Announcements

Live Webinar: Solving the challenges in xEV power conversions | March 7 @9am & 5pm CET. Register now!

Tip / Sign in to post questions, reply, level up, and achieve exciting badges. Know more

cross mob
yjc0101
Level 1
Level 1
5 replies posted 5 questions asked 10 sign-ins

ALL HI.

The current project plans to use two IAUS300N08S5N012T in parallel as switches to support 200A current. How can I calculate the junction temperature of IAUS300N08S5N012T? The data obtained from the manual is shown below.

yjc0101_0-1688375282286.pngyjc0101_1-1688375323946.png

Can the total thermal resistance be expressed as RthJA=RthH+RthCH+RthJC according to the method of windowing on the top of the chip?How will I get RthH, RthCH in the actual circuit design?

       The operating temperature of MOS tube is -40℃~85℃, the voltage is 24V, the current is 200A, is it possible to meet the design requirements by using two IAUS300N08S5N012T in parallel? How should I design the circuit to dissipate heat?

 

 

0 Likes
1 Solution

First of all, this connection is wrong.  Body diodes conduct current at Off condition. With a voltage drop of 0.7V and 200A, it will burn... With ORING I would also think about LinearFET.

RthCH + RthJC = 0.5°C, RthJC =0.4 (datasheet, the most important parameter), RthCH = 0.1 I used the best silicone (thin). Probably with case mounting you will use a thicker thermal foam, you take this from the material spec. I did not consider bottom path (It´s 5-10% of the top side), only top cooling.

It´s a good idea to dissipate power to the case. Probably you need thermal simulation. 24W total power is huge... 

View solution in original post

0 Likes
4 Replies
Abhilash_P
Moderator
Moderator
Moderator
50 likes received 500 replies posted 250 solutions authored

Hi,

  Thank you for posting on the Infineon community.

1. Can the total thermal resistance be expressed as RthJA=RthH+RthCH+RthJC according to the method of windowing on the top of the chip?How will I get RthH, RthCH in the actual circuit design?
Soln: Yes, your understanding is correct, the total thermal resistance from junction to ambient is described as Rthja = Rthjc + Rthja where Rthja depends on the thermal resistance of the pcb material used and the heat sink. This varies from application to application. Hence in a datasheet, the pcb material is considered before giving the Rthja value. 

Abhilash_P_0-1688378682060.png

2. The operating temperature of MOS tube is -40℃~85℃, the voltage is 24V, the current is 200A, is it possible to meet the design requirements by using two IAUS300N08S5N012T in parallel? How should I design the circuit to dissipate heat?
Soln: This can be calculated as below,

Tj = (Rthja*P)*Ta

Here the total power loss (P) and the ambient temperature is adjusted so that the junction temperature doesnt exceed 150C .

For a switched mode application, the pulse width of the operation has to be verified with the SOA and Zthja curves of the MOSFET. 

Regards,
Abhilash P

cynic_bandera
Employee
Employee
5 questions asked First question asked 100 sign-ins

Your calculation is right. To be more precise, you can add in parallel RthJA to the bottom side. RthCH depends on the thermal interface you use. At the best scenario (thermal paste) it´s about 0.05K/W. If you use the best thermal silicone it might be 0.1-0.2K/W, thermal foam (thick) is about 1-2K/W. RthH depends on the heatsink and conditions (airflow and so on). I will show you how to calculate it. 

With two MOSFETs you dissipate 24 W totally, 12 W by each MOSFET. If you use a common heatsink for both at 85°C (you mentioned) with abs max Tjmax=175, you can allow dT=Tjmax-85°C, so dT = 90°C. To dissipate the power you need the heatsink of dT/Ptot (90K/24W) = 3.75K/W. RthCH+RthJC = 0.5°C roughly, So you need at least 3.25K/W heatsink. You also have to consider Rdson degradation over temperature using the Figure Rdson vs temperature at page 5 (after calculation you have to put the correction). You should not use it at abs max junction temperature, TJ=100-120°C is OK (use this number in the formula instead of 175°C). You find a proper heatsink (from Fisher for example) with the proper thermal resistance and form factor. If you see your heatsink is huge you should think about 3-5 pcs in parallel.

If you have a system which operates in a linear mode for a short time (battery protection, hot-plug and so on) with huge power dissipated in a short time, 1-10 ms, you have to consider dynamic thermal stress by SOA at page 4. In this case LinearFET series is the right one.

0 Likes

HI.cynic_bandera.

Thank you for your reply.

Let me explain my usage. The product is a car product, the car power is 24V source is the car DC-DC and the battery parallel, need current is 200A. I need to use two groups of MOS tubes to provide two power paths for DC-DC and battery respectively, through the protection of the ideal diode controller to achieve the function of ORING. So there is no problem of hot-swapping.

Because of the miniaturization of the product, I plan to dissipate the heat from the MOS tube by directing it to the device housing and adding a thermally conductive silicone pad between the MOS tube and the device housing, which may not be easy to calculate the thermal resistance of the heat sink.How is this situation how should I choose the number of field tubes to use?

yjc0101_1-1688447605370.png

For example, how is RthCH + RthJC = 0.5°C obtained in the following response? Is this an empirical value?All the heat is transferred from the top of the MOS tube (with exposed metal surface)? Without considering the power dissipation at the bottom of the MOS tube?

yjc0101_0-1688447327594.png

As shown in the figure below, is it possible to calculate the junction temperature of MOS from a given thermal resistance?

yjc0101_0-1688454473846.png

 

0 Likes

First of all, this connection is wrong.  Body diodes conduct current at Off condition. With a voltage drop of 0.7V and 200A, it will burn... With ORING I would also think about LinearFET.

RthCH + RthJC = 0.5°C, RthJC =0.4 (datasheet, the most important parameter), RthCH = 0.1 I used the best silicone (thin). Probably with case mounting you will use a thicker thermal foam, you take this from the material spec. I did not consider bottom path (It´s 5-10% of the top side), only top cooling.

It´s a good idea to dissipate power to the case. Probably you need thermal simulation. 24W total power is huge... 

0 Likes