Calum
09-09-2008, 10:57 PM
Here's the deal- the map tracing in Nissan DataScan gets the TP and RPM values via a special consult command that requests data by its address. The problem with this is unlike the 'gauge' command you can only get so many addresses, where as the 'gauge' command can stream up to 20 bytes. The new ECUs (USDM B14/S14 etc) support an extended data set for 'gauge' streaming that includes TP (load). That means you can have TP streamed with all the other gauges, assuming your consult software supports that. Here's how to add that functionality to a B13/S13 etc ECU-
First, we have to find the 'consult register' table in the ROM. This is the table that contains the addresses of the data that can be streamed via the 'gauges' command. Unfortunately unlike most tables the location of this table moves on different model years for the same ECU. Suck. Maybe a list of the table address vs year would be a good thing? :) Anywho, to find it, open your bin in a hex editor. I'm using XVI32, which is free and cool!
http://www.chmaas.handshake.de/delphi/freeware/xvi32/xvi32.htm
Search for the RPM MSB and RPM LSB RAM addresses, with the bytes inverted. To get those, look up your ecu in this post-
http://www.sr20forum.com/calumsult/174326-map-tracing-cool.html
So for a B13 those are 4475 and 4474, or inverted its '75 44 74 44', like so-
http://www.sr20-forum.com/usergallery/5448c7242808e71.png
Note the address of the first byte (the one highlighted in the picture), this is the address of the consult register table for the bin your working with. In this case its '7F50'.
Now we switch to TunerProRT. I've uploaded a new xdf for the B13 with the consult register table. If your using a S13 etc you need to patch this table into your XDF (I'll update the S13 etc XDFs later this week). If your consult register table is at a different address than the bin I'm working with, edit the xdf and change the address (right click on the table Consult Data Register, then Edit XDF Item Info, then put your address in the address box. Whew! Btw, the other B13 bin I looked at used '1010'.
Ok, now open up the table, and it should look like this (you should see your RPM MSB and LSBs as the first two entries 00, 01)-
http://www.sr20-forum.com/usergallery/5448c72427adf15.png
Every entry that is 'FFFF' is an entry that isn't used (stock) by the ecu. All the entries correlate to some standard data (mostly). You can look-up what they usually are here-
http://www.plmsdevelopments.com/images_ms/Generic_Nissan_ECU_Sensor_or_Data_register_table_V er_2.pdf
Now we just insert our TP MSB address into one of the unused spots. I like to use '06', which is normally for the second MAF sensor. Get your TP MSB address from the same thread above as the RPM MSB and LSB. For a B13 its 4292.
Ok, now run the 'CalumSult - TP streaming beta' exe, and on the start screen change the register index for TP from '33' to '06'. Change the other one to '06' as well (ignore this channel). You need to do this everytime you start the program. I'll have a more flexible version out this weekend where you won't have to do that.
http://www.calumsult.com/calumsu/CalumSult/
http://www.sr20-forum.com/usergallery/5448c7242879357.png
So with all that done you should be able to do a window'd pull and get a nice plot of TP vs RPM like this-
http://www.sr20-forum.com/usergallery/5448c72ab18446d.jpg
First, we have to find the 'consult register' table in the ROM. This is the table that contains the addresses of the data that can be streamed via the 'gauges' command. Unfortunately unlike most tables the location of this table moves on different model years for the same ECU. Suck. Maybe a list of the table address vs year would be a good thing? :) Anywho, to find it, open your bin in a hex editor. I'm using XVI32, which is free and cool!
http://www.chmaas.handshake.de/delphi/freeware/xvi32/xvi32.htm
Search for the RPM MSB and RPM LSB RAM addresses, with the bytes inverted. To get those, look up your ecu in this post-
http://www.sr20forum.com/calumsult/174326-map-tracing-cool.html
So for a B13 those are 4475 and 4474, or inverted its '75 44 74 44', like so-
http://www.sr20-forum.com/usergallery/5448c7242808e71.png
Note the address of the first byte (the one highlighted in the picture), this is the address of the consult register table for the bin your working with. In this case its '7F50'.
Now we switch to TunerProRT. I've uploaded a new xdf for the B13 with the consult register table. If your using a S13 etc you need to patch this table into your XDF (I'll update the S13 etc XDFs later this week). If your consult register table is at a different address than the bin I'm working with, edit the xdf and change the address (right click on the table Consult Data Register, then Edit XDF Item Info, then put your address in the address box. Whew! Btw, the other B13 bin I looked at used '1010'.
Ok, now open up the table, and it should look like this (you should see your RPM MSB and LSBs as the first two entries 00, 01)-
http://www.sr20-forum.com/usergallery/5448c72427adf15.png
Every entry that is 'FFFF' is an entry that isn't used (stock) by the ecu. All the entries correlate to some standard data (mostly). You can look-up what they usually are here-
http://www.plmsdevelopments.com/images_ms/Generic_Nissan_ECU_Sensor_or_Data_register_table_V er_2.pdf
Now we just insert our TP MSB address into one of the unused spots. I like to use '06', which is normally for the second MAF sensor. Get your TP MSB address from the same thread above as the RPM MSB and LSB. For a B13 its 4292.
Ok, now run the 'CalumSult - TP streaming beta' exe, and on the start screen change the register index for TP from '33' to '06'. Change the other one to '06' as well (ignore this channel). You need to do this everytime you start the program. I'll have a more flexible version out this weekend where you won't have to do that.
http://www.calumsult.com/calumsu/CalumSult/
http://www.sr20-forum.com/usergallery/5448c7242879357.png
So with all that done you should be able to do a window'd pull and get a nice plot of TP vs RPM like this-
http://www.sr20-forum.com/usergallery/5448c72ab18446d.jpg