CSD18534KCS: CSD18534KCS

Hello Sabir,

Thanks for your interest in TI FETs. The top FET in a synchronous boost converter will conduct from the input (source) to the output (drain) through the intrinsic body diode when the converter is not switching and the input voltage is applied. This is because Vin > Vout and the body diode is forward biased. This is expected behavior and should not cause damage to the FET or synchronous boost controller. I suspect that may be what is occurring in your application. Please check it out and let me know if this resolves your issue.

Best Regards,

John Wallace

TI FET Applications

I don't think that is the issue. As when i power it on with CSD18534KCS the board just shows short. And when i use other fets it works as it should. 

And if the issue is with body diode then it should start switching, but instead it shows short. 

Hi Sabir,

Thanks for providing updated information. I have additional questions/comments:

1. Are both FETs displaying this shorted behavior in-circuit or is it just one of them?
2. How are you diagnosing the FETs after they are removed from your board?
3. A DMM can be used to measure drain-to-source & gate-to-source resistance and to measure the forward drop of the body diode from source-to-drain.
4. I would also recommend comparing FETs removed from the board to new devices as well as the competitor's FETs.
5. Can you share your schematic?
6. What is the competitor's part number?
7. I usually recommend a gate-to-source pull down resistor to prevent unwanted turn-on of the FET if the gate is left floating. Please see app note at the link below.

https://www.ti.com/lit/an/slpa021/slpa021.pdf

Thanks,

John

Thanks john for the reply.

1. Yes both fets show same short circuit.

2. Using DMM

3. Already tested.

4. Shows same resistance on DMM as new ones.

5. I can but i don't think it is different from ti recommendation. Also i have already reviwed here with a guy named Niklas. And he had not issue with my schematic.

6. Already used with 55N06 IRF3205 IPA086N10 IPA093N06 CSD18511KCS.

The board in question is currently fully working with 55n06. And have been using it since 2 days. And as soon as i replace the fets with CSD18534KCS, the short appears again

Tried 10k down to 1k  as a pull down resistor for gate to source. But as soon as i solder the gate pin to the circuit drain to source short appears.

Hi Sabir,

Hmm, this is a bit of a mystery. There is nothing inherently different about the TI FET that would cause a drain-source short when installed on the board. I'm sure you checked the soldering to make sure there are no solder bridges causing the short. Do the FETs measure short from D-S with a DMM when installed on the board with no input voltage applied?

Thanks,

John

Hi John, 

I have checked it thoroughly, and there are no shorts or solder bridges on board. If i do not solder the gate pin on board then there is not short from drain to source and i have checked it with DMM. 

I am thinking of building another board to check whether this phenomenon is occurring on this specific board. I don't know whether it will make any difference.

But will check and get back to you. It will take 2-3 days for me to build another one. So you won't be listening back from me for that time.

Thanks

Hi John, 

I took some time out to test a new board. This issue does not occur on new board. But i am still unable to get these mosfets to work. 

Can you guide me with what values are to be adjusted to get these mosfets to work?

I tried my best to make them work. But even on other board i can get other mosfets to work. I got 88% efficiency with 55n06. 12v in 37v out. 396w input and 350w output by connecting power supply to resistor load.

And other mosfets work like a charm. 

I have followed TI recommendations very closely, yet facing these issues and only after buying certain mosfets(CSD18534KCS)

Can you please help me out?

Thanks 

Hi Sabir,

Can you please provide the full part number for the 55n06 so I can do a datasheet comparison to the CSD18534KCS. TI FETs tend to be lower gate charge and faster switching than some of the competition. Maybe a small value series gate resistor can be added to slow down the FETs. Also, a R-C snubber can reduce switch node ringing. However, it seems that you may not be getting any switching at all. Lastly, can you  provide the product markings from the top side of the case? We can check to see if there is anything unusual about these lots. Did you acquire these FETs directly from TI or from a reputable distributor?

Thanks,

John

FQPF55N06. I have tried 3.3r to 47r with no success. I have bought directly from Texas. And 1000pcs at that.

I get the switching with new board but it does not take any load. Voltage drops from 36v to vin as soon as apply load. 

I will update you with part photo tomorrow 

Hi Sabir,

It sounds like there is an interaction between the FET on the boost controller causing it to shutdown the converter. If you can, please capture the switching waveforms: VDS & VGS. The TI FET has much lower gate charge than the FQPF55N06. I suspect there may be more switching noise due to the faster switching that could be affecting the controller perhaps causing overcurrent shutdown when adding load to the output. I may have to reassign this thread to the controller applications team to assist you further. If at all possible, please share your schematic. You can send it to me via private message or regular email: jwallaceri@ti.com.

Thanks,

John

Hi Sabir,

I am sorry we have not found the root cause for the shorted MOSFETs yet.
I still have your schematic within our chat.

If the problem comes from the controller, it would be great if we could see waveforms from the MOSFET gate signals.
This might show us if there is a voltage due to some noise problem, that turns on the FET unintentionally.

Would it be possible to measure the gate voltage (both at the controller pin and at the MOSFET pin) during the short circuit state?

Thanks and best regards,
Niklas

Hi Niklas,

I will send you private message