Author Topic: Circuit still sends voltage surge through when plugging into Power Supply  (Read 18646 times)

0 Members and 1 Guest are viewing this topic.

Offline Falcon69Topic starter

  • Super Contributor
  • ***
  • Posts: 1482
  • Country: us
Re: Circuit still sends voltage surge through when plugging into Power Supply
« Reply #50 on: December 27, 2014, 04:48:11 am »
i see a 12v zener.  How's that work with variable input? i.e. 7-24volts.

and both the circuits you shown, go BEFORE the IN pin of the LM2596?
 

Offline Alex30

  • Regular Contributor
  • *
  • Posts: 197
  • Country: au
Re: Circuit still sends voltage surge through when plugging into Power Supply
« Reply #51 on: December 27, 2014, 05:03:21 am »
Yes so my suggestion is to slow start the input (Vin) of the lm2596 and use the low voltage lockout to make the lm2596 turn on at ~2.5V. Then the voltage will slowly continue to rise until the slow start has fully turned on.

As shown in the datasheet http://www.ti.com/lit/ds/symlink/lm2596.pdf :

UNDERVOLTAGE LOCKOUT
Some applications require the regulator to remain off until the input voltage reaches a predetermined voltage. An
undervoltage lockout feature applied to a buck regulator is shown in Figure 34, while Figure 35 and Figure 36
applies the same feature to an inverting circuit. The circuit in Figure 35 features a constant threshold voltage for
turn on and turn off (zener voltage plus approximately one volt). If hysteresis is needed, the circuit in Figure 36
has a turn ON voltage which is different than the turn OFF voltage. The amount of hysteresis is approximately
equal to the value of the output voltage. If zener voltages greater than 25V are used, an additional 47 k? resistor
is needed from the ON /OFF pin to the ground pin to stay within the 25V maximum limit of the ON /OFF pin.

So try a 1.8-2V zener, or even up 4V. Then when it turns on you won't see the output of the lm2596 jumping any higher than the voltage on the Vin side.
 

Offline Alex30

  • Regular Contributor
  • *
  • Posts: 197
  • Country: au
Re: Circuit still sends voltage surge through when plugging into Power Supply
« Reply #52 on: December 27, 2014, 05:17:29 am »
So in the picture shown, I have put a red box on areas where the lm2529 may turn on and then drawn a line at the point at which the lm2596 is set to turn on due to the zener diode (this one is a 4.7V diode). At this point Vin is about 5-6V. The switching regulator can only theoretically supply a maximum of the voltage at Vin. As you can see it's impossible to see higher voltages than ~6V at the output until the voltage at Vin is higher.
« Last Edit: December 27, 2014, 05:19:05 am by Alex30 »
 

Offline Falcon69Topic starter

  • Super Contributor
  • ***
  • Posts: 1482
  • Country: us
Re: Circuit still sends voltage surge through when plugging into Power Supply
« Reply #53 on: December 27, 2014, 05:36:40 am »
what is the supply voltage? did you try it with 7 volt input and 24 volt input?  Does it work as in the picture above you just tested?
 

Offline Falcon69Topic starter

  • Super Contributor
  • ***
  • Posts: 1482
  • Country: us
Re: Circuit still sends voltage surge through when plugging into Power Supply
« Reply #54 on: December 27, 2014, 05:46:25 am »
is the Cin capacitor no longer needed with this circuit?  Before, I was using a 220uF aluminum.  Ti Data sheet shows a 680uF for it for the lockout circuit, however, this is not with the added soft start circuit you added.
 

Offline Alex30

  • Regular Contributor
  • *
  • Posts: 197
  • Country: au
Re: Circuit still sends voltage surge through when plugging into Power Supply
« Reply #55 on: December 27, 2014, 05:56:14 am »
what is the supply voltage? did you try it with 7 volt input and 24 volt input?  Does it work as in the picture above you just tested?

