?

According to GUFB, it's simply N/V or engine RPM over vehicle speed in 1st gear.

sailorbob
01-28-2006, 04:26 AM
I pretty much 100% sure than the address 0x9448 is correct because the code does this: if rpm >= (Average speed * RPM to MPH Ratio for First Gear) than (rpm / 2) then skip Torque Reduction Table lookup

sailorbob
01-28-2006, 10:00 AM
You guys need to try out the moates emulator......
Tried the FordEmu out today with the Ostrich, didn't get chance to play for long but all I can say is that it worked excellently and I cannot wait to get back to it. My ecu doesn’t work with the F1 adapter so I was a bit concerned that there would be a problem but Craig’s done another great job with this and all went smoothly.

Cougar5.0
01-28-2006, 11:20 AM
I pretty much 100% sure than the address 0x9448 is correct because the code does this: if rpm >= (Average speed * RPM to MPH Ratio for First Gear) than (rpm / 2) then skip Torque Reduction Table lookup

OK, so that's the logical operation that decides whether to use FN1119 (Torque Reduction Table) if RPM is greater than or equal to VSBAR (average speed) * TTNOV (RPM to MPH ratio for first gear).

And you're saying that the scalar value for TTNOV is contained at 0x9448 which is great!

I just want to verify why the definition for A9L (TunerPro) created by Clint lists P_7448 (which is equivalent to 0x9448 - no?) as Idle Spark Multiplier - SPKMUL the default value FF (x/256) - is calculated at 1.00. The definition file on the TunerPro page lists 9448 (hex) with a value of B4 or 180 but no Item Name. But for some reason, not a single default value makes any sense when I use the new definition on the TunerPro page (dated 1/23) - perhaps someone can tell me what I have to do to correct this since now the hex addresses read 9xxx etc but the bin values are nonsense. I'm sure it's something simple that us non-digital engineers don't know. I see a note about "padding" the file by 8k to make it 64k - don't know how to do this...

Edit:

OK, I did a "base offset" of -2000 (hex) and that seemed to realign things (which aligns with the 8k "padding" - why not set up the header like this - or is the header independant of the "def" file?). Was that the right way to fix the issue and can someone explain "base offset"?

This number may not need a multiplier. 255 would tend to disable the function since you would need 6.30:1 gears with a T5 to have it kick in - lol. Now if it's x/2, then it would just kick-in with a T5 & 2.73's (or bigger). I guess I'll be doing some experimenting!

If anyone's curious, I'm trying to figure out how to help me launch at the track & eliminate tire spin with a PD blower...

sailorbob
01-28-2006, 01:43 PM
Hmm, don't know why Clint has got Idle Spark Multiplier at 0x7448, he's also got Spark Retard For Knock Sensor there too. Please bear in mind it's easy to make a mistake in these xdf's, there's a lot of stuff in there! I'm always grateful when someone points these things out.

I think you've sorted out the offset and different bin file size problem, I've posted a reply on the TunerPro forum in more detail on your query.

The value of 255 is, as you say, a disabling value. As there is no check in this part of the code to see if the car is manual or auto the only way to avoid doing the Torque Reduction Table lookup is to set the value to a near impossible (impractical?) value.

Cougar5.0
01-28-2006, 01:51 PM
Hmm, don't know why Clint has got Idle Spark Multiplier at 0x7448, he's also got Spark Retard For Knock Sensor there too. Please bear in mind it's easy to make a mistake in these xdf's, there's a lot of stuff in there! I'm always grateful when someone points these things out.

I think you've sorted out the offset and different bin file size problem, I've posted a reply on the TunerPro forum in more detail on your query.

The value of 255 is, as you say, a disabling value. As there is no check in this part of the code to see if the car is manual or auto the only way to avoid doing the Torque Reduction Table lookup is to set the value to a near impossible (impractical?) value.

lol - so you're the same person who moderates the TunerPro forum - I was wondering :D I really appreciate the help - this could be a really helpful to me if it will work. I don't suppose you know if TwEECer will write the information not contained in it's editable inputs?

sailorbob
01-28-2006, 05:29 PM
I don't suppose you know if TwEECer will write the information not contained in it's editable inputs?Sorry, no. I don't know much about the Tweecer.

86GT
01-28-2006, 11:34 PM
Sorry for the confusion guys, I had two parameters marked as genius level in the PCMX application and this 7448 is one of them. the other is 6EAA (VIPSPK). I used the Genius level in PCMX for address that I did not know or confirm. Again sorry for the confusion.

I have updated My def files to reflect this change.

86GT
01-28-2006, 11:50 PM
Oh yah I forgot to mention that the address is 7430 if you are using the A9P and 7448 if you are using the A9L.

For the most part the scalars are off by 18h in the A9P

fireguy50
01-29-2006, 12:58 AM
According to GUFB, it's simply N/V or engine RPM over vehicle speed in 1st gear.
What RPM is the TTNOV calculated at, 2000RPM?

Cougar5.0
01-29-2006, 11:13 AM
What RPM is the TTNOV calculated at, 2000RPM?

Well, I calculate it at 2000 RPM for convenience, but it is a constant for each gear.

e.g. 2.73's, 3.35 1st gear ratio, 25.6" diameter tire.

MPH @ 1000RPM - 8.4MPH, 2000RPM - 16.8MPH, 3000RPM - 25.2MPH

N/V @ 2000RPM = 119, @ 2000RPM = 119, @3000RPM = 119

TTNOV is there so that the ECU can determine if the vehicle is in 1st gear (or not). If it is, FN1119 spark reduction is applied.

N/V in second gear is 69.

For this example above, you could set TTNOV at, say 100 & the FN1119 would thus pull timing in first, but not second.

If N > TTNOV * VSBAR (avg speed) then apply FN1119 - 'C', else 'C' = 0

Spark formula;

SAF = Y * ['S' + B - C + FN1133(ECT, LOAD)] +KPS1

Notice how 'C', the torque truncation spark table, FN1119, is subtracted from the calculated spark?

This table will be applied any time TTNOV is set low enough that when multiplied by the rolling average speed of the car, it is smaller than the engine RPM at that speed.

86GT
01-29-2006, 12:55 PM
I don't suppose you know if TwEECer will write the information not contained in it's editable inputs?

CalCon will write the entire 2000h adress and above to the tweecer. I have changed things like the injector reference, which is not in CalCon, and it worked. I only use the CalCon to open the binary and download it to the tweecer.


sailorbob Nice find!!!

Cougar5.0
01-29-2006, 04:12 PM
CalCon will write the entire 2000h adress and above to the tweecer. I have changed things like the injector reference, which is not in CalCon, and it worked. I only use the CalCon to open the binary and download it to the tweecer.


sailorbob Nice find!!!

Did you mean CalEdit? Anyway, I spent the time to manually convert my latest tune in to a binary in TunerPro. I burned it on CalEdit and it was a total corruption-fest. Fuel pump would not shut off & even when I started the car with the switch back on the unchanged old tune, much of the data shown in Calcon was garbage. I have to re-burn all of the possible payloads to get the data-collection back to normal. I'm pretty bummed right now as I'm not savvy enough to know what happened. I do know that comparing the binaries in TunerPro show many changes that I never even made - some to "undefined" addresses and some to functions I never ever touched!! Just pasting in my MAF curve over the existing one changes 167 bytes, only 18 of which are directly linked to the MAF change. I'm baffled to say the least.

Cougar5.0
01-29-2006, 08:00 PM
Well, I was able to create a tune where I just generated the constants in TunerPro & then copied all the functions from a known good tune in TwEECer. The car started & ran fine but I still can't get Calcon to cooperate - 1/3 of the parameters are garbage now. I'm annoyed at this point.

Dale McPeters
01-29-2006, 08:51 PM
I don't suppose you know if TwEECer will write the information not contained in it's editable inputs?

Oops I see Clint has already answered this......:D

86GT
01-29-2006, 10:52 PM
Did you mean CalEdit?

Yep!!

Which binary did you start with? are you sure it was the A9L and not the X3Z or vise versa?

fireguy50
01-29-2006, 11:22 PM
Well, I calculate it at 2000 RPM for convenience, but it is a constant for each gear.
When I do the math it varies at different RPM.

4.10's, 2.40 1st gear ratio, 28.88" diameter tire.

Speed calc done with EEC Analyzer 2.7.0.0
1000RPM = 9MPH, 2000RPM = 17MPH, 3000RPM = 26MPH, 4000RPM = 35MPH, 5000RPM = 44MPH

N/V @ 1000RPM = 111, @ 2000RPM = 117, @ 3000RPM = 115, @ 4000RPM = 114
, @ 5000 - 113

86GT
01-29-2006, 11:34 PM
fireguy it could be because of rounding. The difference between the 111 and 117 at 2000 RPM is 1MPH.

Cougar5.0
01-30-2006, 12:04 AM
Yep!!

Which binary did you start with? are you sure it was the A9L and not the X3Z or vise versa?

Yeah, it was A9L for sure. All of the data looked as expected in CalEdit - and I have changed many functions/scalars/tables - I verified all the data as I made changes - Caledit occasionally had issues with filling the highest data point if it was also the highest number for the number of that byte (i.e. 65534, 65535 etc...) - strange roundoff issues happening here.

