cancel
Showing results for 
Search instead for 
Did you mean: 

Viper UVP when configured as high-side buck

engineer55
Associate

Hello,

We are using Viper31 to step down 250VDC to 12VDC in a buck configuration as shown in Figure 42 of the datasheet:

engineer55_0-1757433775850.png

This is also the same topology used in STEVAL-VP319X1B:

engineer55_1-1757433815520.png

We are working with STEVAL-VP319X1B to try to implement the UVP function with a UVP threshold of 230V. However, we have been unsuccessful in our efforts. We currently have the RH and RL voltage divider connected to Vin, UVP, and Viper31 GND. However, Viper31 GND is different from the circuit GND, because this is a high-side buck topology. We suspect this is why we have been unsuccessful. We have RH = 10M and RL = 20k. With this setup, when Vin = 150V, we see the chip draws up to 2mA of current and back down to near zero current and then back up to 2mA over and over continuously. It looks like it is attempting to turn on every 30ms, indicating perhaps that we are triggering the UVP 30ms timers.

We suspect the issue is that Viper31 GND is different from circuit GND, which makes the UVP implementation more complex. Is it possible to implement a UVP circuit on Viper31 when it is configured as a high-side buck?

Thank you in advance for your support.

3 REPLIES 3
AScha.3
Super User

Hi,

as the V+ has to be constant > "+200V" , a cap is needed; maybe try this :

AScha3_1-1757437582531.png

 

rh+ rl could be 10M + 20k , cap 10nF (400V) , diode 1N4007 or similar.

If you feel a post has answered your question, please click "Accept as Solution".

Thank you, can you explain the purpose of the diode? Why is it necessary and what is it doing?

 

And what purpose does the capacitor serve?

We will implement and let you know what we observe.

Thanks!

Ok....diode and cap keep the loaded actual V+ , also for the time, the switcher(viper) is ON, and the voltage drops to

~ zero ; the UVLO would trigger...but now the 10nF holds the voltage at about V+ level. 

If V+ drops , cap is loaded to actual voltage and so, if < 200V , will have not enough voltage to prevent UVLO taking its job, --- stop/off. The cap just gives a average delay of about 300ms , then its going off.

(If i calculate correct, in my head. :)   )

btw.

I have no viper chip, just did some repairs , with exploded chips....common problem on some washing machines ect.

If you feel a post has answered your question, please click "Accept as Solution".