I made a test board of a power supply that I'd like to output +/- 15V at a max of 1.3A from a 24V AC-DC converter. I followed this app note:
https://www.vishay.com/docs/76946/creatingnegativeoutputvoltage.pdf
Instead of the converter in the app note I used LM46002, which is also a synchronous converter. It drops out of regulation at higher current and I'm trying to find out why.
Here's the circuit:
It worked fine at 0.3A. At 0.6A it did this for about 25 seconds, then snapped to -15V:
At 0.9A it took longer to become regulated, and at 1.2A it wouldn't regulate at all when it started up with the load. If I switch the load in after startup, it drops out of regulation in a few seconds.
The shape in the above scope image has a frequency around 100Hz, and is audible when it happens. One thing I noticed in the LM46002 datasheet is an overcurrent "hiccup" mode which shuts down the regulator for about 5.5ms in the event of overcurrent in the low side switch. It shouldn't be triggering an overcurrent condition, which has at least a 1.8A threshold, based on the calculated inductor ripple current and the load current. But if it somehow is, the shape makes me think the lowest magnitude of the voltage is where the regulator turns back on, and the highest magnitude point is where it turns off again.
The converter used in the app note (SiP12116) also has overcurrent protection but doesn't have the longer shutoff period of the LM46002. Also the overcurrent threshold for the low side switch of the LM46002 is lower than the rated output current (but still higher than my calculated peak current), while the threshold for the SiP12116 is above the rated current (which is the same as the load current in the app note design).
Could this behavior be overcurrent protection? If so, what do I do about it? I don't see how it would be over the limit in the first place. Also on the same board I used the same design for a +15V output, and it works fine at all load currents.
https://www.vishay.com/docs/76946/creatingnegativeoutputvoltage.pdf
Instead of the converter in the app note I used LM46002, which is also a synchronous converter. It drops out of regulation at higher current and I'm trying to find out why.
Here's the circuit:
It worked fine at 0.3A. At 0.6A it did this for about 25 seconds, then snapped to -15V:
At 0.9A it took longer to become regulated, and at 1.2A it wouldn't regulate at all when it started up with the load. If I switch the load in after startup, it drops out of regulation in a few seconds.
The shape in the above scope image has a frequency around 100Hz, and is audible when it happens. One thing I noticed in the LM46002 datasheet is an overcurrent "hiccup" mode which shuts down the regulator for about 5.5ms in the event of overcurrent in the low side switch. It shouldn't be triggering an overcurrent condition, which has at least a 1.8A threshold, based on the calculated inductor ripple current and the load current. But if it somehow is, the shape makes me think the lowest magnitude of the voltage is where the regulator turns back on, and the highest magnitude point is where it turns off again.
The converter used in the app note (SiP12116) also has overcurrent protection but doesn't have the longer shutoff period of the LM46002. Also the overcurrent threshold for the low side switch of the LM46002 is lower than the rated output current (but still higher than my calculated peak current), while the threshold for the SiP12116 is above the rated current (which is the same as the load current in the app note design).
Could this behavior be overcurrent protection? If so, what do I do about it? I don't see how it would be over the limit in the first place. Also on the same board I used the same design for a +15V output, and it works fine at all load currents.
Last edited by a moderator:
