Hi, I'm designing a PCB to control a motor and a couple of accessories, it has a main power supply of 13.4v but I will never know the polarity of the voltage, so I'm using a diode bridge as a polarity correction, the controller is an arduino (5v) so I'm using a buck converter to convert to 12v to 5v.

It has two main requirements:
1. needs to be a very small layout
2. needs a uninterruptible power suppl y

because is a small layout I want to use a super capacitor as a UPS, it gives me enough power to keep the controller on (I don't care about the motor) when the main power is disconnected (that is not more than a couple of seconds) the problem is that the current is flowing to the motor discharging pretty quick the capacitor and I'm afraid of damaging the buck converter due that is working backwards, I tried with a schottky diode between the buck converter and the capacitor (not a problem having ~4.8v) but apparently it doesn't work, when I disconnect the 13.4v PSU I can read voltage coming from the capacitor in the other side of the buck converter and can see it dropping very quick

If Somebody can explain me what I'm doing wrong or a better solution I will really appreciate it, Thanks

Extra points: I'm using a MAX809 as a voltage monitor, when the voltage drops to less than 4v I want to shutdown the whole system, I read that I can do it with MOSFETs but I have never use a MOSFET before and I'm afraid of not feeding enough voltage or current to the control circuit

 

Anyone?

 

Can you simplify your diagram into a single network, and also include the supercap?

 

Yes, we need to see how all the parts are connected together in one diagram.

 

Both comments match what I was going to ask for. A whole circuit may immediately show the problem, littleportions will very seldom show anything.
But now, to isolate the controller power, you can use a second bridge rectifier from the 1hirteen volt source, feeding the buck regulator. with a separate bridge for the motor power. Then the super cap can be the source of feed for the switching regulator feeding the controller. THAT arrangement should work. You will have a problem with traces on the circuit board being skinny and having excessive voltage drops, though.

 

Both comments match what I was going to ask for. A whole circuit may immediately show the problem, littleportions will very seldom show anything.
But now, to isolate the controller power, you can use a second bridge rectifier from the 1hirteen volt source, feeding the buck regulator. with a separate bridge for the motor power. Then the super cap can be the source of feed for the switching regulator feeding the controller. THAT arrangement should work. You will have a problem with traces on the circuit board being skinny and having excessive voltage drops, though.

Also, it's very important to limit the charge of the supercap to a recommended 20% below its rated voltage, otherwise its useful life would suffer noticeably.

 

Sorry, I just use the schematic that I made for the pcb, hope this can explain better what I'm trying to do

 

Also, it's very important to limit the charge of the supercap to a recommended 20% below its rated voltage, otherwise its useful life would suffer noticeably.

It's a 5.4V supercap, and with the diode I'm charging the supercapto ~4.8V that is like 87%, I really don't want to go lower because I'm using the MAX809 to shutdown the power at 4.0V, maybe I can add another supercap in series to have 8.1V and be in a below 20%

 

Sorry, I just use the schematic that I made for the pcb, hope this can explain better what I'm trying to do

View attachment 156317

That's better ... where's the motor and the arduino?

 

Both comments match what I was going to ask for. A whole circuit may immediately show the problem, littleportions will very seldom show anything.
But now, to isolate the controller power, you can use a second bridge rectifier from the 1hirteen volt source, feeding the buck regulator. with a separate bridge for the motor power. Then the super cap can be the source of feed for the switching regulator feeding the controller. THAT arrangement should work. You will have a problem with traces on the circuit board being skinny and having excessive voltage drops, though.

I agree with you, actually my first design was something like that, but the layout was too big for my needs, that's why I ended up removing the second bridge

 

That's better ... where's the motor and the arduino?

this is a very simplified version of what I need, the actual circuit it's more complex, but this can give you an idea of where the arduino and the motor are, all the extra components that I'm not showing are a transistor array and the communication system

I'm using a resistor as a representation of the motor

 

this is a very simplified version of what I need, the actual circuit it's more complex, but this can give you an idea of where the arduino and the motor are, all the extra components that I'm not showing are a transistor array and the communication system

I'm using a resistor as a representation of the motor

View attachment 156318

Have you tried placing D1 before (and not after) the regulator? Also, I'd consider placing a capacitor between the rectifier and the regulator.

