SEPIC current control

Hello,

our LT8710 is a controller for SEPIC topology (beside others) which features possible output current control:

https://www.analog.com/en/products/lt8710.html.

One of the examples:

kind regards

Markus

Markus,

I have never worked with a SEPIC type topology before.  

Will you please furnish me with an answer to the following design question for LT8710?

Design requirements: 

Input(V):      Output(V):    Current limit(A):
12V            12V           0.7
12V            24V           1.0

Can L8710 furnish the above power conversions under nominal operating conditions and endure an indefinite short circuit to ground without damage to itself or auxiliary components?  Will circuit recover?  I did not observe any mention of shorting in data sheet.      

Hello,

the two configurations are just fine.

SEPIC topology with its main capacitor has no DC path from input to output, one big advantage of this topology.

The LT8710 is a current mode controller, hence the peak current is measured each switching cycle and controls switching scheme, this already delivers quite a huge safety margin for overcurrent scenarios. On top of this LT8710 features frequency foldback, which reduces the amount of energy which is delivered into a shorted load further on. The sense voltage limit of the current sensing of CSP/CSN pins  is reduced depending on Dutycycle. CSP/CSN sense voltage is controlled with SS pin voltage as well. Given that I would assume that the part would withstand a short circuit already without the output current sensing which you can place on top of your circuit robustness.

if you check out the state diagram on page 11 you'll see that for every abnormal operationg condition part will go into reset (SS pin is discharged to GND, for proper and healthy startup after Reset condition is gone). Short circuit for example will be detected at its FBX pin.

If the short circuit is removed the part will go back to normal operation again.

So if you choose the external components (FETs, inductor) accordingly to the described design step/formulas you'll be safe.

kind regards

Markus

Markus,

I thank you for your time!  I will recommend LT7810 for our next design.

John

Markus,

Digi-key lists available inventory as 0(zero). Yikes! Has lt7810 been discontinued?  Also do you know of the existence of  LT8710 demo boards?

Hello,

availability suffers only from increasing leadtimes in the overall semoconductor market, no obsolescence to be afraid of.

Demoboard info is always accesible via the general product page, click on the appropriate tab:

https://www.analog.com/en/products/lt8710.html#product-evaluationkit

DC2067A

Markus,

Will you please suggest an additional linear technology  part or parts?

 I would like to have a circuit :
1) attached to output of a switching 12V supply
2) attached to output of a switching 24V supply
3) Powered by independent 5V(?) power supply (nice to have)
When enabled, the circuit:
1) disconnects switching power supply from load.
2) After output disconnect, the circuit safely (current limit)  discharges any energy stored by the load capacitance (capacitance might exceed  300uF assume ceramics)
3) When disabled the circuit limits inrush current to a predefined value (similar to LTC4232)

Markus,

Will you please suggest an additional linear technology  part or parts?

 I would like to have a circuit :
1) attached to output of a switching 12V supply
2) attached to output of a switching 24V supply
3) Powered by independent 5V(?) power supply (nice to have)
When enabled, the circuit:
1) disconnects switching power supply from load.
2) After output disconnect, the circuit safely (current limit)  discharges any energy stored by the load capacitance (capacitance might exceed  300uF assume ceramics)
3) When disabled the circuit limits inrush current to a predefined value (similar to LTC4232)