Pololu Robotics & Electronics
My account Comments or questions? About Pololu Contact Ordering information Distributors

LC voltage spike app. note



I suspect that your problem is not being caused by an LC voltage spike (if it was, the symptom would be problems with your load and you would want to put your capacitor on the other side of the switch as close to your load as possible). Rather, it sounds like your CFL transformer is drawing a large initial burst of current, which is straining your power supply and causing voltage to drop and your computer to reset. In that case, putting a large capacitor on the supply side of the switch (where you have it in your diagram) can help. I think you should try a larger capacitor; if you continue to have problems, you probably need a more powerful power supply.

By the way, just so there is no misunderstanding, is the 1mF in your diagram referring to milli- or microfarad? I have seen microfarad represented as mF before, and larger capacitors are typically represented in units of microfarads (e.g. 2200uF).



Thank you for the help and info! The problem was indeed a large initial current draw. Using a bigger capacitor did reduce the occurrence of the crash a bit, but interestingly, replacing the rocker switch I was using with a push-button switch eliminated the crashes entirely.
The capacitor I was using was indeed 1000uF. I now have a 3200uF capacitor connected right before the switch.
Again, thanks for the help!


jan, Thnks for sharing this Information.
I am facing same problem while I am using the relay to charge and discharge capacitor bank. Charging voltage is 80V DC and current 5 to 10A DC. While i am trying to monitor the voltage with my controller and voltage monitor IC, IC and controller gets burnt.Please suggest solution.


Hello, swami.

Without knowing about more about your setup, it is hard to say whether your problems are related to LC spikes. Also, 80V at 10A is pretty high and potentially very dangerous to work with, and especially since things are already getting damaged, I strongly suggest that you find someone knowledgeable with high power electronics to help you if you are not completely sure of what you are doing.

- Jeremy


Actually my task is to monitor super capacitor board, charging-discharging voltage and current. For few cycles of iteration it works fine. Even though I have provided zener to to clamp voltage at input pin of voltage monitoring IC. Still after some operation IC gets damaged.


Lots of good pictures of how NOT to do it.

How about a circuit diagram and photograph of how to do it right?

Thanks :smiley:



Thanks for the suggestion; we’ll try to make it clearer. The main solution is, if you can, stick a large electrolytic cap across your power input, close to your ceramic capacitor.



I have some LED motion sensor lights that use 4 D size batteries that I use at a cabin, they draw 143mA. I want to hook these up to my 24v battery bank. I ordered some D24V5F6 voltage regulators (6V, 500mA Step-Down). I tested the lights using the voltage regulators with a bench top power supply and used something like 100v 50μF electrolytic capacitor to protect the regulator. The lights and regulator worked fine. My cabin is connected to my batteries with 2 sets of 10 gauge cables run in parallel about 140 feet long (I needed larger than 10 gauge but already had a bunch of 10 gauge landscaping wire so I just double it up). My question is how large of a capacitor do I need if I’m connecting these voltage regulators to these long large cables? Or even better how to I calculate it?



Hello, Nick.

We are missing some details about your system, like why you are using such huge cables and what you are powering with your battery bank, so it is difficult to determine how much capacitance you will need. You will probably have to test the regulator and LED motrion sensor lights in your system to determine what works. You might start with a few hundred uF.

- Jeremy


The battery bank is four 6v deep cycle batteries connected in series to power lights in my cabin, which is 140 feet from my battery bank – if you’re asking why I just don’t move the batteries closer to the cabin it’s because the cabin is in the trees and the batteries are next to the solar panels which are in a clearing in the sun. Two 10 gauge (5 mm^2) wire equates to about a 7 gauge wire (about 10 mm^2). So with 2% lose I can still only push 2 amps at 24vdc or a whopping 48 watts. Anyway there is a path along which this cable runs that I’m going place these motion sensor lights, each of which pulls about 143mA.



It is unclear to me how you calculated 2A for 2% loss in your system. With a single 10 gauge wire, the resistance of the wire is about 1mΩ/ft, so with a 280ft cable (down and back), the voltage drop across the cable flowing 2A would be about 0.56V, which is about 2.3% loss. Also, where is the 2% threshold coming from? With something like our regulators, you get some of that efficiency back when their input voltage is lower. Anyway, it sounds like you already have it all installed, and the fact remains that the bulk of your system power is going to the lights in your cabin, and we are not sure how they would interact with the cables and what kind of voltage fluctuations there would be. Still, a few hundred uF at the inputs of the regulators will probably be fine.

- Jeremy


I used this wire size calculator:


I’ll try using 470uF @ 100v. I don’t have an oscilloscope to see if there are any voltage spikes or anything so we’ll just have to see if they fry or not. Thanks for your help.



Good luck on your project. Feel free to let us know how it goes.

- Jeremy



There is something i did not really understand.
I have also followed this doc

explaining why as low as possible ESR is better for bypass capacitors and you also say in your tuto that “extremely low ESRs are generally a benefit”

but you add :
“but do very little to dampen LC oscillations”

so there seems to be two phenomena that need to be reduced :

  • LC oscillations requiring large ESR
  • something else * requiring low ESR

The article above clarifys a little bit this something else

  • The first line of defense against unwanted perturbations on the
    power supply is the bypass capacitor. A bypass capacitor
    eliminates voltage droops on the power supply by storing
    electric charge to be released when a voltage spike occurs. It
    also provides this service at a wide range of frequencies by
    creating a low-impedance path to ground for the power supply.

It’s not so clear for me what distinguises the two phenomena and why they require completely opposite (conflicting) properties of the capacitors (Low vs High ESR)




ESR limits the current in and out of the capacitor. Low ESR allows the capacitor to give a lot of current to the local circuit if it needs it, which is generally good. (The “something else” “pheonomenon” in your second bullet point is your circuit not working because it did not get the power it needs.) However, low ESR means that the capacitor can also draw a lot of current when you initially apply some voltage to it, and when you do that with long wires, that’s where you add inductance to the system and get the oscillation problems. You don’t “require” large ESR to deal with that; there are several other things you can do. It’s just that low ESR does exacerbate the problem, and if you don’t need it that low, adding some back can be the easiest way to balance keeping the oscillations good enough while also keeping the power to your circuit good enough.



Thank you,

I recently bought this 6V 500mA polulu regulator

and use it to feed an arduino from a Lipo 4S (16.8 V max) in a quadcopter.
At the end of your page there is the recommandation:

“If you are connecting more than 20 V or your power leads or supply has high inductance, we recommend soldering a 33 μF or larger electrolytic capacitor close to the regulator between VIN and GND. The capacitor should be rated for at least 50 V.”

Though my Voltage is lower than 20V, i’m still wondering if with ~ 15cm leads (typical breadboard wires) it would be more cautious to add the electrolytic capacitor in parallel with the polulu regulator … ?



For your system, you probably do not need to add an electrolytic capacitor, but adding one would not hurt either, so if you have space for it, you might add it just as a precaution.



I have a Pololu 6V 2.5A step down regulator D24V25F6. VIN connects to 12V solar panel and OUT is connected to a CN3065 lithium battery charge module using a 2.3 metre cable.

In my project, I am finding that the CN3065 boards are experiencing problems after a few weeks of use. Normally the CN3065 would stop charging the lithium battery once the voltage reaches 4.2v. But after a while the boards are supplying higher voltage (approx 5v) to the battery. This is a problem as these batteries should not be charged higher than 4.2v.

The datasheet for the CN3065 states that the maximum input voltage is 6V which the Pololu regulator is providing, so I am wondering if voltage spikes on powerup could be damaging the CN3065. There is a mechanical switch near the lithuim charge module to switch power on/off.


It is always a risk to run at the “maximum input voltage” of a chip. The slightest spike on the input will exceed that voltage.

I suspect that your use of 6V input to the charger is the major cause of the problem, and suggest to use a 5V regulator instead.


A post was split to a new topic: Problem with electrical fault in UPS installation