Hello everyone!
Our company is developing new energy automotive oil pumps with the following parameters
Voltage: DC, 280V~435V
Peak power 2000W
Speed 6000 RPM
Communication: CAN
Temperature Range - 40 ~140
Hall-free sensor,square-wave drive
I would like to ask the following questions:
1、According to the data of NXP official network "AN11967, S12ZVM Family in 48 V Applications", referring to the application of 48V system, HsX and HD signals are processed by 25:1 voltage divider and connected to the chip. Is this method feasible?
2、Whether NXP has other complete solutions for 400V system brushless DC motor drive of new energy vehicle, by the way, the temperature requirement is 140 C.
Do you have any experts to answer this question? Thank you.
Hi,
I would not recommend this for your project. Please use S32K1x instead. Even S12ZVM @ 24VDC or higher doesn't make much sense. S32K1x has better performance, and it is just starting its life cycle. SW support is here as well.
BR,
Matej
Hi,
Thank you for recommending S32K1x,But I have a question about it. According to the description of S32K1xx Data Sheet,
– Ambient temperature range: -40 °C to 105 °C for HSRUN mode, -40 °C to 125 °C for RUN mode
My product is used in engine bay ambient temperature requirement 140°C. If S32K1xx is used in this environment, will there be hidden dangers?
Hello,
unfortunately, S32K1 is not Grade0 complient. Getting back to the S12ZVM device, have you considered to enable the PWM outputs on pins? This way, you would be able to leave the GDU off or in a limited operation, having the power consumption better and reducing the BOM of unnecessary GDU-supporting parts. The only drawback is that you would have more pins occupied for the PWM.
Hello,
I'm not aware of any MagniV solution in 400V applications.
However, AN5082 MagniV in 24 V Applications shows a solution using only the S12ZVM PWM outputs without GDU and thus without the dividers in 24V applications.
Regards,
Daniel
Hi,
Thank you for your reply. In document "AN11967, S12ZVM Family in 48 V Applications",He also talked about the way you described it. But I think there is something wrong with the description in Chapter 5.2.
(1) I think CP VCP must be connected or HGx will not be generated.
(2)The HGx signal after frequency division should be a floating level, which can not drive the external pre-driver directly.
Hi,
I don't think there is a problem.
The charge pump generates voltage > V_HD and the AN says in that chapter:
"Most of the external pre-drivers accept an input (logical) voltage from 2.5 V to 10 V."
Regards,
Daniel
Hi,
Thank you for your attention. Maybe I misunderstood AN.
According to AN description:
“In the case of the HGx pin where the output voltage is around 20 V, the voltage going into the pre-driver’s inputs will be around5 V. For LGx pins the output voltage is 12 V so the voltage going into pre-driver inputs would be 3 V.”
If HGx tries to generate about 20V voltage, I think bootstrap capacitors must be connected (not shown in the figure),
Or using charge pump circuits, VCPCP should be used。That's why I thought the charge pump circuit would be used.
I also have a question about the possibility of using a charge pump circuit instead of a bootstrap capacitor so that I can work at 100% PWM.
Thanks very much。
xumax
Hi,
Thank you for your reply. I think CP VCP must be connected, But the CP VCP in AN is disconnected
Hi,
Why do you think the charge pump is needed?
Normally, the charge pump generates higher voltage for the HS MOSFETs.
But in this case, the external driver accepts only an input (logical) voltage 2.5 V to 10 V as the AN describes.
What external driver are you going to use?
Thanks,
Daniel