Questions regarding BLDCSHIELDIFX007T and IFX007T

Announcements

From sunburn to sun earn – we’ve got the power! Watch our #poweringgreen videos now.

Tip / Sign in to post questions, reply, level up, and achieve exciting badges. Know more

cross mob
User18465
Level 1
Level 1
First question asked First reply posted
I am considering using IFX007T Half bridge ICs in a motor control design. For evaluation of these ICs I have bought a
BLDCSHIELDIFX007T board, which contains 3 of them. I have done testing and measurements with this board in combination
with an Arduino loaded the provided software examples (https://github.com/Infineon/IFX007T-Motor-Control). There are several
things that I would like to ask regarding this board and regarding the IFX007T devices

1. Output capacitors
During testing of the example Arduino code for sensorless BLDC control, I noticed the devices getting very hot. After running
for a few minutes, the devices would heat up to over 90 degrees Celsius, even though the BLDC motor was running slowly and
without load.

I noticed that in this code, some special settings are done that allow a PWM frequency of 31250Hz. From some of your examples
I made an Arduino code that switches one of the IFX007T at this frequency. The other two devices on the board were idle. I noticed that:

1. The supply current, without any load on the output, is around 230mA (at 16V supply, so 16*0.23 = 3.68W !!!
2. After letting the setup run for 5min, I measured a temperature of 90 degrees Celsius with a thermocouple

Additional testing indicated that the power consumption without load attached is PWM-frequency dependent. With the Arduino,
I supplied one of the IFX007T devices with a certain PWM frequency and measured the supply current. The other devices were
disabled. A supply voltage of 18V was used.
Current consumption vs. PWM frequency for 1 IFX007T (board unmodified)
3892.attach

It appeared that the supply current at 0 load, and therefore the used power, is directly proportional to the input PWM frequency.
In your datasheet, it is mentioned that a PWM frequency of 25kHz easily be reachable. But I did not expect a power loss of
18V * 0.2A = 3.6W for just one of the IFX007T ICs. When I let all 3 IFX007T ICs switch at this frequency, the power loss also triples.

I noticed that on the BLDCSHIELDIFX007T board, capacitors are placed between the output and both Vsupply and GND. This is
according to the application notes document, which states that these capacitors are needed to improve EMI performance. The
recommended capacitance value of both capacitors is 220nF, this is rather large for use in such an application.
220nF capacitors on the output
3894.attach

When an increasing PWM frequency is used, these capacitors are charged and discharged more often, which requires current to do.
Therefore I tried to desolder these capacitors on one of the IFX007T devices on the board. This resulted in the following PWM
frequency vs. Supply current graph:
Current consumption vs. PWM frequency for 1 IFX007T (desoldered output capacitors)
3893.attach

When the capacitors are removed, the high supply current consumption is also gone. At frequencies above 25kHz, the current
consumption with zero load is increased more than 20 times (!!).

All of these measurements were taken with zero load on the output, the consumed power is dissipated mostly in the IFX007T
devices, therefore they get very hot. And when the output capacitors are removed, the problem is solved completely.

I now have several questions regarding the recommended output capacitors:

1. Why are output capacitors recommended in IFX007T designs when they will cause a large power loss if moderately high
PWM frequencies are used?
2. In your BLDC BEMF Arduino example code, a PWM frequency of 31.250kHz is used, therefore you imply that the IFX007T
can handle this frequency. But the power loss and excessive heat generating at this frequency is not mentioned anywhere. Why is that?
3. Can I make a motor controller design which uses IFX007T devices as its power stage, without using the output capacitors?
Or perhaps capacitors with a much smaller capacitance to significantly reduce power loss?


2. Active freewheeling
In your product advertisement, it is mentioned that the IFX007T is capable of active freewheeling. This would mean there is no
need for additional freewheeling diodes. However, I have been unable to find if this feature is automatically carried out in the
digital logic inside the IFX007T, or if it has to be done manually by controlling the IN and INH pins in the device. If so, how can
this be implemented? If I am not mistaken, this is not implemented in the Arduino example code.

Thanks in advance for the help.
0 Likes
3 Replies
JingN
Moderator
Moderator
Moderator
First like received First solution authored 25 replies posted
Thanks very much for your interest.


Please find some comments below:

  • PWM frequency: 31250Hz is only the base frequency. The resulting frequency is equal to the base frequency divided by
    the given divisor. Please choose the needed PWM frequency by using the proper divisor. Sorry for that, if this base frequency makes you confused. We don't recommend you to use this device with the frequency higher than 25kHz.
  • Power dissipation: Could you please clarify how you measured the supply current? In your case, there is no load connected, we can calculate how much current you need to charge/discharge the output capacitors. And the switching power lose can be calculated according to the application note(page 57 - page 59). Your calculate is not correct. The current doesn't flow from VS to GND in IFX007T. Most of them flows from VS to the output cap or from the cap to GND. Therefore please follow the application note to calculate it. I suggest you to try our power dissipation tool in Infineon tool box. We support power calculate for BTN8982TA in the tool. Physically BTN8982TA and IFX007T are identical. Only the specs are provided with different conditions. Please use it to get some hints for the power dissipation of IFX007T. How to install: Infineon Toolbox-->Manage Tools-->Power Dissipation

  • Output capacitors: 220nF is recommended in the application note to improve the EMC performance. But customers can select a proper value according to their own requirements and applications.
  • Active freewheeling: Yes, the active freewheeling and the cross current protection is integrated in the device. You don't need to put additional effort there. E.g. When you INPUT is switched from HIGH to LOW, the device will keep measuring if the high side is completely switched off. Only if the high side is switched off, the low side will be switched on.
0 Likes
User18465
Level 1
Level 1
First question asked First reply posted
Hi, thank you very much for your reply and your time.

I measure the supply current by putting a multimeter on current measure mode in series with the power supply that I use to power the BLDCSHIELDIFX007T board. So the supply voltage multiplied by the consumed current gives me the consumed power. And since I noticed that the IFX007T devices are the only devices getting very hot, this power must be dissipated inside those.

You are correct that the current flows from VCC to the output capacitors when the HS switch is on, and that the current flows from the output capacitors to GND when the LS switch is on. But this current is taken from the power supply and eventually returned to ground.

I have done some calculations to determine the theoretical current that is required to charge and discharge the capacitors at the desired frequency of 25kHz:

Formula for stored charge in a capacitor:
C = Q / V

Where:
- C is Capacitance in Farads
- Q is Charge in Coulombs
- V is voltage in Volts

I use a supply voltage of 18V. So the charge stored in the 220nF capacitor is 220x10^-9 * 18 = 3.96x10^-6 Coulombs

1 Coulomb is equal to 1 Ampere * 1 second (Q = I * t)

So charging a 220nF capacitor to 18V requires 3.96x10^-6 As (Ampere seconds)

When I use a 25kHz PWM signal, there are 25000 charge and 25000 discharge cycles each second.

So then the total needed charge for 1 second of operation is 3.96x10^-6 * (25000*2) = 0.198 As (Ampere seconds)

Since t is 1 second here, the average consumed current for charging/discharging is 0.198A.

When I use my Arduino to generate a PWM frequency of around 25kHz like this (I checked this with an oscilloscope):
3899.attach

I indeed measure a power supply current of around 200mA. This is the case for one of the IFX007T devices on the board with original, unmodified capacitors. When I edit the code to use an IFX007T device from which I desoldered the output capacitors, I measure a current consumption of only 11mA. All of these measurements were done without any load attached on the output. So the high current consumption, and therefore the IFX007T devices reaching very high temperatures, is really caused by the output capacitors.

I downloaded your Toolbox application but I could not find the BTN8982TA listed as a selectable device. However, it is already clear that the zero-load power consumption and heat generation is caused by the output capacitors, and not the switching losses in the IFX007T itself.

Regarding the output capacitors, I think I have enough information for now.


Regarding the Active freewheeling feature, this is clear as well. Thank you!
0 Likes
JingN
Moderator
Moderator
Moderator
First like received First solution authored 25 replies posted
Hi David,

Yes, you are right. With the output capacitors the current should be around 200mA. But the power dissipation of the IFX007T should not be 18V * 200mA. E.g. when we switch on the HS, the switching loss of IFX007T should be Vs/2 * Is. When the HS is on, the power dissipation of HS should be Rdson(HS)*(Is^2).

For Power dissipation tool, could you please try to update it? It was release a few days ago, so it might be only visible in the newest version. I will check with my colleague to see if we have some issue with the release as well, and will come back to you soon.

3901.attach

Thanks
0 Likes