| Electronics > Projects, Designs, and Technical Stuff |
| MOSFETs and Miller plateau |
| (1/4) > >> |
| TreeOone:
Hello! I'm looking for a suitable MOSFET which could be used to turn ON and OFF coil of electro-mechanical relay (rated current 100mA). I am intending to use +3.3V from MCU's output for charging and discharging MOSFET's gate. I have following doubts: MOSFETs have Miller plateau, which should be completely charged, before Vds can drop to lowest values. If in Vgs-Gate_charge diagram Miller plateau is not reached, Vds stays high. I was happy with this understanding until I stumbled upon following datasheet: https://hr.mouser.com/datasheet/2/308/NTD3055L104-D-114990.pdf Now, by looking at following datasheet, two contradictory informations are presented: 1) On page 3 there is "Figure 1. On−Region Characteristics", which clearly states that you can drive this MOSFET with 3V. In fact it says you can also drive my electro-mechanical relay's coil. When gate voltage is at 3V, I should expect Vds to be less than 500mV @ Id = 100mA. That's great, I found a solution. But....there is also.... 2) On page 5 there is " Figure 8. Gate−To−Source and Drain−To−SourceVoltage versus Total Charge" which states this MOSFET's gate will not even reach Miller plateau with 3V, i.e. Vds will be high, you need at least Vgs = 5V. Crap....this will not work :/ Now, could somebody explain how previously mentioned diagrams can exist in the same datasheet?????? I also have this MOSFET: https://hr.mouser.com/datasheet/2/268/TN0604%20D080813-965142.pdf This one has Miller plateau at 1V, but also smaller Id = 0.5A, for Vgs = 3V. Which MOSFET should be used? |
| Zero999:
One question: why not use a BJT? It's cheaper and far easier to get parts which will do this. The plain old BC338 can do this with minimal base drive. A forced Beta of 1/100, so that's IB = 1mA will give a VCE<1V (probably around 200mV), when IC = 100mA. http://www.mouser.com/ds/2/149/BC337-193546.pdf To answer the original question, either MOSFET will do, but they're both way overkill and way more costly, than a simple BJT. If it must be a MOSFET, then go with something like the FDN337N, which is specifically designed for 3.3V logic level operation. In any case it will still be more expensive, than a BJT. https://www.onsemi.com/pub/Collateral/FDN337N-D.PDF |
| MagicSmoker:
^this is the correct answer. A MOSFET is the wrong switch for the job; use a BJT instead. If you are really lazy, use a "pre-biased" BJT which already has the base and base to emitter resistors inside such as MMUN2231 or the like. |
| 2N3055:
NUD3124 |
| TreeOone:
Ok, you convinced me that BJT is also acceptable option. By looking at the datasheet: https://www.onsemi.com/pub/Collateral/BC337-D.PDF ....if I want Ic = 100-200mA (added little redundancy), diagram "Figure 3. DC Current Gain" says I need 1mA in the base for Ic = 200mA. (h = 200 @ Ic = 200mA) Now the question is how to feed that current to base, i.e. how big should be base resistor. Table "ON CHARACTERISTICS" says Base−Emitter On Voltage = 1.2V @ (Ic = 300 mA, Vce = 1.0 V) ....so voltage drop on base resistor should equal 3.3V – 1.2V = 2.1V ...and resistance equals 2.1V / 1mA = 2100 Ohms In my case this resistor should insure Vce < 1.0 V @ Ic = 100-200 mA Is this calculation correct? And finally, would please somebody give answer to my initial question regarding Miller plateau? |
| Navigation |
| Message Index |
| Next page |