Filling the MAF and tracking the changes, the flow rates below point 16 (15 & under) will show up as a change in 1 byte at an appropriate address and changes to two bytes at very low numbered addresses. For example, if I change Y axis cell #15 to "100", I get address 6E20 filled with 3E (versus 1D with the stock value of "90") but address 003C & 003D are also changed from the original values. Seems odd that 3 bytes are changed for these fields. Change all the X axis cells, you'll get something with a ton of low address cells filled, but most of which are associated with another function all together? Seems messed up to me and my engine surely woudn't run with this so something is wrong...

Note: A/C Time Delay vs TP (FN087) will write to the same addresses as MAF Transfer starting at 0000 - seems odd to me... A/C Flow Max vs RPM is doing the same thing starting at 000C & EGR Desired Drop vs Flow starting at 0014.

For example, change FN087 Y-axis point 6 & do a "difference tool" & then change MAF Function FN036 Y-axis point 16 & watch how the value in address 0000 changes.

Something seems to be misaligned somewhere...

Dale McPeters
01-30-2006, 08:47 AM
Tried the FordEmu out today with the Ostrich, didn't get chance to play for long but all I can say is that it worked excellently and I cannot wait to get back to it. My ecu doesn’t work with the F1 adapter so I was a bit concerned that there would be a problem but Craig’s done another great job with this and all went smoothly.

Yep Craig has done an excellent job on this piece of hardware. ;)
By the way the F3 adapter works well with pcm's that the F1
does not support.

86GT
01-30-2006, 10:43 PM
Yeah, it was A9L for sure.

I wish I could answer the weird things that you are seeing but I cant. It does not make logical sense that it would makes changes in more than one function or the MAF making changes to three bytes at a time. Tis could only happen if the def file was not set up correctly or if the memory locations were crossed.

This was in Cal Edit that you saw this in?

You might have to start with a fresh binary and make all of the chnages again. Some how it got corupt. I used PCMX to make all of my binary edits. I have used TunerPro on a few parameters and seems to work. I have yet to try a fresh binary from TunerPro.

Cougar5.0
01-30-2006, 11:56 PM
I wish I could answer the weird things that you are seeing but I cant. It does not make logical sense that it would makes changes in more than one function or the MAF making changes to three bytes at a time. Tis could only happen if the def file was not set up correctly or if the memory locations were crossed.

This was in Cal Edit that you saw this in?

You might have to start with a fresh binary and make all of the chnages again. Some how it got corupt. I used PCMX to make all of my binary edits. I have used TunerPro on a few parameters and seems to work. I have yet to try a fresh binary from TunerPro.

I'm using the A9L that I got from TwEECer site. I'm doing the "difference tool" in TunerPro to see the changes AFTER I change particular functions. Most work fine except for MAF & a couple of others. I've used THREE different definition files (yours, dex's & one I made) and the issues appear every time. I can change the filling by messing with the def file, but I haven't been able to stop the weird behavior with certain cells as described above. Either my computer (work laptop that I do heavy CAD on) is messed up or TunerPro is messed up (I don't think the definition files are messed up but they may be). Maybe you could try using a fresh copy of A9L in TunerPro & make the changes I described above & use the "difference tool" to verify what I am talking about.

NOTE!! The data looks OK in CalEdit when I look at it BUT there are bytes that are changed that don't show up in CalEdit, and the resultant bin from TunerPro corrupted the crap out of my Tweecer - I still can't get CalCon to log correctly!! I doubt anyone else has done a tune entirely in TunerPro as I have - especially MAF curves since I was the first one to find that the file wouldn't work in CalEdit due to integer format for "Y" values. I hope someone can verify this since I'm no expert in digital engineering. Mechanical - yes, electrical - yes, digital - NO! just learning...

Edit: Also note that I took the SAME A9L starting file and changed ALL of the scalars for my current tune in TunerPro & then copied all the functions in CalEdit and the tune works fine also. No bytes got filled "out-of-place" when I just changed scalars.

Oh, and I hope I'm not coming across like a grump - just that I've been sick as a dog for a week and i had a little while beween raindrops to burn my new tune and it was nothing but trouble - I don't mean to take it out on anybody :D

86GT
01-31-2006, 09:32 AM
I will try making a change to my bin in the MAF area to see if it works when I get home tonight.

Dale McPeters
01-31-2006, 10:04 AM
Something seems to be misaligned somewhere...

Drop Mark Mansur an email as he may have an explaination for this..;)

sailorbob
01-31-2006, 01:50 PM
Currently the 'Item Finder' tool and 'Difference' tool in TunerPro will not work for functions, Mark is aware of this and hopefully will release a fix soon.

Cougar5.0
01-31-2006, 07:34 PM
Currently the 'Item Finder' tool and 'Difference' tool in TunerPro will not work for functions, Mark is aware of this and hopefully will release a fix soon.

Glad Mark is aware of this issue. Those tools are working fine according to EEC_Ed BTW - glad the issue is recognized though :chug: