Author Topic: problems controlling voltage spikes when connecting voltage  (Read 1447 times)

0 Members and 1 Guest are viewing this topic.

Offline frichTopic starter

  • Newbie
  • Posts: 4
  • Country: dk
problems controlling voltage spikes when connecting voltage
« on: February 15, 2020, 07:33:11 pm »
Hello

i am new here, and quite new to electronics. as such i am still pretty much at the level of standard passive components, and just play abit with timers and such.
i have a made a small project where i am connecting a small router to the a plug that contains 24Vdc and an ethernet connection. the small router i am using is running on 5Vdc (usb) so i have a small buck converter inbetween to step down the voltage from 24 to 5.

unfortunaly i keep burning up buck converters at random, some times they last 3-400 times, and some times they only last a couple of times.
from what i have been able to figure it, it must be due to a spike during when i plug it in. it is allways at the point of plugging it in that the converter gives up.

i have installed a diode on the negative to protect from reverse polarity connection (1N4007), and i tried installing a zener diode across the supply to cap the spike (1N5360BG).
this also seems to fix the positive spike, but i still get a negative spike from time to time.
i have tried multiple buck converters (all cheap chinese ones) and it is allways the IC on them that blows up, not sure if you can use the markings on them for anything (JWBV 045F, AGCH, 1HRG, MP1584EN, CXW8509) i have tried to look up a datasheet, but without much luck :(

so the big question is, how do i create a proper filter to protect the buck converter or maybe someone can point me towards a better buck converter.
 

Offline Vovk_Z

  • Super Contributor
  • ***
  • Posts: 1478
  • Country: ua
Re: problems controlling voltage spikes when connecting voltage
« Reply #1 on: February 15, 2020, 09:10:59 pm »
For example, use aluminium electrolites with high enough ESR (at inputs and outputs). They will dump spikes.
 

Offline frichTopic starter

  • Newbie
  • Posts: 4
  • Country: dk
Re: problems controlling voltage spikes when connecting voltage
« Reply #2 on: February 15, 2020, 09:30:35 pm »
sound reasonable :)

but not quite sure how to calculate that.

if i place something like a 10uF 63v cap parallel across the zener diode, would that be enough ?
found one here with an internal resistance of 2,3ohm
circuit at that point is 24Vdc nominal, with a current draw of 0,03A

terribly sorry about the seemingly stupid questions :( but i am used to working from 100V to 36kV
 

Online David Hess

  • Super Contributor
  • ***
  • Posts: 17429
  • Country: us
  • DavidH
Re: problems controlling voltage spikes when connecting voltage
« Reply #3 on: February 16, 2020, 11:22:52 am »
If the input capacitance is low, then applying power quickly, through a mechanical connection or transistor, can cause any series inductance in the wiring to produce a ringing waveform which exceeds the absolute maximum input voltage of the device.  One solution is to add enough bulk input capacitance, typically 10 to 100 microfarads per amp, to absorb the energy.

Alternatively if the module uses a solid tantalum input capacitor, its surge rating could be exceeded.
 

Offline Deni

  • Regular Contributor
  • *
  • Posts: 70
  • Country: hr
Re: problems controlling voltage spikes when connecting voltage
« Reply #4 on: February 16, 2020, 11:41:33 am »
Had similar issues on a project few years ago. Inductance of power supply wires caused spikes, so I added TVS diode in parallel to the input (with rating above normal operating voltage, but below max. input
voltage rating of DC/DC converter - 33V in mycase) and, just in case, added 100uF/35V electrolytic cap. No issues afterwards.
 

Offline frichTopic starter

  • Newbie
  • Posts: 4
  • Country: dk
Re: problems controlling voltage spikes when connecting voltage
« Reply #5 on: February 16, 2020, 05:30:37 pm »
i think i will try to make some tests with different caps to see if i can reduce the spikes.

i am quite constrained with space, so realy have to be as small as possible.

just in case that doesnt work, would it be possible to make a custom buck converter, that would be able to handle up to something like 60V input ?
and if so is that something that requires alot of knowledge to even start to look at ? maybe one of you have a good suggestion on a voltage regulator to use ?
 

Online David Hess

  • Super Contributor
  • ***
  • Posts: 17429
  • Country: us
  • DavidH
Re: problems controlling voltage spikes when connecting voltage
« Reply #6 on: February 16, 2020, 07:28:36 pm »
just in case that doesnt work, would it be possible to make a custom buck converter, that would be able to handle up to something like 60V input ?
and if so is that something that requires alot of knowledge to even start to look at ? maybe one of you have a good suggestion on a voltage regulator to use ?

There are buck regulators intended to operate in an automotive environment which support 60 volt surges.  Or a series transistor could be added to the input.  The common solution if a capacitor is not suitable was suggested by Deni; add a suitable TVS.
 

Offline frichTopic starter

  • Newbie
  • Posts: 4
  • Country: dk
Re: problems controlling voltage spikes when connecting voltage
« Reply #7 on: February 16, 2020, 07:37:32 pm »
okay

please enlighten me thou, what is a TVS ?
better yet, is the a list of common abbreviations ? took me quite a while to track down ESR earlier :)
 

Offline Deni

  • Regular Contributor
  • *
  • Posts: 70
  • Country: hr
Re: problems controlling voltage spikes when connecting voltage
« Reply #8 on: February 16, 2020, 07:49:46 pm »
TVS == Transient Voltage Suppressor. You can visualize it as a very fast Zener diode. Also, they can absorb much higher amount of energy than Zeners.
See more here https://en.wikipedia.org/wiki/Transient-voltage-suppression_diode
 

Online David Hess

  • Super Contributor
  • ***
  • Posts: 17429
  • Country: us
  • DavidH
Re: problems controlling voltage spikes when connecting voltage
« Reply #9 on: February 18, 2020, 12:49:30 am »
TVS diodes are exactly like zener or avalanche diodes because that is what they are.  They are not any faster but are processed to have a uniform junction like avalanche rectifiers so that at high currents, hot spots do not develop.  The junction is suppose to breakdown uniformly.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf