Uhm .....
LM2621: never tried it, cause looking at the datasheet, the circuit needed for the FB input is too much complex for be directly substitute with a low value resistor.
LM2577: i discarded it when i seen on the datasheet that it still require a FB of 1.2V (high FB voltage = high power dissipation on the RSENSE), and anyway the current is limited to 800mA
LM5118: very complex, it require 2 external mosfets for work in buck/boost mode, and don't work under 3V
MAX668: Not tried it, cause the min Vin of 3V ..... tried the MAX669, instead ..... it work until some point, then tend to become unstable ..... probably cause for low voltages you need to use it in bootstrapped configuration, that take the power supply from the output, and sometimes it go in self-oscillation (i cooked a test load, just imagine a LD
)
MAX862: never heard about it ..... are you sure you don't mean MAX6825 ? ..... if is the 6825, can be, it's a buck/boost ..... need to get some samples (if Maxim is not tired about all my past samples requests
)
MAX1760: not tried it, but looking at the datasheet, is not too much promising ..... max current 800mA, max Vout 5.5V, and the FB regulation depend from the OUT pin, so you probably cannot directly connect it to a RSENSE in serie with the LD (that anyway is still 1.24V, too high for good efficence using only a RSENSE) ..... also, the ISET pin does not regulate the load current (it was too good, otherwise
), but only the coil current.
LT1310: promising, but still high FB voltage, and need compensation (a little bit more complex circuitry)
SC4501: same as before.
I know i'm playing "devil's advocate", here, but designing a constant current boost driver, must be took in considerations a lot of parameters, if you want to get a working driver, at the end ..... and, trust me, i'm one of the first ones, that want to find a good and cheap alternative