Paul and Sabrina's Cheap 3 Phase Inverter (AC Controller) with Field Oriented Control

[P-hack]

here are the output files using xc16 from the original instructables sources, in the latest mplab (on ubuntu).

[Astro]

Originally Posted by P-hack

here are the output files using xc16 from the original instructables sources, in the latest mplab (on ubuntu).

I had a try at this but received errors during the build. Most likely due to the way i setup MplabX. Until now i have only used the XC8 compiler with my other projects so i may not have added in the XC16 compiler correctly.
Any chance you could post a link to an archive of the project folder? I just want to eliminate the chance that i missed adding a file or put them in the wrong folders.

[P-hack]

yah you will need xc16 for the 16 bit mcu.

here is the zipped project folder (with .xls added due to upload limits on .zip, remove it from the filename)

THIS IS A STATIC SNAPSHOT, make no assumptions about how current it is, like the above, the intent was to test some hardware with the xc16 compiler BEFORE having to sort out the hundreds of megs of install, but hopefully it will help too. The code is surely out of date.

edit, sorry bout the contest paul. here are the finalists if anyone is interested:
MAKE ENERGY: A US-Mexico Innovation Challenge

lots of mediocre, the bioconverters are kinda interesting, liked the Vapour TorQ heat engine. Am pissed the hho cell made the list (Celda seca de hidrógeno) :/ seriously, wtf?


edit, this doesn't have sources in it :/ will try a different folder.

[Astro]

Originally Posted by P-hack

yah you will need xc16 for the 16 bit mcu.
here is the zipped project folder (with .xls added due to upload limits on .zip, remove it from the filename)

I couldn't successfully add that folder as a project. I think it is because it is getting late and my brain went to bed a couple of hours ago.
I will tackle it again tomorrow.

[P-hack]

Ah, the zip is the X folder, which doesn't have source. Will update from git later and try again.

[P-hack]

didn't see any changes in git, and I'm not sure how helpful this is (since I just changed some case names and a filename or two), and there seems to be an impedance mismatch between git and mplabx, so my files might seem like they are in a strange location. But attached are the sources I used on linux. I am not intending to fork anything here, just a snapshot in time.

Also the built files are in post 1771.

here is the clean/build log if it helps too:

PHP Code:


CLEAN SUCCESSFUL (total time: 70ms)
make -f nbproject/Makefile-default.mk SUBPROJECTS= .build-conf
make[1]: Entering directory `/var/home/dave/Desktop/proj/mpcontroller/MPLABXProjects/mpcontrol.X'
make  -f nbproject/Makefile-default.mk dist/default/production/mpcontrol.X.production.hex
make[2]: Entering directory '/var/home/dave/Desktop/proj/mpcontroller/MPLABXProjects/mpcontrol.X'
"/var/progs/microchip/xc16/v1.24/bin/xc16-gcc"   ../../ACInductionController/ACInductionController.c  -o build/default/production/_ext/1746335589/ACInductionController.o  -c -mcpu=30F4011  -MMD -MF "build/default/production/_ext/1746335589/ACInductionController.o.d"        -g -omf=elf -O0 -msmart-io=1 -Wall -msfr-warn=off
"/var/progs/microchip/xc16/v1.24/bin/xc16-gcc"   ../../ACInductionController/UART4011.c  -o build/default/production/_ext/1746335589/UART4011.o  -c -mcpu=30F4011  -MMD -MF "build/default/production/_ext/1746335589/UART4011.o.d"        -g -omf=elf -O0 -msmart-io=1 -Wall -msfr-warn=off
"/var/progs/microchip/xc16/v1.24/bin/xc16-gcc"   -o dist/default/production/mpcontrol.X.production.elf  build/default/production/_ext/1746335589/ACInductionController.o build/default/production/_ext/1746335589/UART4011.o      -mcpu=30F4011        -omf=elf -Wl,,--defsym=__MPLAB_BUILD=1,,--script=p30F4011.gld,--stack=16,--check-sections,--data-init,--pack-data,--handles,--isr,--no-gc-sections,--fill-upper=0,--stackguard=16,--no-force-link,--smart-io,-Map="dist/default/production/mpcontrol.X.production.map",--report-mem 


Program Memory  [Origin = 0x100, Length = 0x7f00]

section                    address   length (PC units)   length (bytes) (dec)
-------                    -------   -----------------   --------------------
.text                        0x100               0x170           0x228  (552)
.const                       0x270               0xff0          0x17e8  (6120)
.text                       0x1260              0x1a60          0x2790  (10128)
.dinit                      0x2cc0               0x174           0x22e  (558)

                     Total program memory used (bytes):         0x43ce  (17358) 35%


Data Memory  [Origin = 0x800, Length = 0x800]

section                    address      alignment gaps    total length  (dec)
-------                    -------      --------------    -------------------
.ndata                       0x800                   0           0x1ac  (428)
.nbss                        0x9ac                   0           0x188  (392)
.ndata                       0xb34                   0            0x58  (88)
.nbss                        0xb8c                   0            0x40  (64)

                        Total data memory used (bytes):          0x3cc  (972) 47%


Dynamic Memory Usage

region                     address                      maximum length  (dec)
------                     -------                      ---------------------
heap                             0                                   0  (0)
stack                        0xbcc                               0x434  (1076)

                        Maximum dynamic memory (bytes):          0x434  (1076)

"/var/progs/microchip/xc16/v1.24/bin"/xc16-bin2hex dist/default/production/mpcontrol.X.production.elf -a  -omf=elf  
make[2]: Leaving directory '/var/home/dave/Desktop/proj/mpcontroller/MPLABXProjects/mpcontrol.X'
make[1]: Leaving directory `/var/home/dave/Desktop/proj/mpcontroller/MPLABXProjects/mpcontrol.X'

BUILD SUCCESSFUL (total time: 2s)
Loading code from /var/home/dave/Desktop/proj/mpcontroller/MPLABXProjects/mpcontrol.X/dist/default/production/mpcontrol.X.production.hex...
Loading completed 


and here is the info from the mplabx dashboard for the project

PHP Code:


/var/home/dave/Desktop/proj/mpcontroller/MPLABXProjects/mpcontrol.X
mpcontrol
Project Type: Application - Configuration: default
Device
dsPIC30F4011
 Checksum: 0xD1AE
Compiler Toolchain
XC16 (v1.24) [/var/progs/microchip/xc16/v1.24/bin]
Production Image: Optimization: gcc 0
Memory
Usage Symbols disabled. Click to enable Load Symbols.
Data 2048 (0x800) bytes
Program 16384 (0x4000) words
Debug Tool
 PICkit2
Debug Resources
Program BP Used: 0  Free: 2
Data BP Used: 0  Free: 2
Data Capture BP: No Support
Unlimited BP (S/W): No Support

Other Project Info
Data Reserved Memory
Production Image
Program Reserved Memory
Production Image

Control Points.

Java NetBeans Properties
MPLAB X IDE v2.30 - mpcontrol : default
                   os.name : Linux
                   os.arch : amd64
                os.version : 3.13.0-32-generic
       sun.arch.data.model : 64
          sun.java.command : org.netbeans.Main --userdir /home/dave/.mplab_ide/dev/v2.30 --branding mplab
              java.version : 1.7.0_67
           java.vm.version : 24.65-b04
      java.runtime.version : 1.7.0_67-b01
java.specification.version : 1.7
             netbeans.user : /home/dave/.mplab_ide/dev/v2.30
             netbeans.home : /var/progs/microchip/mplabx/mplab_ide/platform
     netbeans.projects.dir : /home/dave/MPLABXProjects
            java.io.tmpdir : /tmp
                 user.home : /home/dave
              user.country : US
             user.language : en
             user.timezone : America/Chicago
          sun.jnu.encoding : UTF-8
             file.encoding : UTF-8
            file.separator : /

Java & System Memory
         Java Total Memory =    247 Mb
         Java Used Memory  =    112 Mb
         Java Free Memory  =    135 Mb
CommittedVirtualMemorySize =   3770 Mb
    FreePhysicalMemorySize =    585 Mb
    MaxFileDescriptorCount =      0 Mb
   OpenFileDescriptorCount =      0 Mb
            ProcessCpuLoad =      0 Mb
             SystemCpuLoad =      0 Mb
   TotalPhysicalMemorySize =   3754 Mb

System Path
  PATH = /home/dave/bin:/var/progs/jdk/bin:/var/progs/adk/sdk/tools:/var/progs/adk/sdk/platform-tools:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/games:/usr/local/games:/var/progs/microchip/xc8/v1.33/bin:/var/progs/microchip/xc16/v1.24/bin 


so if you unzip this into directory /home/dave/Desktop/proj/mpcontroller the mplab project posted above should work as is, but I don't recommend it long term.

[Astro]

Well this is all strange.
I extracted those files to a folder in the mplabx projects directory.
Then in mplabx i created a new stand alone project using that folder.
Added the source and header files from the folder to the project.
Did a clean and build and hey presto it worked.
I hadn't changed anything else in mplabx so i am not sure what i was dong wrong before.

CLEAN SUCCESSFUL (total time: 51ms)
make -f nbproject/Makefile-default.mk SUBPROJECTS= .build-conf
make[1]: Entering directory `/home/jase/MPLABXProjects/ACInductionController'
make  -f nbproject/Makefile-default.mk dist/default/production/ACInductionController.production.hex
make[2]: Entering directory `/home/jase/MPLABXProjects/ACInductionController'
"/opt/microchip/xc16/v1.24/bin/xc16-gcc"   ACInductionController.c  -o build/default/production/ACInductionController.o  -c -mcpu=30F4011  -MMD -MF "build/default/production/ACInductionController.o.d"        -g -omf=elf -O0 -msmart-io=1 -Wall -msfr-warn=off
"/opt/microchip/xc16/v1.24/bin/xc16-gcc"   UART4011.c  -o build/default/production/UART4011.o  -c -mcpu=30F4011  -MMD -MF "build/default/production/UART4011.o.d"        -g -omf=elf -O0 -msmart-io=1 -Wall -msfr-warn=off
"/opt/microchip/xc16/v1.24/bin/xc16-gcc"   -o dist/default/production/ACInductionController.production.elf  build/default/production/ACInductionController.o build/default/production/UART4011.o      -mcpu=30F4011        -omf=elf -Wl,,--defsym=__MPLAB_BUILD=1,,--script=p30F4011.gld,--stack=16,--check-sections,--data-init,--pack-data,--handles,--isr,--no-gc-sections,--fill-upper=0,--stackguard=16,--no-force-link,--smart-io,-Map="dist/default/production/ACInductionController.production.map",--report-mem 


Program Memory  [Origin = 0x100, Length = 0x7f00]

section                    address   length (PC units)   length (bytes) (dec)
-------                    -------   -----------------   --------------------
.text                        0x100               0x170           0x228  (552)
.const                       0x270               0xff0          0x17e8  (6120)
.text                       0x1260              0x1a60          0x2790  (10128)
.dinit                      0x2cc0               0x174           0x22e  (558)

                     Total program memory used (bytes):         0x43ce  (17358) 35%


Data Memory  [Origin = 0x800, Length = 0x800]

section                    address      alignment gaps    total length  (dec)
-------                    -------      --------------    -------------------
.ndata                       0x800                   0           0x1ac  (428)
.nbss                        0x9ac                   0           0x188  (392)
.ndata                       0xb34                   0            0x58  (88)
.nbss                        0xb8c                   0            0x40  (64)

                        Total data memory used (bytes):          0x3cc  (972) 47%


Dynamic Memory Usage

region                     address                      maximum length  (dec)
------                     -------                      ---------------------
heap                             0                                   0  (0)
stack                        0xbcc                               0x434  (1076)

                        Maximum dynamic memory (bytes):          0x434  (1076)

"/opt/microchip/xc16/v1.24/bin"/xc16-bin2hex dist/default/production/ACInductionController.production.elf -a  -omf=elf 
make[2]: Leaving directory `/home/jase/MPLABXProjects/ACInductionController'
make[1]: Leaving directory `/home/jase/MPLABXProjects/ACInductionController'

BUILD SUCCESSFUL (total time: 1s)
Loading code from /home/jase/MPLABXProjects/ACInductionController/dist/default/production/ACInductionController.production.hex...
Loading completed

Thank you P-hack.
Now i have a foundation for making any minor changes that the forum or Paul might come up with and building the hex file myself.

