0
\$\begingroup\$

I've this TO-247 MOSFET http://www.datasheet.es/PDF/203551/2SK3271-01-pdf.html

how much current will support this case? I've to control a 12V 80A DC Motor

\$\endgroup\$
4
  • 2
    \$\begingroup\$ What does the datasheet say? \$\endgroup\$
    – Hearth
    Apr 15 at 14:53
  • \$\begingroup\$ that supports up to 100A but does not say anything about the case. I saw other MOSFETs with case TO220 that claim to support up to 150A but the case only supports 70A \$\endgroup\$ Apr 15 at 14:55
  • \$\begingroup\$ The limitation will likely be thermal, so you need to consider maximum ambient temperature, maximum Rds (on) vs temperature and your heat sink performance. I don’t see a “case” limitation as on many TO-220 packages, but it’s quite possible you’ll find a limit like 75A in another document related to the package . \$\endgroup\$ Apr 15 at 14:56
  • \$\begingroup\$ @SpehroPefhany take a look infineon.com/dgdl/… at the page 2 in the point 6 (at the end of page) "Calculated continuous current based on maximum allowable junction temperature. Package limitation current is 75A." ... \$\endgroup\$ Apr 15 at 15:00
1
\$\begingroup\$

It has around 6 mOhm RdsOn so it will dissipate 38W at 80A. This is doable with TO-247 and a proper heatsink.

However the heat sink will be large and expensive. If you put the heatsink money in more MOSFETs instead, and wire them in parallel, it will be cheaper and more efficient. Say you use 3x 3mOhm MOSFETs, each taking 80/3=26A, then each FET will dissipate only 2W, total 6W, you need a much smaller heatsink. And you can also use cheaper TO-220 or even SMD MOSFETs.

\$\endgroup\$
4
  • \$\begingroup\$ cant wire mosfet in parallel if using PWM \$\endgroup\$ Apr 15 at 16:04
  • \$\begingroup\$ That's wrong, it is often done. \$\endgroup\$
    – bobflux
    Apr 15 at 21:29
  • \$\begingroup\$ Low rds mosfets usually have high gate capacitance/charge requirements. Paralleling them obviously makes it even larger usually necessitating a gate driver. \$\endgroup\$
    – Kartman
    Apr 15 at 23:17
  • \$\begingroup\$ Yes, driver is much cheaper than heatsink for 38W... \$\endgroup\$
    – bobflux
    Apr 15 at 23:49
1
\$\begingroup\$

The SOA curve on the datasheet implies it's okay:

enter image description here

I share your suspicion, but it's quite possible that Fuji's EIAJ SC-65 package is a bit better than Infineon's similar package. There are a number of TO-247 packages with limits in the 120A-195A range (as well as some with less than 100A, as you know), so it's quite plausible, and Fuji has a lot of background in power electronics.

See, for example, this IR selection guide's notes. Some of their TO-247 packaged products are limited to 90A by the packages, others to 195A.

The JEDEC/EIAJ standards only cover physical dimensions so you can't really assume much between different products let alone different manufacturers.

If your motor draws 80A continuous, the MOSFET will likely explode or at least fail shorted when you try to start the motor, but you probably know that already. I did a little one-off project where we ended up paralleling two (honest) 300A MOSFETs to get adequate margin.

\$\endgroup\$
2
  • \$\begingroup\$ You're definitely going to need a heatsink for that much current, of course. The package alone would fry instantaneously. \$\endgroup\$
    – Hearth
    Apr 15 at 15:52
  • \$\begingroup\$ As Hearth says, and don't forget that Rds(on) has a strong positive tempco so the power dissipation when hot might be 50% or more higher than the datasheet number at 25°C. \$\endgroup\$ Apr 15 at 15:56
0
\$\begingroup\$

According to the datasheet you referenced:

enter image description here

\$\endgroup\$
1
  • \$\begingroup\$ that supports up to 100A but does not say anything about the case. I saw other MOSFETs with case TO220 that claim to support up to 150A but the case only supports 75A, ie. take a look infineon.com/dgdl/… at the page 2 in the point 6 (at the end of page) "Calculated continuous current based on maximum allowable junction temperature. Package limitation current is 75A." \$\endgroup\$ Apr 15 at 15:01
0
\$\begingroup\$

The max current is limited by the power dissipation (switching and resistive losses) and the size of the die, not the case itself. Note that max current figures spec'd on data sheets are typically assuming only resistive losses and are for case temperature 25 deg C, which generally is not realistic.

You need to ensure the junction temperature of the MOSFET stays below the maximum stated on the data sheet. So, estimate the power loss in the device, then, given the junction-to-case thermal resistance and expected heatsink temperature, calculate the junction temperature. If it's too high, you have to choose a bigger MOSFET.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.