Infineon bldc shields impressions

thanks @Antun_Skuric, that was helpful!

Cool,
Let us know how it goes. Iā€™m really interested too see some results and i really have great hopes for it.
Iā€™d like the btn8982 or ifx007t to be the bases for the higher power version of the foc shield :smiley:

And infineon guys have done a great job!

But i would like it to be a bit smaller, cheaper and bit more modular. :smiley:

To revive the discussion, does the current stable release support the ifx007T shield?

Hey Valentine,
Yes, it supptrs everything except that doesnā€™t not involve current sensing (back emf and foc using high-side currents on pins IS).

You can set three enable pins and three odma pins directly in the constructor of the BLDCDriver class and use it as any other 3pwm bldg driver.

1 Like

@Antun_Skuric,

Thank you, I was wondering about that, they hobbled the original nice design by back-porting it to fit the atmega instead of making future-proof.

Question: Regardless, if I connect an MCU capable of taking input from three external inline ACS772 sensors I can ignore the board weird current sense and should be able to run simplefoc with current sense?

Cheers,
Valentine

Hey @Valentine,

Yep, its a pity they did not continue their design further, there is a real potential in developing the boards like that one!

Yep, that will work.
Iā€™m in the process of developing a bit stronger verison of the SimpleFOCShield which uses a similar topology: GitHub - simplefoc/Arduino-SimpleFOC-PowerShield: A powerful Arduino shield for running BLDC motors using the FOC algorithm

At the end Iā€™ve removed completely the high-side current sense and added inline sensors to the board. It works well with the library. There is an example code in the dev branch at the moment:

@Antun_Skuric,

Thank you, that was very helpful.

Question 1: I see you added only 2 current sense resistors. Perhaps an obvious answer, but if I add three, does SimpleFOC take 3 or I have to feed only the first and third?

Edit: I see it does, found some time to browse the code. Please ignore my question, thanks.

Question 2: The power shield looks very nice. What is itā€™s availability / release date?

Thanks,
Valentine

Hey @Valentine,

Iā€™ve already thested all the versions in the main branch and they seem to work pretty well. I was able to push the shield to 25Amps on 8Volts (the only powerful lipo battery I have is 2S).
So you can actually already order the version v0.2 from the JLCPCB for example.

For them to be available in the shop I want it to be stackable, have the configurable pinout, have an additional LDO, and do some more testing. The boards is already almost there and the version v1.0 will I hope be available in June.
Iā€™ve ordered the latest versions few days ago on JLCPCB and they should arrive next week. The new board is in the dev branch for the moment.

Got around to test the board, I have a couple of these IFX007T boards but last three weeks were very busy with production work. Covid doesnā€™t help either.

The board works with simple foc out of the box, however I see huge power dissipation i cannot explain. In open loop velocity it gets really hot even without load, 0.5A with open leads, and heats up. When I attach a motor it jacks up the current to 4A just turning the motor. Iā€™m afraid attaching a large motor or running for more than a minute because it may pop the mosfets. Also the Mega is right under the mosfets, it gets hot too, not really helping. I will attach a temperature sensor and run the rest of the control loops and pull more data for different velocities and modes of operation. I may have to deep dive into the IFX007T mosfet physics to see why it heats up so much [PWM frequency???] Also need to first find an air cooled thermal solution, because the board is relatively large, and Iā€™m running it as a shield on top of a Mega, so dunking the whole setup in liquid cooling will cost me way more than everything else combined, and then cleaning the liquid is a royal pain. 3d design and print of a fan housing may take some time, too.

If you have any ideas why the power dissipation is so high please let me know.

Iā€™ll keep you informed but canā€™t commit to any timing, as my schedule is rather unpredictable.

Cheers,
Valentine

Hey, my guess is that it heats too much because youā€™re running open loop and maybe your voltage limit is too high. Open-loop is much more inefficient, youā€™re putting full power to the coils and moving the magnetic field hoping the rotor will follow.

I suggest you try to lower your voltage limit and suggest you try close-loop as youā€™re going to have much better performance.

Thank you, highly probable. I am just throwing things at the board at that point, will try this too.