This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

THVD8000: THVD8000 Immunity problem

Part Number: THVD8000

Hello,

We have a problem during EMC test : Conducted radio frequency (IEC 61000-4-6). The severity of the test are : 

Frequency range: 150 kHz – 80 MHz
Amplitude: 3 V r.m.s. h
Modulation***: 80 % AM at 1 000 Hz
Frequency sweep range:
≤ 1,5 × 10–3decades/s (or 1 %/3 sec.)

After that, the test levels shall be increased to 10 V r.m.s. for spot frequencies in accordance with IEC 60945 at 2 ; 3 ; 4 ; 6,2 ; 8,2 ; 12,6 ; 16,5 ; 18,8 ; 22 ; 25 MHz.

We are using a THVD8000 with a 1 MHz carrier and a baud rate of 115200 BAUDS. The schematic is has below

 

THVD8000.png

 

 

We have replace C520 and C521 by 100pF value and we have replace the R512 resistor by two 60 resistors with 220pF capacity between the two resistor and 0V. We have replace the self by 744227 (mor efficient as 744252220)

Our electronic is connected on the another electronic with the same schematic. the cable is shielded and the cable shield is connected to the guard ring on our board. 

When the perturbation is launched (especially during the spots with 10 V r.m.s) we have some communication FAULT with the THVD8000. We can observe that, the LED 501 stay light during all the perturbation.

But if the shield câble is not connect on our ring no problem appears => So we suppose the HF pertubation pass by the ring and small capacity (47pF) place between the ring and our 0V and the noise pass in the THVD by 0V.

 

Did you have some recommandation about that ? Place ferrite between 0V and 0V pin of the THVD8000 ? Do not connect the shield on the ring ?

Best Regards

Malo

  • Hi Malo,

    It seems like you've isolated the issue to the guard ring on your board, rather than the THVD8000 itself. 

    If you can get away with not connecting the shield cable to the ring and still pass emissions, that would be the simplest answer. A ferrite bead between ground and the THVD8000 could also work, though I would probably try to focus on keeping the signal from passing between the ring and ground plane to begin with. 

    If you are able to send over any layout files/information I might be able to help out a bit more.

    -Ethan White