Author Topic: Circuit still sends voltage surge through when plugging into Power Supply  (Read 18644 times)

0 Members and 1 Guest are viewing this topic.

Offline Christopher

  • Frequent Contributor
  • **
  • Posts: 429
  • Country: gb
Re: Circuit still sends voltage surge through when plugging into Power Supply
« Reply #25 on: December 26, 2014, 11:52:53 pm »
If the output is being switched to 24v as it turns on them clamps to 5v then it's not good at all... suppose the only way you'll tell is with a scope or dmm in peak but that might not pick it up.
 

Offline Falcon69Topic starter

  • Super Contributor
  • ***
  • Posts: 1482
  • Country: us
Re: Circuit still sends voltage surge through when plugging into Power Supply
« Reply #26 on: December 27, 2014, 12:00:49 am »
I guess I don't understand.  I'm trying to follow in what that schematic is doing, but I'm just confused.  My chip does not have Sense + or sense -.

I don't want to change the lm2596, as I've already bought alot of them.

 

Offline tautech

  • Super Contributor
  • ***
  • Posts: 28136
  • Country: nz
  • Taupaki Technologies Ltd. Siglent Distributor NZ.
    • Taupaki Technologies Ltd.
Re: Circuit still sends voltage surge through when plugging into Power Supply
« Reply #27 on: December 27, 2014, 12:07:41 am »
You are really just pissing in the dark without a scope.
Nearly anything with 2 channels will show what you want to see.

Get something, anything and explore another world.  ;)
Avid Rabid Hobbyist
Siglent Youtube channel: https://www.youtube.com/@SiglentVideo/videos
 

Offline Falcon69Topic starter

  • Super Contributor
  • ***
  • Posts: 1482
  • Country: us
Re: Circuit still sends voltage surge through when plugging into Power Supply
« Reply #28 on: December 27, 2014, 12:28:55 am »
What about using an P-Channel Mosfet or PNP transistor on the output of the LM2596, like right after the 1uH inductor (L2) on my schematic?  With a cap and resistor on the gate?  That way, the cap will take time to charge to reach the voltage, turning on the transistor.  That would give the LM2596 enough time to properly switch on.  That would use a lot less components than the circuit you had posted Christopher.
 

Offline Christopher

  • Frequent Contributor
  • **
  • Posts: 429
  • Country: gb
Re: Circuit still sends voltage surge through when plugging into Power Supply
« Reply #29 on: December 27, 2014, 12:31:14 am »
I'll do a couple of simulations in the morning for you but if you could post an image of your pcb that'd be good. Doubt it's a layout problem as these chips are pretty resilient.
 

Offline Falcon69Topic starter

  • Super Contributor
  • ***
  • Posts: 1482
  • Country: us
Re: Circuit still sends voltage surge through when plugging into Power Supply
« Reply #30 on: December 27, 2014, 12:35:23 am »
I would appreciate that Christopher.

I will send you a pic of the board.

I was also thinking maybe something like this using a low side transistor or mosfet to switch it on.

Would something like this work?  I can adjust the regulator to put out 6 volts, this should offset the voltage drop across the transistor.

 

Offline Christopher

  • Frequent Contributor
  • **
  • Posts: 429
  • Country: gb
Re: Circuit still sends voltage surge through when plugging into Power Supply
« Reply #31 on: December 27, 2014, 12:57:23 am »
Can you take the circuit back to normal, remove r3 and short the on pin to ground then see what happens?

Then again but removing the cap between the feedback and switching pins
 

Offline Falcon69Topic starter

  • Super Contributor
  • ***
  • Posts: 1482
  • Country: us
Re: Circuit still sends voltage surge through when plugging into Power Supply
« Reply #32 on: December 27, 2014, 01:44:29 am »
Yes, I had done that.  It still does it.  I made pin 5 directly connected to ground. 
 

Offline Andy Watson

  • Super Contributor
  • ***
  • Posts: 2074
Re: Circuit still sends voltage surge through when plugging into Power Supply
« Reply #33 on: December 27, 2014, 01:48:28 am »
as you can see from the circuit, I added the enable pin circuit which was suppose to be a slow on start that I thought would prevent all my LED's in the circuit from turning on for a split second when the circuit board was plugged in.
What is controlling the LEDs?  Is it the same logic that is being powered by this supply circuit? If it is, then the problem is happening when the +5V becomes active. As has been pointed out, the enable pin is an On/Off switch - not a slow start. No amount of delaying the start-up is going to stop the LEDs flashing when the power supply output is activated. Need to be looking elsewhere to cure the problem.
 

