Author Topic: Question about smd components operating temperature  (Read 110 times)

0 Members and 1 Guest are viewing this topic.

Offline jakesvdm27Topic starter

  • Contributor
  • Posts: 10
  • Country: za
Hi,

I am working on an embroidery machine for my mom.  It has a robotics box connected to a screen via a RJ45 to ST40 cable. There is a RS485 chip with PORT A and B, as well as PORT Y and Z on both ends. It uses one set to send data and the other set to receive. I.e A and B sends and Y and Z receives. Or the other way around, cannot remember now.

Background

The error we get on the screen is: The system is unable to communicate.

That is why I opened the screen. I am trying to fix that issue. I know the robotics box is sending data. With the screen unplugged, there is no data flowing. Used a USB to RS-485 converter. Oncer the screen is plugged in, the data flows. I also know the screen is communicating with the robotics box. You can click on a button to test PORT D on the screen, and the robotics box runs the test. My test may not be scientific, but this is just some background on how I got to the issue my question is about.


But, then I noticed the one PC board in the screen gets really hot. It has two boards, one attached to the actual touch screen, and another connected via a ribbon cable. The latter gets hot - image A.

I have identified one smd chip that gets really hot. It is on image C, markings on the chip is A84H. Measured 115 degrees celsius on the chip. If I identified it correctly, is is a fixed frequency LED driver - MP1518. Datasheet is also attached. This thing starts high and climbs quickly to 115.

On the other side of the PC board, off to the edge, is a MP2467 (image B) step down converter. It drops the 24v in down to 5v for the pc board. Now that A84H chip heats up the PC board, and eventually gets this MP2467 chip up to about 103 degrees celsius. The MP2467 starts out at about 45 degrees. But gradually heats up to the 103.

The caps, resistors and schottky diodes around the respective IC's, reach the same temperatures as the IC. In other words, 115 degrees celsius and 103 degrees celsius.

Image A is to explain where these IC's are located. On the bottom right you see the MP2467. To the left of it is a long white connector (don't know what it is called). On the back of the board, "under" that connector is the A84H (MP1518).

My Question:

The data sheets say that the oprating junction temperature is between -40 and 125 degrees celsius for both IC's in question. I know that when I measure temp on the IC, it is the case temp and not junction temp.

My gut feeling tells me that this is way to hot for those IC's and that I have a problem. I just want some confirmation. I did some research on google, but anything temp related with smd components talks to the soldering temperature. That is not what I want to know.

I want to know if 115 and 103 is too hot for an IC to operate at all the time. Temp is in degrees celsius. In other words, do I have a problem with or around the A84H chip. And yes, the inductor gets just as hot.

Thank you for the assistance.
 

Offline Siwastaja

  • Super Contributor
  • ***
  • Posts: 8441
  • Country: fi
Re: Question about smd components operating temperature
« Reply #1 on: Today at 09:46:44 am »
Yeah, 115degC would be a (barely) acceptable temperature for the internal semiconductor junction but definitely too much for a reading out of the plastic package. The plastic does not conduct heat that well, meaning there will be a temperature differential within the plastic. If the package surface is at 115degC, junction will definitely exceed the 125degC rating.
 

Offline MathWizard

  • Super Contributor
  • ***
  • Posts: 1540
  • Country: ca
Re: Question about smd components operating temperature
« Reply #2 on: Today at 10:12:36 am »
So what did you measure the temperatures with? Did you check any voltages? But it sounds like it mostly works. But yeah no one would normally design to have junction temps that close to the max, and you're talking about case temps. So yeah are they accurate? I think near 50C will burn my fingers pretty quick. +100C will boil water at sea level.
 

Offline jakesvdm27Topic starter

  • Contributor
  • Posts: 10
  • Country: za
Re: Question about smd components operating temperature
« Reply #3 on: Today at 10:41:45 am »
Thank you for the responses. You have no idea how much I appreciate it.

I meassured it with my firends FLIR camera. Not sure what model he has, but it is fairly new. And I have a blister on my finger as well, so I do believe it is over a 100 degrees celsius. The blister is just a rookie mistake ... lol

The PC board in that region is also just over 100 degrees. No doubt in my mind about the 115. in fact, the PC board gets too hot to hold in your hand.

Again, thank you very much. Have a fantastic weekend
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf