TPA3255: Temperature calculation

Hi Pawel

Could you please share how you calculate? 

Yes, I took the formula from this document: https://www.ti.com/lit/an/spra953c/spra953c.pdf?ts=1676555124239 I used the equation 5 and to calculate Rcs I used the equation 7. The other variables which I know are:

Rsa=1,5°C/W

Ta=50°C (the maximum acceptable temperature)

Rjc=0,3°C/W

P=35,2W

Like I said previously, I don't know the equation which include the flow of air from fan, which I will use. 

From equation 7 I've got: 

T=2mils=0,0000508m

k=0,88W/mK

A=0,000026m^2

So:

Rcs=T/(k*A)=2,22°C/W

Then from equation 5:

Tj=Ta+((Rjc+Rcs+Rsa)*P)

Tj=50+((0,3+2,22+1,5)*35,2)

Tj=191,5°C

Thank You for the answer. What do You need to do this flotherm simulation? 

And about estimating the final temperature. In the previous project I used TAS5630BDKD and I want to change it for TPA3255. The cases are different, so the thermal resistance is different too. I substituted the variables from the TAS datasheet, to equation, which I wrote in previous post and it gave me 155°C.

That's why I'm afraid that TPA3255 won't handle the 2x180W output power, for 1 minute. 

Hi Pawel

Pawel Berth said:

What do You need to do this flotherm simulation? 

I mean we could provide the thermal model of our device, and customer could do simulation based on their system. Officially, TI will only simulate under JEDEC standard. Different simulation environment will need lots of research to make the simulation results correspond with the realities.  

Pawel Berth said:

In the previous project I used TAS5630BDKD and I want to change it for TPA3255.

I think this maybe a good reference. Because in both device's datasheet, the Junction-to-case (top) thermal resistance value are very similar. Because they are both Pad Up devices, this thermal value is the most important. 

How about change the TAS5630BDKD project's working condition, to be much similar with new design. Check if the thermal meet expectations?

Yes I'm comparing these two IC's. Calculating the Tj from formulas gave me Tj=155°C for TAS5630BDKD and Tj=191,5°C for TPA3255. For how long the maximum power of TPA3255 is measured? Is it continous power or measured in a quick amount of time? I need to provide 2x180W on 4ohms, is it possible with cooling fan?

Hi Pawel

Pawel Berth said:

Calculating the Tj from formulas gave me Tj=155°C for TAS5630BDKD and Tj=191,5°C for TPA3255

So does the using condition changes a lot among these two devices? Because the thermal parameter between these two devices are quite similar, shouldn't cause so big difference. Only reason seems the using condition changes.

Pawel Berth said:

For how long the maximum power of TPA3255 is measured?

We will use validation bench to automatically find the maximum power at every PVDD voltage, so the device could withstand maximum power at least several dozens of seconds. 

Pawel Berth said:

I need to provide 2x180W on 4ohms, is it possible with cooling fan?

The total power is only 360W, much smaller than the maximum power of this device. It is possible.

Shadow He said:

So does the using condition changes a lot among these two devices? Because the thermal parameter between these two devices are quite similar, shouldn't cause so big difference. Only reason seems the using condition changes.

I calculate Rcs from the equation 7 which is Rcs=T/(k*A), which is equal 2,22°C/W for TPA3255 and 0,6°C/W for TAS5630BDKD. The difference is huge because of different thermal pad surface. TPA has thermal pad almost 4 times smaller than the TAS. 

Yesterday I received the evaluation board of TPA3255, I will test it for conditions that it will have to provide (output power for exact time). 

I just measured the amplifier and it can provide 2x180W on 4ohms only for 25 seconds without fan. Adding a fan improved the time to 45 seconds. After these time period, the OTW occured. Could You tell me which thermal grease do You use in evaluation boards? I wonder if I use better one, then the amp will work fine without any warnings.

OTW or just clipping warning. I'm not sure because diode wasn't bright. I will measure it with osciloscope tomorrow.

Ok, thank You. I will measure the evaluation board and will see if any other problem will occur. I hope not :)

let me close it temporarily.

if having any problem, we can re-open this issue.