Post Go back to editing

LT8210 Burn out the mosfet

Category: Software
Product Number: lt8210

Dear ADI:

    I use lt8210 for buck boost charger for battery , But now I use 2.5mΩ RSENSE series to inductor for sample inductor current , And my inductor saturation current about 30A , When input is 39V, output is 54.7V, LT8210 work in boost state and CCM mode ,Then the MOSFET A  and C  is burn out , Maybe Overcurrent  and inductor saturation, But i Check the datasheet page 5 about "Cycle-by-Cycle Inductor Current Limiting" ,The give message is that "VSNSP1 – VSNSN1 = 50mV"  If 50mV is correct , When I use 2.5mΩ Rsense , The Maximum current  just only 20A , I wanna know the VSNSP1 – VSNSN1= 50mV is right or not ,TKS!

Parents
  • Dear Chris: Now i have Fixed startup issues, But now i have find the problem as following:

          I use 4 mosfet, Every mosfet Qg about 50nC Then operation frequency is 200Khz, So i count out the current comsumption 

    I = 4*50nc*200K = 40mA, And external power 13V supply to EXTVCC and GATVCC about 10V ,When output current less than 0.5A/50.4V, GATVCC Ripple is very good, And LTC8210 efficient arrivate at 98.5% in CC mode, But if increase output current, For example 1A ,The lt8210's  GATVCC is very unstabity, So many problem is out, can you help me find out the reason

        I use external Switch Power IC power to EXTVCC pin ,can output 1A, When GATVCC unstabity, EXTVCC pin signal is very stablity

  • Hi fhlxUpdate,

    This sounds like the capacitor on either GateVCC or EXTVCC is to small. Try increasing the size of the capacitor.

  • Top layout

    inner for GND

    Inner for sigal

    bottom layout

    component and Top layer

  • PDF

    This sheet come from ADI Demo board 

  • Hi fhlxUpdate,

    The layout I think is OK. 

    I look at your schematic. You have a GATEVCC cap of 10uF and boost caps of 0.47 uF. Please check these values. If the ratio of GATEVCC cap to boost cap is not very high, then you can have difficulty getting a good GATEVCC voltage. As a rule of thumb, the GATEVCC cap should be 100x the boost cap. Please check to make sure that you have adequate GATEVCC capacitance.

    Thank you,

    Chris

  • Hi  cgass, Very thankful you at least, I Try to add GATEVCC caps to 20uF, But no good effect In large current, Cux PCB's room limited, Big caps can't add, So stop! I have designed LT87xx series Chips etc, just GATEVCC is 10 times is okey, For Lt8210, gateVCC is so high, maybe Capacitor is high too.

    In fact, I have Test 5 PCS board, First 54V/2A output , Efficiency is very good, But my project At least  400W and more!As the current increases, the MOSFET suddenly burn, and then GATVCC is Abnormal next times!!! Now expect new Sample IC for prototype, Tks you any ways

  • You can also reduce the boost cap value to 0.1uF. Either increase GATEVCC or reduce boost.

  • Dear cgass: my Project requirement about 500W power supply, If  boost cap value to 0.1uF, Just only use lower Qg value Mosfet, Isn't it ? Low Qg valuce MOSFET Rdson is high, And the very hot, And discrease the efficient? 

      Another problem, GATVCC capacitor size is increase, and output is stablity, And can work in  400W output, But Inner LDO Get issue, Very easy break; especially startup, Damage to LDO due to transient inrush current When high caps!Maybe Series limiting resistor can do it ,I have Broken 8 pcs chips Since the LDO can't work, Maybe high power output for LT8210 is not good

  • Dear cgass: 

      The datasheet page 29" The bootstrap capacitors CBST1 and CBST2 need to store roughly 100 times the gate charge (QG) required by top switches A and D. In most situations, a 0.1μF to 0.47μF, X5R or X7R, 25V capacitor is adequate. The bypass capacitance from GATEVCC to ground should be at least ten times the value"

        1. If use 10 times CAPS, only output 48V*2A is good select, If increase power, GATEVCC ripple is out 

         2. If use  100 Times CAPS, Can output 48V*10A, But GATEVCC Broken Easily in startup.

  • Hi fhlxUpdate,

    What are the MOSFETs you are looking at? I would like to review the datasheet.

    I have another design working in the lab with 2xBSC072N08. Same MOSFET on high side and low side. Each MOSFET has 25 nC of gate charge, so 50 nC total gate charge. I have 4.7 uF GATEVCC cap and 0.1 uF boost caps. This circuit works well. It runs 35 Amps of inductor current.

  • hi, I will try to use 0.1uF boost capacitor next time, We use CSD19533Q5A, If single Mosfet(0.47uF boost,10uF GATEVCC is all okey) ,it Qg =27 nC,  But if Double Mosfet, Fault is out as above! Cux i wanna use 10A -12A output ,single MOS is dangerous! Our problem as following:

        1. Maybe LT8210 GATEVCC LDO is not strong! How to protect it when produce?

         2. Another Ways, LT8210 is easy burn when output is short suddenly, I have used your Buck-boost chips similar 3790, 8708, 8390 etc, They short protect is good CHARACTERISTICS , But 8210..., Maybe should take time test it!!!!!

  • Hi fhlxUpdate,

    I am sorry you are having these difficulties. Do you have a demoboard of the LT8210? Even though the current is lower than your target, you can test the short circuit capability and other protection features. 

Reply Children
No Data