Unclear voltage regulation circuit

[bnz99]

Hi,

I'm a beginner and trying to analyze a PCB of an purely digital extension card for an older electronic instrument from the early 2000s (containing two flash nors). There is a part of a circuit which I'm curious about what it is doing exactly and why. The context is that the mainboard, where this extension card is plugged on, can deliver either 3.3V or 5V through it's connector. It depends on which exact model the mainboard is. The 3.3V or 5V are not provided through different pins, but always through the same VCC pin which I have found out by measuring when the extension card was plugged on those different mainboards.

The circuit looks like this:



VCC (either 5V or 3.3V is coming from the connector pins at the top. It goes directly into the right pin of the SOT-23 Q2 IC or through a 220 ohm resistor to the left pin of the Q2 IC. The marking on D1 cannot be read properly. There is a bar though at side side pointing to Q2. The other end of D1 is connected to ground. I'm thinking D1 might be a 3.3V zener diode that clamps the 5V to 3.3V or otherwise let the 3.3V pass through. However, I have measured the voltage (using a 5V mainboard) and pin 1 of Q2 measured 4.2V, if this was a zener diode, I would have expected the measure about 3.3V. What I don't get is the part with Q2 and Q1. Both are marked as "DD" which doesn't help much. The Q markings may indicate a transistor or mosfet, but I'm not sure what kind of transistor circuit this would be. Typical pinouts I have found would be Pin 1 (left pin) as the base, Pin 2 (right pin) to be the emitter and Pin 3 (single pin on the other side) to be the collector. I'm puzzled though as no matter whether it is an NPN or PNP channel resistor and no matter whether it is switched, the voltage would still flow between emitter and collector as there is continuity between the right pin of Q2 to the right pin of Q1 (at the one end of R3). That would probably make the whole circuit pointless, wouldn't it? I have found that there are also sot-23 zener diodes with a DD marking, but that didn't make much sense to me either. R4 is a 470 ohm resistor connected to ground.

I would be very glad if someone with more knowledge about circuit design could give me a clue what this is doing and possibly why. Thanks a lot!

[bnz99]

I have added the voltages to the pins. Maybe this helps.

[magic]

2.4V zeners exist.

If these are P-ch MOSFET, with 3.3V input the 2.4V at D1 may be insufficient to turn on Q2, allowing R4 to turn on Q1 which passes 3.3V to the output. With 5V input Q2 turns on, Q1 turns off and 3.3V at Q1 drain must be coming from somewhere else... maybe another pin?

[inse]

I was thinking of PNP transistors, for MOSFETS, the voltage threshold might be inappropriate, unless its low threshold types.
The circuit looks like in the attached picture.

[magic]

MOSFET has only one diode between S and D when G=S.
BJT always has diodes between E-B and C-B.

It's easy to tell them apart with a DMM.

My bet is they are MOSFET. 1.5V threshold is not impossible for a logic level type. The series connection of BE junction and zener to ground that you suggested would be quite dangerous with a PNP.

[bnz99]

Thanks! These are great hints that I can follow up on and it's starting to make sense to me. Next thing I do is try the board in a 3.3v mainboard and measure the voltages. After that, I'll desolder one of the Q ICs and try to find out whether these are mosfets or bjt. I in fact forgot to mention an ldo on the other side of the board. The holes on the output of Q1 are connected to the LDO output. That must be where the 3.3v are coming from.

[magic]

Yep, this circuit appears to bypass the LDO when on 3.3V motherboards and that's it.

No need to desolder those transistors, measure them in circuit with power off. MOSFETs are quite easily damaged by overvoltage or ESD so I wouldn't mess with them without need.

[inse]

Maybe it’s this one?
https://www.alldatasheet.com/datasheet-pdf/pdf/101036/TOSHIBA/SSM3J02F.html