TIDM-1000: One phase current is distorted when PWM turned on (Build-1)

Hi Naim,

I recommend using 500 ohm as this is the first time you are running the board. Please run at similar conditions as stated in design design and validate the hardware is good enough. From the first image, in without PWM case, your guiIrms1/2/3 are NOT closer while guiVrms1/2/3 is good, this needs to be debugged first. 

Then move to with PWM scenario. Below can be tried on the setup: 

1. Please double check the current sensing circuitry if any changes been made?

2. Swaping your AC signal (RYB to RBY) to check if the distortion is moved onto different phase?

3. Can you confirm you didnt connect AC neutral to VBUS mid point?

4. are you using the 3phase 3 wire system here or not?

Let me know your observations to debug further.

Best Regards,

Uttam Reddy

Uttam,

Thanks for your response. I have changed the load resistance to 500 ohms as per your recommendation. Here is the answer to your querries:

1. I could not see any change to the current sensing circuit. 

2. Without doing the PWM, here is the SS of the debug window:

After swapping Phase-1 and Phase-3 (respect to initial condition):

After swapping Phase-1 and Phase-2 (respect to initial condition):

With PWM at this point (Phase-1 and Phase-2 swapped):

It seems the current balancing improves with swapping the input lines without PWM. However, distortion persists with I3 if the PWM turned on. 

3. I connected AC neutral to VBUS mid-point following HW setup guideline as Figure 16.

4. I am using output from a VARIAC, which is 3-phase 4 wire system.

Please let me know your thoughts. 

Thanks,

Naim

Uttam, 

In addition to my previous inputs, I have done some measurements output of the Hall sensors (TP_IL1,TP_IL2,TP_IL3 to TP_GND3). I measured these points with a low voltage differential probe. Lower waveforms are indicating input currents (Y, G, B). Red is the Test Point voltage. The PWM is ON during these instances. 

I also measure with different three differential probes (high voltage) simultaneously. But had to do some low pass filtering as these were getting noisy. I swapped 1^3. The noise was shifting (R to Y). 

I hope it would give you better insights. Please let me know your thoughts. 

Thanks,

Naim

Hi Uttam,

Thanks for your response. I do not think it is causing from the VARIAC. Here is the voltage with no PWM. 

I cleared the trip (=1) and could see the PWM signals for related tests. This is Build-1. Don't you think the input current is matching with the manual?

Besides, I tested the output voltage from Hall sensors. Here are my test results:

I can see significant noise from the output of the opamp. Here are my test results:

Could you please explain why I am having this much noise at U22_Pin14?

Thanks,
Naim

Hi Naim,

Variac is clean.

Looks like U22 hs gone bad, can you please swap the U22 component and redo this test? I assume you are using the same firmware from DPSDK without any changes?

Thanks & Regards,

Uttam

Hi Uttam,

Thanks for your response. I have not changed the original firmware. Just following the instructions through.

I can try replacing the U22. But would it make any issue if the board finally needs to be returned for a replacement?

Thanks,

Naim

Hi Naim,

Understood. At this point, I think its best to replace the op amp since the other 2 phases are good. I still think it will solve your problem. 

we can think of alternative later, either me/ Matthew will take care of this. 

Thanks & Regards,

Uttam

Hi Uttam,

Thanks for your confirmation. It would be very helpful. I will keep you posted. 

Naim