Paul does amazing work but it is nice to know that i can make minor changes myself.
I am not saying i will but it is nice to not feel reliant on Paul's generous nature.

[Astro]

Originally Posted by thingstodo

High V side

Battery pack neg -> precharge resistor (with precharge contactor across it) -> B- on controller

Battery pack pos -> 80 amp DC braker -> primary contactor -> B+ on controller

thingstodo, i am not sure about the pre charge contactor set up you have described.
I had a quick look at Pauls code and it looks like the pre charge voltage is turned on, then after a 1 second delay the main contactor voltage is turned on. But then, after a further delay of 0.2 seconds the pre charge contactor voltage is removed.

This seems designed to work with something like circuit 2 below whereas your description sounds more like circuit 1.



Circuit 1 would require the main contactor close first. Controller charges through pre charge resistor. Then the pre charge contactor is closed and the main contactor stays closed. Both contactors must stay closed whilst the controller is driving the motor.

Circuit 2 has the pre charge contactor closing first. Controller charges. Then the main contactor closes. The pre charge contactor is now bypassed by the main contactor and so can be de energised (opened). Only the main contactor needs to remain closed while the controller is driving the motor.

If you have already wired your vehicle then the start up sequence could be changed in the controller software.
Remove the two lines that cause the pre charge contactor to open.
And swap the main and pre charge variables.

Before changes

    O_LAT_PRECHARGE_RELAY = 1;  // 1 means close the relay.  Now the cap is filling up.
    DelayTenthsSecond(10); // savedValues.prechargeTime);

    // High pedal lockout. Wait until they aren't touching the throttle before closing the main contactor.
    do {
        ReadADInputs();
    } while (ADThrottle < savedValues.minRegenPosition || ADThrottle > savedValues.minThrottlePosition);  // I'm using a wig wag hall effect throttle.

    O_LAT_CONTACTOR = 1;    // close main contactor.
    DelayTenthsSecond(2);   // delay 0.2 seconds, to give the contactor a chance to close.  Then, there will be no current going through the precharge relay.
    O_LAT_PRECHARGE_RELAY = 0;  // open precharge relay once main contactor is closed.

After changes

    O_LAT_CONTACTOR = 1;  // 1 means close the main relay.  Now the cap is filling up.
    DelayTenthsSecond(10); // savedValues.prechargeTime);
    
    // High pedal lockout. Wait until they aren't touching the throttle before closing the main contactor.
    do {
        ReadADInputs();
    } while (ADThrottle < savedValues.minRegenPosition || ADThrottle > savedValues.minThrottlePosition);  // I'm using a wig wag hall effect throttle.

    O_LAT_PRECHARGE_RELAY = 1;    // close precharge contactor.
//    DelayTenthsSecond(2);   // delay 0.2 seconds, to give the contactor a chance to close.  Then, there will be no current going through the precharge relay.
//    O_LAT_PRECHARGE_RELAY = 0;  // open precharge relay once main contactor is closed.

A disadvantage to running with a precharge circuit like circuit 1 is that both the pre charge and main contactors must be rated to carry the full traction current. Whereas with circuit 2 the pre charge only has to carry a fraction of the current and only for 1 second. So the pre charge contactor can be a much lower cost device.

[P-hack]

awesome you got it compiling (I hate being the only one ).

reminder the compiler has probably changed dramatically and the binaries still need hardware re-validation, as well as validating the optimization level is correct and operates the inverter hardware correctly.

[thingstodo]

Originally Posted by Astro

thingstodo, i am not sure about the pre charge contactor set up you have described.

Paul showed me that the pre-charge circuit I have will not work with the controller as programmed.

I will change my circuit for testing to match circuit 2.