Controlling 3V3 relay from 2V8 (VL53L1X) GPIO output with PNP transistor

Question

I need high side switch for 3V relay (exactly RM40-3021-85-1003 with 45ohm coil -> need ~65mA). I have multiple VL53L1X (ST's ToF distance sensor, 2V8 logic levels) connected with theirs OC GPIO outputs ("INT") to one common wire. Normally the common wire high (2V8) all the time, but when any of the sensors trigger (something comes closer than distance limit) - the common wire logic level goes from 2V8 down to 0V (VL53L1X open collector outputs are active low, so, this is why I can't use low side switch). Is it possible to solve this by only one PNP? (+resistor+flyback) And not using aditional NPN as suggested in question here... The GPIO output of the sensor has level 2V8 (as sensor is powered by 2V8 LDO on PCB, can't be changed to 3V3 in this situation). But the 3V relay (emitor of PNP) is powered by 3V3 source (no 2V8, nor 5V supply possible). Its only 0,5V difference, is it secure to do it simply like this circuit: or by adding a diode to base to prevent "highside switch fail"?

Please can you also give me exact values of all parts (resistors, PNP number, diode 1N4148?) as I need something very reliable for many years of lifetime and yet simple (I thought the more parts the more to fail during time), as I'm good in ESP and PIC (SW+HW), but not that good in bipolar transistor simulation... (The result device must be very secure to stop moving parts before a human may be hurt. It's HW backup solution if the SW stop fails.) Thanks.

Answer 1

You can "offset" the 2.8V digital control signal upwards by about half a volt, with a single diode, D1 here:

^{simulate this circuit – Schematic created using CircuitLab}

D1 develops half a volt or more when it passes any current, a potential difference which is added to \$V_{IN}\$. X is therefore half a volt or so higher than \$V_{IN}\$.

When \$V_{IN} = 2.8V\$, you can expect X to rise close to 3.3V, easily enough to turn Q1 off.

Answer 2

No - That's not going to be 'guaranteed' to work always. The problem is your logic control signal can only go to max 2.8V, but the VCC is 3.3V. Like you say, that's "only 0.5V" and "in theory" the transistor won't turn on until there's 0.7V difference there. But with such a small margin, you can't be assured that always the transistor would be completely off. They really don't have a hard-fixed cutoff voltage like that - It's more of a range. That's why there are extra transistors in the other referenced link. They allow for the base of the PNP transistor to get up to the same voltage as "VCC", thus ensuring that transistor is off.

Understand your concern about 'more parts = higher likelyhood of failure". That's not really my experience.... IMO it's more like "The harder you drive the parts, the more likely to fail". If you're being very gentle with the parts - operating them very far from their maximum voltage, current & power limits - resistors and transistors will last pretty much forever. Relays are another story.... Do you really need to have a relay - i.e. you really need the electrical isolation??? Relays are mechanical devices and as such, they definitely will eventually fail.