0

I am working on a power supply for a stepper motor driver which requires 48 V and 12 V to operate. The 12 V rail must present all the time if it was cut off for any reason then the 48V should also be disconnected! I am feeding the 48 V to a DC-DC Buck to get the 12 V rail.

enter image description here

I do not want to use a mechanical relay so I came up with this circuit to disconnect the 48 V in case the 12 V output of the buck converter was shutdown for any reason.

enter image description here

I have noticed that it has a very slow response time, it may take up to 600 ms starting from disconnecting the 12 V till the 48 V disconnects(!) which is not accepted; I want it to be in the range of few milliseconds! I tested this circuit with 48 V and 12 V lab power supply with load up to 100 W.

So how to improve this circuit to have a faster response or maybe if it is not suitable at all, may be I should replace it with something else!

EDIT

The 48 V circuitry with the buck converter are on a sperate rail and the p-mos switch circuit will be the main entry point supply for the stepper motor driver circuit and will be fed directly from the 48 V (after reverse polarity protection and the 12 V source), so in normal conditions, the 48 V rail with the buck converter will work and supply the p-mos switch circuit with the required voltage which will supply the driver circuit but if the buck converter stopped working for any reason, the p-mos switch circuit will not work and the the driver circuit will not be supplied with the 12 V and the 48 V. The power supply circuit is still under development and is not meant to be a final product yet! so I added several components just for extra protection in the development phase!

EDIT

I found that my bench power supply is the cause for the long time response, it would take up to 500 ms to go below 0.7 V .

Now I would like to know if the the pmos switch circuit has any hidden issues with-in the current configuration or there is any other better solution .

Muhammad Nour
  • 527
  • 6
  • 16
  • You realize that the left-most TVS, the left-most LED and the PMOSFET have an on-going competition over who will go off first in case of reverse polarity? My bet is on the TVS since it has nearly zero resistance. Should be a nice pyrotechnic effect at 3A, the fuse will not save the day. And the P MOS is just a BoM filler brick. – Lundin Aug 10 '23 at 14:32
  • 1
    How can your circuit ever begin if it requires 12 volts to be present to activate the 48 volts and, 12 volts needs the 48 volts to be present? – Andy aka Aug 10 '23 at 14:44
  • @Lundin No competition -- that's a unidirectional type, so the reverse is about a volt. Conveniently, the PMOS can be skipped; the fuse won't be happy about it though! – Tim Williams Aug 10 '23 at 15:35
  • At what load current was the 600ms "disconnect" measured? On both the buck reg and the 12V supply. – Tim Williams Aug 10 '23 at 15:37
  • @Lundin some components are added for extra protect during the development phase – Muhammad Nour Aug 10 '23 at 16:18
  • @Andyaka, I have made edit to the post and added extra explanation about what you have mentioned – Muhammad Nour Aug 10 '23 at 16:18
  • I tested the circuit with 48V @2.3A – Muhammad Nour Aug 10 '23 at 16:19
  • The 12V supply was 100 mA – Muhammad Nour Aug 10 '23 at 16:22
  • You need a 10 V zener diode in series with R38 if you don't want to wait until C22 is down to 0.7 V. – Jens Aug 11 '23 at 01:53
  • @TimWilliams Hmm, it really ought to be drawn as one (or as a zener to begin with). And I don't see how the PMOS can be skipped for polarity protection, otherwise the rest of the PCB will get energized backwards and the path of least resistance will take the blow. – Lundin Aug 11 '23 at 06:27
  • The first TVS is to prevent voltages over ~ 48 V and the pmos for the reverse polarity, removing the pmos will definitely allow reverse polarity current to go flow. – Muhammad Nour Aug 11 '23 at 10:48

0 Answers0