LAUNCHXL-F28069M

Patrick Fisher2 said:

-Wht's the eta on motorware examples for driving dual motors with this LP?

The plan is to release a 2 inverter control software project in MotorWare _15, scheduled for this summer

Patrick Fisher2 said:

-How independently can the motors be driven with a single processor? Can a highly dynamic load on one motor adversely effect the other as a result of the mcu being tied up?

real world system events have no effect from the standpoint of MIPS or interrupts.  There will be some restrictions based on the overall control loop times for each inverter so you don't run out of MIPS. There will also likely be some restrictions on the PWM (and hence ADC end of conversion interrupt) rates for the two inverters to guarantee proper sampling and no overlap in information into each control system. 

Patrick Fisher2 said:

-Any obvious problems with tying two identical inverters together to effectively double the current rating? -Would you process the feedback from both inverters and 'average' that feedback or disregard one set off feedback signals completely? Would it be dangerous to assume a fairly even current division among the two boards or would the signal quality be too damaged by slightly uneven current flows? -Maybe more difficult than I first thought???

This is out of my expertise. I'll see if I can get the DRV guys to respond. This is very atypical though...normally you would // just the FETs and not have an entirely duplicate 3PHI.

Patrick Fisher2 said:

-What's the state of the comms peripheral drivers in motorware (mainly CAN & I2C) - I had some trouble six months ago when I2C wasn't quite ready in motorware?

Minimal changes.  Most of the software files were cleaned up / aligned to an internal standard / effort but there are still major gaps in ease of use / completeness from a driver / API level IMO.

Thanks Chris

ChrisClearman said:

This is out of my expertise. I'll see if I can get the DRV guys to respond. This is very atypical though...normally you would // just the FETs and not have an entirely duplicate 3PHI.

I'm shooting for a purpose multicopter esc so driving two motors from one mcu is fairly useful - in any case, it's a pretty cool gimmick. Ostensibly it seemed pretty simple to tie the two 3PHI together to drive a single motor for use in helis/cars/etc but I might just put that idea on the back-burner for the time being. I'm also going to dump the 8301 pre-driver so it's probably some dead fets in waiting...

ChrisClearman said:

Minimal changes.  Most of the software files were cleaned up / aligned to an internal standard / effort but there are still major gaps in ease of use / completeness from a driver / API level IMO.

Is there any impetus at TI to get the MW drivers up to speed or am I better off going it alone? -Anyone on e2e that's made a start on CAN bus that you're aware of?

I'm about to start on CAN - I've tried to probe TI from multiple fronts on this issue (missing MW drivers), but got no reply. Soon I've exhausted all other things I can work on the project, so I'm going to have to attack this quite soon.

Chris, can you give me (and Patrick) some pointers on how to get started? I guess we're going to have to integrate the ControlSUITE CAN register definitions into the MW "environment" and work from there? What's the easiest way to achieve this?

I thought that new drivers would be coming when I saw the built-in CAN tranceiver on the new LaunchPad, but I guess it's for sometime in the future, then...

I've sent a note directly to Patrick.

We are still delinquent on certain drivers. I'm hoping we make progress in the next two updates this year.

Patrick,

Is is generally not recommended to parallel two seperate inverters unless they have been closely matched. You can have issues with mismatch in the turn ON and turn OFF of the gate drivers. This can lead to one driver still actively driving HIGH while the other is actively driving LOW which will cause some nasty problems.

The other less serious side effect (as you pointed out) is mismatches in the current flow but this is a less immediete concern.

Hey Chris

I too have a need to drive one device with two synced bldc motors to maintain target speed with increased torque.

How close is the MW version 15 to release, and will it contain the dual motor lab as you indicated?

Mathworks "Simulink" allows the exact syncing as seen in this video www.youtube.com/watch;feature=youtu.be using the F28069M LP, so we know it's doable!

Can or will CCS allow for simple "property setting" for dual motor syncing as in Simulink?

Where can I find documentation for the F28069M LP / BOOSTXL-DVR8301 labs that define the intent of each lab like there is for the F28027 LP / BOOSTXL-DRV8301?

Thanks, Bruce

MW 15 is releasing end of this week, but the 2 motor example is NOT included. We will release the example in MW 16 which will release by early January.

The Simulink package is not using the instaspin libraries (which is the major SW challenge) it is just using a standard encoder based FOC similar to examples in controlSUITE.