I just chose 16V for that one but I have attached one showing 7V input and 24V input to allow you to get your head around it. I have not properly designed the power transistor or the RC circuit so you will have to work that part out. But as you can see by doing this method, your effective maximum Vin is going to be your input voltage (power adapter) - 0.7V roughly as you have a voltage drop on the transistor. But you should still be able to provide a 5.25V output at 6.3V Vin (7V supply).

is the Cin capacitor no longer needed with this circuit?  Before, I was using a 220uF aluminum.  Ti Data sheet shows a 680uF for it for the lockout circuit, however, this is not with the added soft start circuit you added.

You certainly do. the schematics given in the datasheet often only relate to the part that is being discussed. I recommend you go through the LM2596 Series Buck Regulator Design Procedure section in the datasheet to make sure your component values are correct if you haven't already.


I think using this method should be fine as long as you aren't pulling lots of current, otherwise the power transistor will cook.
« Last Edit: December 27, 2014, 06:00:52 am by Alex30 »
 

Offline Falcon69Topic starter

  • Super Contributor
  • ***
  • Posts: 1482
  • Country: us
Re: Circuit still sends voltage surge through when plugging into Power Supply
« Reply #56 on: December 27, 2014, 06:02:50 am »
with the on/off (pin 5) now occupied, any suggestions where a switch can be used to turn off the circuit.  This circuit board will be inside a control panel.  As the control panel and machine are turned off, this board, as well as other circuit boards in the control panel, need to be turned on and off at the same time.

Any suggestions as to where?  I'd hate to use a big Solid State relay to do this.
 

Offline Alex30

  • Regular Contributor
  • *
  • Posts: 197
  • Country: au
Re: Circuit still sends voltage surge through when plugging into Power Supply
« Reply #57 on: December 27, 2014, 06:16:02 am »
If the switch has to go near the on\off pin just put a switch between D1 and R3. When it is switched on (Short circuit) the supply will be on, when it is off (open circuit) Q2 will be driven low by R3 and Q2 will switch off, then the on\off pin will be driven high by R4, turning off the circuit.

EDIT
Sorry don't do that. As the slow turn on is dependent on when the supply turns on, so your switch must be before the slow turn on circuit.
« Last Edit: December 27, 2014, 06:25:00 am by Alex30 »
 

Offline Falcon69Topic starter

  • Super Contributor
  • ***
  • Posts: 1482
  • Country: us
Re: Circuit still sends voltage surge through when plugging into Power Supply
« Reply #58 on: December 27, 2014, 06:27:31 am »
I don't think that will work?  That would leave C1 fully charged, thus, when the switch is switched to "ON" position, the current/voltage will flow, without a soft start, then I am back to this same problem.

What do you think of coming off the junction of the 50ohm resistor and C1, put a 47K resistor and sink to ground? Then placing the switch between the 50ohm resistor and the supply voltage?  On simulation, it seems the switch is only about 3.5mA, so, a it should be a problem for the control box switch.    The 47k resistor would be so that the cap drains fast, ready for the circuit to turn on again.
« Last Edit: December 27, 2014, 06:30:24 am by Falcon69 »
 

Offline Falcon69Topic starter

  • Super Contributor
  • ***
  • Posts: 1482
  • Country: us
Re: Circuit still sends voltage surge through when plugging into Power Supply
« Reply #59 on: December 27, 2014, 06:53:14 am »
so, something like this I was thinking.....
 

Offline Christopher

  • Frequent Contributor
  • **
  • Posts: 429
  • Country: gb
Re: Circuit still sends voltage surge through when plugging into Power Supply
« Reply #60 on: December 27, 2014, 11:22:15 am »
They really should supply a spice model for this chip!

Are you sure the problem is with the buck regulator and not the logic at the output?

Can you bypass the buck with a 5v wall art and see if the problem still occurs?
 

Offline Falcon69Topic starter

  • Super Contributor
  • ***
  • Posts: 1482
  • Country: us
Re: Circuit still sends voltage surge through when plugging into Power Supply
« Reply #61 on: December 27, 2014, 06:08:43 pm »
I will try that now.

