UCC28951-Q1: UCC28951-Q1

no other changes we made except those values - sentence correction sorry

Hi Rajashekar:
Can you fill in the parameter in the UCC28951 EXCEL calculator and send it to me?
here I need the driving circuit to see whether the circuit is correct.
please also send the DF1 and DF2,and QA datasheet to me.


I recommend to use 600/650V CFD series mos instead of using 1000V rating mos. it is because usually the higher voltage ratio mosfet used, the worse recovery performance has.
I guess the blown out of mos is because of the over current or over temperature, not over voltage. So try to enlarge the deadtime of the mos.

hello David

I didnt find the UCC28951 excel calculator. please give me some path to find it.

here iam attaching the driving circuit6165.gate driver.pdf

one correction in the above circuit that DT pin left open and EN pin is connected to the supply voltage with the 10K ohm pull up resistor.no other changes we made.

Iam also attaching DF1 , DF2 both are same superfast recovery diode.pdf

QA datasheet is C3M0065100K input mosfet 1000V 35A 65m-ohm.pdf

Thanks and regards

Rajashekar

Hi Rajashekar,
You can use the UCC28950 excel calculator.
Below is the link.
www.ti.com/.../getliterature.tsp

Hello 

Here iam attaching the excel TI 600W design mine.xls

Please correct where im going wrong.

Thanks and regards

Rajashekar

Hi Rajashekar,
It is ok to me. You may be not need the shim inductor, but please leave a place for it.
Does the mosfet broken issue been solved?
What kind of mos is used this time?

Hello David 

this issue of broken mosfet is not solved, we used mosfet part nmbr: C3M0065100K 1000V 35A .

the failure for mosfet has only three reasons 

1. voltage rating

2.current rating

3. junction temperature

We had found that there is no heating of mosfet ,in addition we attached sufficient heatsink for it.

when we talk about voltage , we had given 300V dc input to converter but it is very less than the rating of mosfet.

With the same input voltage 300V ,input primary current is less than 10A which i have seen from the calculation i.e, around 5-6 A.

Does the Transformer had gone to saturation where the excessive current had flown into the mosfets?

I will specify transformer specifications 

Transformer core used EE65 ferrite core N87material

Np : 16turns ,single strand taken on primary.

Ns1 : 1 turn

Ns2 : 1 turn

Wire used : 42/200 SWG litz wire  i.e 42 SWG 200strands litz wire both on primary side and secondary side of transformer

5 strands taken on each secondary as we are testing for 25A output , so it can carry 30A easily

Please check in ,if iam going wrong and correct me.

Thanks and regards

Rajashekar

Hi Rajashekar,

As I told you before, Can you please try 600-650V CFD series mos to see the performance?

here I attached one mos I used before.  the part number is IPW65R080CFD.

I do not have the experiment using SIC mos, but here my thinking is :

it is not possible mosfet broken is caused by voltage, except that the spike on the mos is too higher than 1000V.

The biggest suspect is the over current that caused Junction temperature over spec. you can also scope the mosfet's current the through Drain pin of the upper mosfet. Using trigger mode you should be able to see the current  and the voltage of the mos .

you can also scope the primary current of the transformer to see whether it runs into saturation.

Infineon-IPW65R080CFD-DS-v02_04-EN.pdf

5123.Control circuit edited new 1.pdfHello David

the mosfets had blown due to exceed of its rated current 35A.

Proper controlling is not happening at the input side of converter. What might be wrong ? what changes should i do.I have attached control circuit at the first.

can you check the compensation part also

thanks and regards

Rajashekar

Hi Rajashekar,
I think I had told you clearly, please try the component and scope the need waveform to me, otherwise we can not know why the mosfet is still broken.
You should be able to test the GS waveforom of both upper and lower mos during the broken moment. this is the way you can find why there is over current on the mos.
one information I need to add is: from the driving circuit, you use a MGJ2D121505SC DCDC to offer the power supply to UCC21521. Can you please check this device again?the input for this device is 12V, but here you use 5V.

Hi Rajashekar,
do you have any update on this post?