50V Drain Supply Voltage - LNK304

8 posts / 0 new
Hi, I'm using a LNK304 in a buck-boost configuration. Input voltage designed for is 25V-265V. The supply starts up at 19VDC. The problem is, that I'm not sure (because of the minimum 50V drain supply voltage spec in the datasheet) if I can run the supply from this low voltage or not. I just want to know if it will run ok, I know it's not recommended, but I really need it to run at least safely from 25VDC. I designed for 150mA. But I'm only drawing about 75mA maximum. Can you also please give more information on the 50V minimum drain supply voltage? Is it a must? Or is there a absolute minimum ? Regards, Quintin

Hello Quintin,

PI can only guarantee the operation of the device over the full device temperature range at a drain voltage of 50 V minimum. This ensures that sufficient current is available to operate the IC (charging of the BP pin capacitor during the MOSFET off time).

You can extend the operating voltage range by supplying current to the BP pin externally. The following DI shows the principal.


Hope that helps



Thanks for the reply. I will implement the extra circuit in the di152 pdf. Will let you know if I get my supply to work from that low voltage. Regards, Quintin.
hellosir, pls may i have you mail id
Hi Quintin and PI, I have almost the same requirement (input voltage range 18VDC to 200VDC), output voltage 8.5V at 100mA. I am using a LNK305 at the moment and it works ok most of the time. What happens with my setup at the moment is, that everything works ok, if I disconnect the input voltage and connect again during the next x hours. If I wait for some days, the circuit will not always startup again. I see timeouts (after ~100ms?) and regular restarts every 0.8s. This goes on for some 20s and then everything is working ok again. The reason for this strange time-dependant behaviour seems to be that after more then ~1 day all capacitors are completely discharged to a level where the circuit will not startup. If I directly short the (input) caps for some seconds I get the same behaviour. Only way found up to now to make it work is to touch pin 2 with an oscillosope probe. Then everything works and the circuit will also run down to ~15V with full output regulation. Please inform me if your circuit is working in the end, I will do the same. Regards, Jörg.

Hello Jorg,

thanks for sharing your experience - makes my role on the forum that much more enjoyable! Could you share your schematic? Depending on the configuration it may be possible to tell if the issue is power delivery (to charge the output caps within the auto-restart time as you correctly pointed out) or insufficient voltage to keep the LNK305 operating. If it is too low an input voltage you can monitor the BP pin voltage with a scope - but caution in needed depending if the SOURCE node is a switching node or not (hooking up the scope ground may cause a short). If the BP pin goes below 4.8 V during startup the device will stop switching until the voltage charges back to 5.8 V - this would indicate insufficient voltage to operate the IC. Again PI only guarantees operation to 50 V.

If you post your schematic I can check if that'll work.



I have attached part of my schematics. It is from a control card for a universal lead acid battery charger for 12 to 48 cells (24V to 96V nominal voltage). DC Input is therefore ~18V to 180V. The +8.5V output drives some LEDs, the LCD backlight and some relais. The ~4.5V VCC2 is for the LCD logic supply, VCC1 for a microcontroller with RS232 (backed up by a lithium cell in sleep mode without battery connected). This design is already used in a 10-20k/year product (although i knew about the input voltage limitation to >50V) and all the prototypes worked very well. Now, in production, about 5% of the boards show the slow startup after a battery is connected. This is not a complete disaster, but at least annoying. Hope you can find a way to help me with this design. Jörg.
The 50 V requirement is to make sure that the internal current source can charge up the BP capacitor and start switching the internal MOSFET Alternatively you can try feeding a current into the BP pin in the form of a floating current source. This should enable you to start up from lower input voltages. See design idea 152 or see this link for more details - http://www.powerint.com/sites/default/files/PDFFiles/di152.pdf