EDIT:  Interesting.  It does it still.  Is it possible that I am still getting feedback from the inductors?

The only thing I can think of is that debounce circuit.  I have an LED that is on that circuit, and that too is flashing for a fraction of a second, which is only suppose to turn on if the switch is pressed.
« Last Edit: December 27, 2014, 06:16:11 pm by Falcon69 »
 

Online Andy Watson

  • Super Contributor
  • ***
  • Posts: 2074
Re: Circuit still sends voltage surge through when plugging into Power Supply
« Reply #62 on: December 27, 2014, 06:29:24 pm »
What precautions have you taken in the design of your LED control circuit to ensure that it powers-on in to a known state?
 

Offline Falcon69Topic starter

  • Super Contributor
  • ***
  • Posts: 1482
  • Country: us
Re: Circuit still sends voltage surge through when plugging into Power Supply
« Reply #63 on: December 27, 2014, 06:31:07 pm »
well, with the debounce circuit, the inverter is floating, unless the switch is hit, then it goes to ground.  WHen switch is open, the inverter is floating until power is applied, then it goes to power, for the input.  So, upon plug-in, it is floating
 

Online Andy Watson

  • Super Contributor
  • ***
  • Posts: 2074
Re: Circuit still sends voltage surge through when plugging into Power Supply
« Reply #64 on: December 27, 2014, 06:36:20 pm »
I think I will need to see more of the circuit to understand what your are trying to achieve with the inverter.
Where does the inverter get its power from?
 

Offline Falcon69Topic starter

  • Super Contributor
  • ***
  • Posts: 1482
  • Country: us
Re: Circuit still sends voltage surge through when plugging into Power Supply
« Reply #65 on: December 27, 2014, 06:52:07 pm »
This is the debounce circuit currently. ("Debounce Circuit")

I have the older boards, that have no debounce circuit (no inverter either), it is just a tactile switch, 1k resistor, and a capacitor ("Debounce Circuit 3") and this one does not have the LED flash when using it.  However, I can not use that, as I DO NEED a debounce circuit; the LED's all act funny when the tactile switch is pressed.

SO, I've at least isolated the problem to the Debounce Circuit now.  Now, any ideas how to fix it?

EDIT: that should be a 0.1uF capacitor. Mistake there.
« Last Edit: December 27, 2014, 06:54:19 pm by Falcon69 »
 

Offline Falcon69Topic starter

  • Super Contributor
  • ***
  • Posts: 1482
  • Country: us
Re: Circuit still sends voltage surge through when plugging into Power Supply
« Reply #66 on: December 27, 2014, 07:33:12 pm »
Maybe I can do this.

It would only add 1 more component (the other 0.1uF cap, the inverter comes in a dual chip, the same size), And this would allow the inverter to default to ground and only have power supplied to the input WHEN the switch is pressed.

I think if I don't have the Schmitt Trigger of the Inverters, I would have a problem with switch debounce still, otherwise I would just eliminate the inverters.

What do you think?

 

Offline eetech00

  • Regular Contributor
  • *
  • Posts: 64
Re: Circuit still sends voltage surge through when plugging into Power Supply
« Reply #67 on: December 27, 2014, 09:23:26 pm »
Hi

In your debounce circuit (attached)...Try changing R6 from 1M to 82k
See if that helps.

But I would have used these values:
R6 = 82k
R5 = 18k
C4 = 1uf, 25v

that would give enough debounce time to cover just about any crummy switch.

eT
 

Offline Falcon69Topic starter

  • Super Contributor
  • ***
  • Posts: 1482
  • Country: us
Re: Circuit still sends voltage surge through when plugging into Power Supply
« Reply #68 on: December 27, 2014, 10:13:16 pm »
The Debounce circuit works, it's just that when plugging in the voltage source (i.e. 12volt walwart), a voltage spike happens and the debounce circuit causes the LED's to Flash.

That's why I was thinking that if I take away the power supply to the input, as in the schematic with the two inverters, it should eliminate that problem, I hope. It should make the inverter's input default to a low state as the only connection it would have is ground.
 

Offline eetech00

  • Regular Contributor
  • *
  • Posts: 64
Re: Circuit still sends voltage surge through when plugging into Power Supply
« Reply #69 on: December 27, 2014, 11:22:09 pm »
The Debounce circuit works, it's just that when plugging in the voltage source (i.e. 12volt walwart), a voltage spike happens and the debounce circuit causes the LED's to Flash.

That's why I was thinking that if I take away the power supply to the input, as in the schematic with the two inverters, it should eliminate that problem, I hope. It should make the inverter's input default to a low state as the only connection it would have is ground.

During and after power up, the inverter's output should remain low...right?

I think the voltage isn't rising fast enough at the input to the Schmitt inverter.
Basically, the Schmitt is initializing faster than its input, so you see a momentary flash.

You don't need another circuit...
Try changing the 1M resistor to something much lower...like 82k (or even 100k)

eT
 

Offline Falcon69Topic starter

  • Super Contributor
  • ***
  • Posts: 1482
  • Country: us
Re: Circuit still sends voltage surge through when plugging into Power Supply
« Reply #70 on: December 27, 2014, 11:33:08 pm »
Okay, I will try that right now.

EDIT:  Nope, changing to 100k did not fix it.  I will try 47K, I don't have anything between that.

EDIT AGAIN: Nope, changing to 47k or 10k did not fix it either.

Thanks for the suggestion eetech00, it was worth a try.  :-+
« Last Edit: December 27, 2014, 11:46:53 pm by Falcon69 »
 

Offline eetech00

  • Regular Contributor
  • *
  • Posts: 64
Re: Circuit still sends voltage surge through when plugging into Power Supply
« Reply #71 on: December 28, 2014, 12:17:07 am »
Okay, I will try that right now.

EDIT:  Nope, changing to 100k did not fix it.  I will try 47K, I don't have anything between that.

EDIT AGAIN: Nope, changing to 47k or 10k did not fix it either.

Thanks for the suggestion eetech00, it was worth a try.  :-+

Ok...but I still think something similar what causing the momentary flash. Somewhere there is a input that should be initialized high, to keep an output low, is not initializing fast enough.

Using your debounce circuit and values, I was able to duplicate the flash. Changing the resistor value fixed it which makes perfect sense. So this same issue is probably happening somewhere else in your circuits.

eT
 

Offline Falcon69Topic starter

  • Super Contributor
  • ***
  • Posts: 1482
  • Country: us
Re: Circuit still sends voltage surge through when plugging into Power Supply
« Reply #72 on: December 28, 2014, 12:23:15 am »
that's what I thought as well, but, this single LED is based of that Debounce circuit, and it too is flashing. i.e. OR Gate A input tied to inverter's out, and OR Gate B input tied to ground, LED and resistor on output of OR Gate.  That LED is flashing, along with all the other LED's in circuit controlled by other Logic Gates.
 

Offline eetech00

  • Regular Contributor
  • *
  • Posts: 64
Re: Circuit still sends voltage surge through when plugging into Power Supply
« Reply #73 on: December 28, 2014, 04:03:50 am »
that's what I thought as well, but, this single LED is based of that Debounce circuit, and it too is flashing. i.e. OR Gate A input tied to inverter's out, and OR Gate B input tied to ground, LED and resistor on output of OR Gate.  That LED is flashing, along with all the other LED's in circuit controlled by other Logic Gates.

Try this...just to test the theory..

If you can...connect the input of the inverter directly to +5 volts...with nothing else connected to the inverter input. Connected this way, there is nothing slowing down the voltage rise on the inverter input. Then connect/disconnect supply power while watching the LED. The inverter output should never go high and the LED should never light or flash.

eT
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf