STP45N65M5 damaged

[arsaenz]

Hello. I am using your STP45N FETs in our power supply design for an audio amplifier and we are getting field returns with the STP45N65M5 part damaged. It appears that the gate and source have been shorted but I am not 100% sure. Do you have any insight as to what may cause this or is there a facility where I can send a couple of these damaged FETs to get some failure analysis done? Thank you and please let me know what additional information you may need.

[PGump.1]

Hi,

Unfortunately, nobody here will be able to answer your questions with limited description you have given.

I suggest you get an Ohm meter so you can be 100% sure if there is a short.

Here is a clue - GENERALLY, over voltage creates short circuits, over current creates open circuits. However, short circuits can create over current...

Good luck!

Kind regards
Pedro

[arsaenz]

Hello. I took a DMM to an undamaged unit and logged the following readings:

- VGD = Open

- VGS = Open

- VDS = Shorted

I then took the same measurements on the damaged unit and these were the results:

- VGD = 2.15Kohm

- VGS = 2.15Kohm

- VDS = Shorted

Please let me know if these readings could have any indication as to what happened (over-voltage spike at the gate, etc.) or if there is any other measurements/information that would be helpful. 

In addition, I would be really curious to get a Failure Analysis report, would this be possible and if so, where could I send the damaged devices?

Thank you!

[PGump.1]

Hi,

"VDS = Shorted" in both cases tells me 1 of 4 things:

1. you are measuring the device while under bias
2. you are measuring the device while under load
3. the 'undamaged' device is in-fact damaged
4. you have the DVM leads reversed...

"VGD = 2.15Kohm, VGS = 2.15Kohm" - Here is a clue - GENERALLY, over voltage creates short circuits.

I hope that helps.

Kind regards
Pedro

[arsaenz]

Hi Pedro,

I am taking these measurements when the devices are not biased and are not under any load. I am simply touching the leads on a DMM to the device terminals when the part is resting on a table. Also the "undamaged" device is brand new, has not been used and was just shipped from Mouser.

I am assuming that it was an over-voltage condition which caused this, however, I am trying to understand if it was a VGS or VDS over-voltage condition so that I know what circuitry in my application I should further investigate. Can you recommend any other measurements I can take which will help me confirm where/how this happened?

Also, is it possible to ship the damaged parts somewhere where Failure Analysis can be done? Thank you.

 

Regards,

Alex

[PGump.1]

Hi,

"VDS = Shorted" in both cases tells me 1 of 4 things:

1. you are measuring the device while under bias
2. you are measuring the device while under load
3. the 'undamaged' device is in-fact damaged
4. you have the DVM leads reversed...

If you have ruled out 1, 2 and 3 - it must be 4!

Kind regards
Pedro

 

[arsaenz]

Hi Pedro,

OK so reversing the leads and the new results are below. This seems to make a bit more sense given that there is no short on the New Device. Is there anything you'd be able to tell me about the damaged devices? Assuming it was an over-voltage condition that damaged the devices, do we know if it was a VDS or VGS OV? Thank you.

 

	RGD	RGS	RDS
New Device	Open	Open	1.481M
Damaged Device #1	2.371K	2.378K	Shorted
Damaged Device #2	213.2K	217.4K	Shorted

[PGump.1]

Hi,

---

@arsaenz wrote:

is it possible to ship the damaged parts somewhere where Failure Analysis can be done?

---

 What do you expect them to tell you?

I'll save you a lot of money, because any report will simply tell you - "It was a BIG electron(s) going VERY fast that punched a hole(s) in the junction..."

Here is a clue - the Analysis doesn't need to be done on the device, it needs to be done on your design.

Kind regards
Pedro

[arsaenz]

Hi Pedro,

Yes I fully understand that the analysis needs to be done on my design thank you. The only information I am trying to gather from STMicro is if we can deduce whether or not the over-voltage condition occurred on the drain or gate. If this is not possible, or you need to see other measurements to make this determination, or you can make an educated guess based on the information provided, please let me know. Thank you.

 

Regards,

Alex

[arsaenz]

Hi Pedro,

I am following up to see if there is any indication in the data/observations I'd presented in this discussion that would indicate if the damage was due to an over-voltage at the gate or drain. Thank you.

 

Regards,

Alex

[PGump.1]

I have nothing more to add. Have you?

Kind regards
Pedro