VN751S Switch Off State and Short Circuit Detection

[Okeskin]

Hi,

I will use VN751S. However, I have a problem. I use feedback to detect open load condition. Use of voltage divider resistors and vn751 are attached.

The problem is; For 24v, the result of the voltage division should be about 19V, but when you create the circuit with VN751S (IN = 0V switch off ), the voltage becomes 13V. My expectation is that the switch has a very high equivalent resistance in the off position and it will not affect my feedback circuit. What is the reason of this?

Another problem is short circuit protection. While the VN751S output is shorted to ground, the output current will never be 0. It allows about 200 mA of current to flow from the output to ground. Shouldn't the output be turned off completely in a short circuit to ground (Current > 6A)?

 

These tests and their results are very important. Since I will decide accordingly, I need to find an answer immediately.

BR,

Onur.

[Domenico RAGONESE]

@Okeskin​ 

I don't see your computation about the open load detection:

independently by VN751S the R30 should be much higher if you expect Vfb.io17 = 19V at Vcc.bat = 24V. .

However, I can see ~350uA flowing into VN751S when Vcc.bat =24V and R22 = 27k.

So, some pull down circuit (~40k) is also affecting your measure: you've to take care about it.

About the short circuit behavior, maybe you're talking of 200mA average current measured by a multimeter?

In fact, the VN751S implements a current limitation circuit working as follows:

the IC maintains ON the output switch limiting the output current to ILIM until the Thermal protection is triggered.

Then the IC switches off the high side switch until the internal Tsense goes below the reset value.

After that it retries to supply again the load.

This is repeated until the short on the load is removed or IN goes to 0.

Hope it helps.

[Okeskin]

Hi Domenico,

I think it was misunderstood. I'm not talking about Fb.IO17. The place I'm talking about is exactly the juntion point of R22 and R26, that is, the output of VN751S. Actually 19V occurs at this point when 24v is applied.

That is, there is an equivalent resistance of 105Kohm parallel to the NV751 output. I could not see an explanation in the document about the pull down (~40k) you mentioned. Are you telling based on my measurement result?

I will test the current issue in detail.

BR

[Domenico RAGONESE]

@Okeskin​ 

We did a quick check on our side and measured around 40kΩ.

We're waiting for the feedback from designers about the internal structure between OUT and GND.

I'll keep you posted and according to the feedback we'll consider the possibility of a datasheet update.

BR,

Domenico.

[Okeskin]

Hi @Domenico RAGONESE​ 

For a high side switch, I would expect the impedance at the off position to be higher.

I will be waiting for your return. It is very critical for me.

BR,

Onur

[Domenico RAGONESE · Accepted solution]

Hallo @Okeskin​ .

Have you ever considered IPS160H (or IPS160HF) for your design?

These part numbers integrate the wire break detection feature: you just need to set an external resistor, while the voltage comparator is integrated.

Also, the pin-to-pin compatibility with IPS161H (and IPS161HF) could allow you to design an equivalent platform for 500mA applications, not only for 2A ones.

Going back to VN751S, it is a product launched on the market many years ago, designed with the available technologies at that time.

Some pull-down effects between OUT and GND remains, although some design tricks have been implemented to limit the parasitic effects of the technology.

Hope it helps.

Best regards,

Domenico.

[Okeskin]

Hi @Domenico RAGONESE​ ,

The products you recommend look good. But unfortunately I need a pin to pin product.

Thanks for your help.

[Domenico RAGONESE]

Hallo @Okeskin​ ,

apologizes for the late feedback.

We spent some time in the characterization of the characteristic Iout vs. Vout between OUT and GND pins of VN751S.

As you can see from the first sheet of the attached excel file, this is not linear.

For you "open load detection" resistor network we suggest to set it such that the detection threshold at OUT pin remains ~3V.

The activation threshold of the comparator can be fine tuned by splitting in two resistors the pull down resistor of the detection network (see simplified block diagram in the second sheet of the excel file).

Also, we did an exercise trying with a resistor setting and verifying what happens at Vcc =24V (+/-20%): see attached excel.

Finally, in short term (< 3months) the datasheet of VN751S will be updated wit the relevant information to drive the open load / off state resistor network.

Hope it can help.

Best regards,

Domenico.