TIDM-02009: Current Sense Circuit Questions

Hello Gokhan,

I will take a look at your questions and get back to you tomorrow. 

Gokhan,

Gokhan G. said:

1) I think whole circuit for DCDC side is as below and Ipri is connected to ADC pin in order to control DC link current for peak current mode. I think CT_N and CT_P is floating in normal operation. Could you please confirm?

Not exactly. The 3.6kW DC-DC converter module can accept a C2000 controlCARD directly. In that case, CT_N and CT_P would be unconnected, and Ipri would be used by the C2000 device on the DC-DC converter board as the PCMC signal.

However, note that there is also mother board connector which accepts CT_N and CT_P. These signals connect to the mother board (see EVT_001E2(001)_MotherBoard_Sch.PDF, sheet 5) and these signals are routed to the exact same circuit (but on the mother board), which are then fed to a C2K_ADCIN14 net. This is the CT_OUT measured on the TIDM-02009 design guide (don't ask me why a different name is used! ). 

The mother board also accepts a C2000 controlCARD. In this configuration you can control the DC/DC stage plus traction inverter stage using a single C2000 device (the whole value prop for TIDM-02009).

Gokhan G. said:

2) How does this circuit achieve positive signal during reverse mode of operation. I think the output of OPAMP will be zero.

There is a nice video on this type of sensing here. Not exactly the same, but close.

https://www.ti.com/video/6130991705001

Basically you bias the op-amp to some reference voltage to allow the output of the op-amp to vary between some min voltage at min negative current and a max voltage at max positive current.

Gokhan G. said:

3) What is purpose of using R66? How do you decide R66 as 22kohm?

R66 and R73 will set a reference voltage. I believe here the designer did not want to pick exactly mid range since the desire was to measure higher positive currents vs negative currents. I'd need to look through the firmware to see if it lists min and max current sensing values.

Gokhan G. said:

4) Why did you choose R73 as 1.05k and not 1k? How do you decide R73 as 1.05kohm?

Not sure, would need to track down the original designer for this circuit. However, if you are interested in this type of sensing, would be better to turn this conversation over to our high-speed amplifier team. TI also has current sense amplifiers which can be used.

Hi Gus,

I think that the current sense circuit in TIDM-02009 cannot achieve reverse current sensing. Because, it has resetting diode (D2) in PSFB mode. That diode will block current during reverse mode of operation. Therefore, I think this circuit will only add offset reference to sensing in PSFB mode.

Could you please confirm?

Thanks in advance.

Hi Gokhan,

You are right in the case of CT1/CT2 being used. However, for CT_N/CT_P this is not the case. 

Hi,

OK, I won't use CT1 and CT2 pins for bidirectional operation. However, I couldn't see where the real current transformer which has CT_N and CT_P pins. I need to see and review schematic circuit diagram for it to check I can put it to my board for my bidirectional operation.

Thanks.

Gokhan,

I was able to get more information on TIDM-02009 from the original designer. 

1. You are correct about U1/D2. The purpose of U1 is to measure current but only in the forward direction (HV to LV). U1 is not setup to measure current in the reverse direction. There is an offset added to the non-inverting input of U1, but this is only to get better linearity on that op-amp. D1 is added to rectify the input current such that U1 does not see negative current as the PSFB legs switch.

2. For reverse direction you'd want to measure current on the LV side. For that there is U9 (INA240A) on the board which can measure negative current. You can see there is an offset provided to the INA to allow for positive and negative current flow. The INA will add delay so implementing PCMC will not be feasible. However, the thought is that you could do average current mode control instead to charge the HV side. 

Hi Gus,

Thank you for your effort. I got some points.

1) In TIDM-02009, why is there need to add additional OPAMP circuit to the rectified signal. It can be provided directly to the ADC pin. Is it done to protect ADC from negative signal during reverse mode of operation? However, in this case, OPAMP input will face negative voltage on its pins.

2) I couldn't see where the real current transformer which has CT_N and CT_P pins. Is there really that current transformer on the power path?

Thanks in advance.