Offline Falcon69Topic starter

  • Super Contributor
  • ***
  • Posts: 1482
  • Country: us
Re: Circuit still sends voltage surge through when plugging into Power Supply
« Reply #34 on: December 27, 2014, 02:15:30 am »
yes, that is what we are trying to track down now Andy.  I was hoping for an easy fix, but doesn't seem so.
 

Offline Andy Watson

  • Super Contributor
  • ***
  • Posts: 2074
Re: Circuit still sends voltage surge through when plugging into Power Supply
« Reply #35 on: December 27, 2014, 02:38:41 am »
Ah, I must have missed a bit of the thread ;)  In that case could you give some more details about the LED drive/control? Are you looking for a positive or negative enable signal ?
 

Offline Falcon69Topic starter

  • Super Contributor
  • ***
  • Posts: 1482
  • Country: us
Re: Circuit still sends voltage surge through when plugging into Power Supply
« Reply #36 on: December 27, 2014, 02:52:23 am »
i need a positive 5.3 volts from the LM2596.  I could just use a 5v walwart instead of this headache, but I'm trying to offset the voltage drop from length of wire that goes to some hall effect sensor switches. 

As said before, the whole circuit works fine, except for when initially plugging in the board.  All the LED's flash for a fraction of a second.  I'm trying to stop that from happening.

If I understood Christopher right, The LM2596 is supplying on the output at plug-in the full supply voltage of the input for a fraction of a second, i.e. 24 volts if it was the supply amount, until the voltage regulator fully switches on, and supplies the 5.3volts.

That could prove fatal to the logic gates and inverters on this board and on the switches.
 

Offline Alex30

  • Regular Contributor
  • *
  • Posts: 197
  • Country: au
Re: Circuit still sends voltage surge through when plugging into Power Supply
« Reply #37 on: December 27, 2014, 02:56:47 am »
Falcon, what voltage are you running your tests at when you plug it into your power supply?
 

Offline Falcon69Topic starter

  • Super Contributor
  • ***
  • Posts: 1482
  • Country: us
Re: Circuit still sends voltage surge through when plugging into Power Supply
« Reply #38 on: December 27, 2014, 02:58:09 am »
right now, i'm using a 12volt walwart plugged into it.  But, the circuit needs to be able to handle between 7-24volts input for the voltage regulator.
 

Offline Alex30

  • Regular Contributor
  • *
  • Posts: 197
  • Country: au
Re: Circuit still sends voltage surge through when plugging into Power Supply
« Reply #39 on: December 27, 2014, 03:07:30 am »
This might seem dumb as I don't fully understand the inner workings of the LM2596 (but then again the slow startup circuit should work in my opinion), would you try removing C4 entirely and short circuit R3 to the voltage supply of the LM2596? In this configuration the reg should never switch on, even partially, if it does then no amount of fiddling with pin 5 will work. I hope we can fix this as I will be using this for some of my work.
 

Offline Falcon69Topic starter

  • Super Contributor
  • ***
  • Posts: 1482
  • Country: us
Re: Circuit still sends voltage surge through when plugging into Power Supply
« Reply #40 on: December 27, 2014, 03:16:02 am »
no, the regulator will not turn on then, as the voltage then supplied to the lm2596's pin 5 will be above its 1.3v threshold. In order for the lm2596 to turn on, pin 5 must have a voltage below 1.3v.
 

Offline Alex30

  • Regular Contributor
  • *
  • Posts: 197
  • Country: au
Re: Circuit still sends voltage surge through when plugging into Power Supply
« Reply #41 on: December 27, 2014, 03:33:31 am »
Exactly, but there are some internal capacitors within the lm2596 - just not sure how it's set up and how it impacts this slow startup circuit. My point being that using the slow startup circuit given in the datasheet should work, given that C4 is a short circuit in initial startup. I'm trying to prove the configuration in the datasheet will never work by asking you to change C4 to a short circuit.
 

