We currently face an issue with voltage regulator LM2679-ADJ. It is part of the design of an ECU controller board for use in trucks . The boards contain 3 pieces of that regulator each, i.e. 3 channels..
ECUs of that design (current layout and component placement) have already been in operation for more than 5 years. With some hundred copies the error I have to report, did not show up so far. Since the production of a new batch of 150 pieces this fault occurs within this new series at about 10 out of 100 devices. Two of the three channels (as well as the previously made devices) work flawlessly - the third channel (always the same) fails. The majority of the devices worked properly during and after commissioning. The fault occurred only after a few power on/off cycles.
In one case a consecutive functioning, fault and functioning again was observed: function -> fault -> function -> fault (persistent)
The fault is, that there is no working output. The setup is for a 13.8V DC output, on a 28V DC input. Instead of this, I habe something about 3.5V output (without any pulsing), if I connect a load to the output there is only a very small decrease in voltage, but the additional output current ist the same as the additional input current.
When measuring on the regulator we noticed that the SS pin of the module does not supply power - the SS capacitor is at 0V, even after removing the capacitor (short circuit excluded), no current can be detected.
The values of the surrounding parts and the solder joints were tested on multiple copies: no errors are detected. However, exchanging the regulator brought the whole circuit back to life. The failure always occurred immediately when switched on (or during storage?).
As long as the reason for this fault is unknown the batch is on hold.
We therefore ask if you could kindly review the schematic and layout documents attached. On the bottom side of the circuit board an aluminum plate heat sink is applied. Coming in voltage, ground plane an going out voltage lines are located to the bottom side.
Do you detect any systematic fault here?
Could the error be explained with increased thermal stress during reflow soldering?
Would you expect any critical transients on the input or output based on the circuit design?
Other ideas?
-
Ask a related question
What is a related question?A related question is a question created from another question. When the related question is created, it will be automatically linked to the original question.
