2

I ordered some MOSFETS from AliExpress and I think I've been duped.

The diagram below shows my setup:

enter image description here

If I'm not mistake, this means that the R_DS(On) of this N-channel MOSFET is 0.042/0.62 = 67 milliohms, at a V_GS of 3.83V.

The datasheet for this AO3416 says that even at a V_GS of 2.5V the R_DS(On) should be less than 26 milliohms. The MOSFET doesn't warm up, so its temperature is near ambient 25C, like the datasheet.

Have I been duped, or am I measuring R_DS(On) incorrectly?

edit 1: At V_GS of 4.83V under a new load I measure 0.58A and 0.030V voltage drop, so RDS_On is 51.7mOhms, still higher than what it should be at <22mOhms...

edit 2: At V_GS of 3.96V under another load I measure 0.28A and 0.0165V voltage drop, so RDS_On is 58.9mOhms...

I think it's fair to say that the product does not meet the specifications?

UnknownDagger
  • 143
  • 7
  • The setup looks good. However, are you sure that the voltage drop you are seeing is purely due to the MOSFET and not due to parasitic resistances due to board traces? – sai Jun 05 '23 at 15:12
  • @sai the board traces are extremely short, it is simply a SOT-23 MOSFET on a breakout board – UnknownDagger Jun 05 '23 at 15:18
  • 1
    The datasheet characterizes Rds_on at maximum 26 milliohms for a drain current of 5.5A and ambient temperature of 25C - check the table on page 2. Your conditions are different. – Adam Lawrence Jun 05 '23 at 15:50
  • @AdamLawrence does RDS_On decrease at high drain currents? – UnknownDagger Jun 05 '23 at 17:18
  • Are you measuring across the device pins on the breakout board? Or elsewhere in the circuit? – Tim Williams Jun 05 '23 at 18:25
  • It doesn't take much thin trace on a breakout board to add up to 30mΩ. Measure right at the SOT23 package pins with multimeter probes or with thin wires attached there. There are other effects, but probably not as material. – Spehro Pefhany Jun 06 '23 at 02:32
  • It's generally not recommended to buy parts from aliexpress, ebay, amazon, etc; you should look to well-recognised and authorized stocking distributors like digikey, mouser, farnell, etc. – Hearth Jun 06 '23 at 02:33

2 Answers2

2

Have I been duped, or am I measuring R_DS(On) incorrectly?

The method is correct, although RDS-on may or may not include the lead resistances but let's ignore them for now as I don't think the SOT-23-3 lead resistances can be as high as 10 mΩ for less then 1 Amp of current. Also the RDS-on may change with drain current, especially when the gate-source voltage is closer to the threshold voltage, VGS-th. Judging from the characteristic curves given on p.3 (e.g. Fig.3) I can say this pretty much not the case in your setup, so we can ignore this as well.

I personally prefer switching instead of applying continuous DC, just to prevent the junction temperature rising because it has significant effect on RDS-on. The speed and sensitivity of the equipment becomes important here, also it may require using a data logging capable equipment.

We can expect some deviation from the datasheet values as the conditions do not perfectly match. But this deviation is not normally +100%.

Anyway... You should think twice before buying electronic components from unknown sources. And you've learnt this in a harder way.

Rohat Kılıç
  • 26,954
  • 3
  • 25
  • 67
0

With these numbers I would question the test equipment and the actual connections. The measurement for voltage DS should be at the leads of the package, not the PCB. Also are you using at least a 5.5 digit meter that is in calibration?

When doing this for qualification we placed them in a temperature controlled oil bath and let them stabilize for about 1/2 hour. Measurements were pulsed, micro seconds if my memory is correct. The current source was extremely accurate good to at least 0.0001 mA. I also used a current meter in series with it. All metering connections were kelvin.

Gil
  • 4,951
  • 2
  • 12
  • 14