Offline Falcon69Topic starter

  • Super Contributor
  • ***
  • Posts: 1482
  • Country: us
Re: Circuit still sends voltage surge through when plugging into Power Supply
« Reply #42 on: December 27, 2014, 03:37:30 am »
As others have enlightened me, the delay start is just that, a delayed start, meaning, it will delay the turn on of the LM2596.  However, it seems the problem I am having is not that, and needs to be fixed using some kind of a Soft Start circuit on the output of the LM2596.  This is where I am confused, but happy to learn.

I am trying to search for 'LM2596 Soft Start' circuit schematics, but I'm running out of luck.  I am a visual person, and I need to see stuff drawn in order to understand it.
 

Offline Alex30

  • Regular Contributor
  • *
  • Posts: 197
  • Country: au
Re: Circuit still sends voltage surge through when plugging into Power Supply
« Reply #43 on: December 27, 2014, 03:50:09 am »
Oh I see. Is undervoltage an issue for any micro controllers or other chips you have in your circuit?
 

Offline Falcon69Topic starter

  • Super Contributor
  • ***
  • Posts: 1482
  • Country: us
Re: Circuit still sends voltage surge through when plugging into Power Supply
« Reply #44 on: December 27, 2014, 03:51:24 am »
all the chips I have (logic and inverters) operate on 2-6 volts.
 

Offline Alex30

  • Regular Contributor
  • *
  • Posts: 197
  • Country: au
Re: Circuit still sends voltage surge through when plugging into Power Supply
« Reply #45 on: December 27, 2014, 04:18:05 am »
Well the simplest slow startup application may be of use here. But I'm not 100% sure if this is the best thing to do here as I am only a hobbyist. Simply a power transistor (perhaps a darlington) with a cap at the gate as shown on the picture. The turn on time can be altered by changing the RC time base to your needs. My thinking would be to use this before the lm2596 then use the under voltage lockout circuit given in the datasheet (attached) to suit your needs.

Things to look out for:
If you have lots of changing loads in your circuit you could cause oscillation by using smaller capacitor values
Voltage drop over the power transistor could increase your power requirement
Power rating of capacitor
 

Offline Falcon69Topic starter

  • Super Contributor
  • ***
  • Posts: 1482
  • Country: us
Re: Circuit still sends voltage surge through when plugging into Power Supply
« Reply #46 on: December 27, 2014, 04:22:57 am »
I was thinking about the same thing, but using a mosFET instead, and AFTER the inductors.  That would be the trace going out to the rest of the circuit.

The problem is the inductor is messing with the feedback for a fraction of a second when the board is plugged in using the walwart or another supply voltage.

If I can somehow delay the current/voltage from going to the rest of the circuit (after the inductors) until the voltage regulator turns on completely, maybe a 1/2 a second delay, then I think everything will work just fine.
 

Offline Alex30

  • Regular Contributor
  • *
  • Posts: 197
  • Country: au
Re: Circuit still sends voltage surge through when plugging into Power Supply
« Reply #47 on: December 27, 2014, 04:29:03 am »
Well it shouldn't be a problem if you set up the lockout to happen at 2.5 volts as there should only be 2.5 volts available on the power rail anyway until the slow start up is fully initialized.
 

Offline Falcon69Topic starter

  • Super Contributor
  • ***
  • Posts: 1482
  • Country: us
Re: Circuit still sends voltage surge through when plugging into Power Supply
« Reply #48 on: December 27, 2014, 04:36:40 am »
i think you are confused.  The input with the lm2596 will always be 7-24v (depending on which walwart is used).  There is no undervoltage here.

I need to control the output somehow, so the current and voltage is delayed until the lm2596 turns fully on, thus, eliminating the problem of the surge current and voltage at plug-in.
 

Offline Alex30

  • Regular Contributor
  • *
  • Posts: 197
  • Country: au
Re: Circuit still sends voltage surge through when plugging into Power Supply
« Reply #49 on: December 27, 2014, 04:43:08 am »
No, my suggestion is to put the slow startup circuit on the Vin side of the LM2596. The voltage at Vin therefore will slowly ramp up. So use the lockout pictured above to ensure that the LM2596 does not turn on until the voltage at vin > 2.5V. At this point, even if the feedback isn't working properly and tries to draw tens of volts, there is only ~2.5V available to the lm2596 anyway, so you won't see any high voltage transients when the lm2596 switches on.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf