DC-DC converter trouble - any suggestions?

[FANT]

DC-DC converter trouble

Hi guys, I need your help !
I am trying to build a switching voltage regulator for this application:
input voltage : from 1,2 to 6 Volt from 4 * 1,2 V 4Ah NI-CD batteries
Output voltage : about the same as the imput if the input voltage is above 4,5 V, regulated at 4,5 Volt if the battery voltage is below ( until 1,2 V )
Output current : 0,8 A for a incandescent lamp ( consider that at the tur-on the resistance of the lamp is very low )
The regulator has to be switching to improuve the efficency.
The application is for the lamp used on the headset of the miners friends of mine.
I have already used the Maxim MAX1709, but as the input voltage is higher than 4,5 Volt, the IC is usable only to fry the eggs !

Thanks for any suggestion

 

[goodboy_pl]

take a look at:

**broken link removed**

maybe useful!

BEST!

 

[nebisman]

you can try a design with old lm3578 from national semiconductors. Its a good switching regulator.

 

[ted]

I would try to design the device as a two-step functionality: A uup-converter for voltages below 4.5V, and a pass switch (I would try a MOSFET switch) for voltages over 4.5V.

In this manner the very low resistance switch (high efficincy) would be used instead of a converter when the batteries are full, and the converter with much lesser eficiency (maybe around 90 % for a good design) only when needed.

When you operate on so low voltages and still need some current synchronous rectification might improve the eficiency remarkably.

Of course, we are talking about a bit costly components here, so all depends also on the target price.

Btw. I react a bit on use of NiCd battery: you get much more energy storage per volume/weight and still retain quite low cost by using NiMh cells.

 

[FANT]

First of all thanks to everybody for your help.
Answering to ted, the target price is quite important; anyway the most important thing is the reliability ( in any case is foreseen a pass switch to jump the regulator if ths fails ) and the efficency.
About the NI-CD batteries it is a must because they have already the battery packs and this is a improuvement of the existing system.
The strange think is why the max 1709 burns as soon as the battery voltage is higher than the set point ?
sholdn't the the switching transistor simply turn OFF and leave the output voltage be as the input voltage ?
In the data sheet is not said anything !
Can anybody expalin this ?
Thanks in advance

Mandi

 

[johnyaya]

Regarding the use of NiCd batteries... If you take the 4 cell pack down to 1.2 V you will end up ruining the batteries sooner. NiCd battery output falls off quickly, so there isn't much run-time left to warrant going that low. If emergency light is required you could devote a cell or two as backup. An approach that I recently used was to carry 2 battery packs and then use automatic control to switch from the dead battery pack to the live one. By the time the 2nd pack is going dead the 1st pack will float-up from not having a load. With this technique the operator can see when half of there power is gone, then they get a little bit more at the end from switching back and forth until both packs are used up.

The type of switching regulator configuration you need is buck-boost. The "buck" is for when the input is above setpoint and the "boost" is when the input is below setpoint. Sorry, no other comments on your switching regulator problem. Good luck!