 

Have you tried placing D1 before (and not after) the regulator? Also, I'd consider placing a capacitor between the rectifier and the regulator.

I place the diode after the regulator because I don't want to have any reverse voltage in the regulator, I'm afraid of damaging the regulator applying voltage in the low end, I didn't use any capacitor between the bridge and the regulator because is a buck converter, I think that it has everything that it needs.

This is the buck converter that I'm using

 

I place the diode after the regulator because I don't want to have any reverse voltage in the regulator, I'm afraid of damaging the regulator applying voltage in the low end, I didn't use any capacitor between the bridge and the regulator because is a buck converter, I think that it has everything that it needs.

This is the buck converter that I'm using

View attachment 156319

What exact regulator are you using? I'm unable to tell from the data you've posted.

 

I don't see how the motor can be discharging the capacitor.

What is the size of the supercap?

There is no base resistor for the first NPN transistor (you should give your parts identification labels so they can be referred to) so its current is limited only by the maximum output current of the MAX809. The transistor needs to have a series base resistor to limit that current.
In addition, the 1k collector and 220 ohm base resistors are taking a significant amount of current.
All those currents are likely what's discharging the cap.

If you replace the NPN's with MOSFETs you could reduce those currents to essentially zero.

 

The datasheet was suggested that resistor for the reset is 100K, but you just used 1K and that is too small, you can use 10K to replace 1K, the R_PD1 is also too small, you change to 10K, and at least in series with a 1k Rb for 2N2222, use the resistor more higher and to reducing more power dissipation.

 

What exact regulator are you using? I'm unable to tell from the data you've posted.

This is the buck converter that I'm using
https://www.amazon.com/eBoot-MP1584EN-Converter-Adjustable-Module/dp/B01MQGMOKI

 

The datasheet was suggested that resistor for the reset is 100K, but you just used 1K and that is too small, you can use 10K to replace 1K, the R_PD1 is also too small, you change to 10K, and at least in series with a 1k Rb for 2N2222, use the resistor more higher and to reducing more power dissipation.

Thanks, I will do that

 

I don't see how the motor can be discharging the capacitor.

What is the size of the supercap?

There is no base resistor for the first NPN transistor (you should give your parts identification labels so they can be referred to) so its current is limited only by the maximum output current of the MAX809. The transistor needs to have a series base resistor to limit that current.
In addition, the 1k collector and 220 ohm base resistors are taking a significant amount of current.
All those currents are likely what's discharging the cap.

If you replace the NPN's with MOSFETs you could reduce those currents to essentially zero.

Me neither, that's why I came here, I don't see any logic way for the super cap to discharge that fast when the motor is connected

I have two 2.7v 10F supercaps in series
https://www.ebay.com/itm/Brand-New-...rad-Super-Ultra-Capacitor-Blue-Z/292508660227

the output current of the MAX809 is 20mA, that's why I didn't use a base resistor, do I need to add one even with that output current?

all my problems are only when the main PSU is not connected, that is when the supercap is feeding all the power, if I try to turn on the motor in this conditions the voltage drops pretty quick, in less than a second but if do the same test without the motor connected the voltage drops mostly at the same rate that when the motor is off

but what really bothers me is having voltage before the D1 when the supercap is feeding the voltage

I apologize if I made to many mistakes here, my experience with electronic circuits is very limited, especially making schematics

 

Me neither, that's why I came here, I don't see any logic way for the super cap to discharge that fast when the motor is connected

I have two 2.7v 10F supercaps in series
https://www.ebay.com/itm/Brand-New-...rad-Super-Ultra-Capacitor-Blue-Z/292508660227

the output current of the MAX809 is 20mA, that's why I didn't use a base resistor, do I need to add one even with that output current?

all my problems are only when the main PSU is not connected, that is when the supercap is feeding all the power, if I try to turn on the motor in this conditions the voltage drops pretty quick, in less than a second but if do the same test without the motor connected the voltage drops mostly at the same rate that when the motor is off

Please post the motor's datasheet as well, if available. I doubt it has a series resistance of 1k, as you previously posted.