+86 400-003-5559 CN


Measuring Capacitors Over Their Working Voltage | Hackaday

tagshv capacitor

Ceramic capacitors are small, they don't leak, they are convenient, but they are strange. Certain types of capacitors lose capacitance based on their operating voltage. If you use ceramic covers for filters, DC to DC power supplies, bypass covers, or any devices that require precise capacitance in the circuit, problems may arise.

[Mathieu] proposed a tool that can measure the capacitance of the cover over its entire working range. He is calling it

, Although the name may be a bit confusing, the function is not. This small box will measure the capacitance of the part in the voltage range of 1.3V to 15.5V.

By connecting SMD tweezers or test clips to the capacitor, OpenCVMeter can increase the voltage and measure the capacitance of the part throughout the test cycle. Then dump these data to the Chrome application (a very popular platform for testing device applications), and display on the screen to determine the ability of the capacitor to work in the circuit

If you have ever used antique electronic equipment, then you will know that the first thing that deteriorates in any piece of equipment is the bottle cap. Either the bottle caps had extremely loose manufacturing tolerances at the time, or they were actually

The key, but a clumsy hat may paralyze everything from tube amplifiers to computers. This is a very concise tool and does not actually exist in a single dedicated device.

I believe that the old but still powerful ESI VideoBridge can actually perform these types of tests.

Not only are electrolytic capacitors prone to aging due to the drying of the liquid electrolyte? (Although all caps may fail for various reasons).

I think ceramic capacitors are a separate issue.

Paper wax will also expire with the passage of time, and many antique appliances are using them. I believe that mica may also fail with age, and when they do, they will also show some very, very strange symptoms, such as not showing their failure immediately or testing well on the meter, but there is a problem when placed When in the circuit and housing, it will not.

Tantalum caps can also fail due to age-related issues, especially after being unplugged for a long time.

When they fail, they will fail quite a bit. Suddenly, I have a big power source. Suddenly, a hot tantalum heat is emitted, and it falls on the carpet. I scramble to find something to pick it up.

So, why should the PSU be removed from the case in the first place?

This is an open power supply.

…Able to test ceramic covers in the entire 1.3v – 15.5v working range. Does the working voltage of the ceramic cover not exceed 50v?

I am always thinking.

The voltage of the micro SMT is lower, so the size is smaller.

There is no problem with the size of 50V 0402. 15v is a big limitation. There should be an external voltage input.

I designed the device to measure capacitance at the most commonly used voltages.

Compared with other ceramic capacitors, some ceramic capacitors have very high capacitance and low voltage. Just like 10uF at 15V. However, the capacitance they bear varies greatly with the applied voltage.

Why not just use electrolytic capacitors, you can get higher capacitance? Since aluminum foil is made by winding aluminum foil into a cylinder, it has high parasitic inductance. ESR is often greater than ceramic capacitors.

The high-value ceramic cover is made of many parallel metal and dielectric ceramic layers, so the parasitic inductance is very low. This is very important for bypass applications such as switching power supplies and circuits with peak-to-valley ratios.

Higher-value ceramic capacitors use ceramic formulations with higher dielectric constants. The disadvantage is that the breakdown voltage is low, the piezoelectric effect makes them detect and emit sound at the same time, the capacitance temperature changes and the capacitance changes with the applied voltage.

I would never use one of these high-value ceramics in any situation that requires strict tolerances. Like active filters or passive filters, in addition to general low-pass or high-pass filtering, more needs to be done.

Is there a ceramic cover with a rated voltage not exceeding 15V? Usually at least 60V.

Many I think 6.3 V is the lowest universal rating.

Yes, 6.3V is quite common, but there are several 4V and even 2.5V types.

High-value MLCCs have low voltage ratings. When you push C up, there is only so much dielectric in a small package. 1/2 CV ^ 2 I saw 4V 10uF in 0603.

I didn’t know that they made ceramics with such a high capacity, thinking it was just electrolysis or tantalum. Technology, right? Yes, I was thinking about the traditional round terminal type, not the compact SMD type.

SUUUUPER is useful. The derating of SMD ceramic capacitors is so large that I would not even consider decoupling capacitors. Through-hole 0.1uF @ 5V decoupling capacitors may require a "1.0uF 0402" @ 5V to similarly filter out 5V power noise. To make matters worse, from most data sheets I have seen (I’m basically just a workshop manufacturer with a rated CV curve now), it’s not even better at higher rated voltages (in fact, CV The curves tend to be the same; you can just put more voltage on it safely)

For the decoupling of digital circuits, packaging and PCB layout dominate the capacitance to achieve high frequencies. Your through-hole part has capacitance, but it will be lost in the long lead, which is the mounting height above the PCB (larger loop area). The 0402 part works better in modern circuits.

This is a very interesting article, which details why this is a very cool device:

Wow... this is a good find, thank you!

What about people who claim to be "open source" but not open source? Someone told me that I was stupid and published this kickstarter design file?

This is what I can find:

Mathieu Stephan

In the next few days, I will write a detailed article on the schematic website.

Interesting project, but I also want a larger voltage range. We work a lot at the 24V and 48V supply voltages of the equipment, so 50V and 100V rated capacitors are common to me. Even the standard 100nF decoupling capacitors in the 0402 package we use are rated at 50V. If we need large-volume products in small packages, then we will only reduce the voltage rise. Well, maybe once they release the hardware design, we can make an improved version in which you can set a higher voltage and reduce some accuracy, or work at a lower voltage to get a higher resolution.

What I would also like to see is an extension of the complete RLC meter, which can give some estimates of the series inductance and resistance in the capacitor. If it is not calibrated correctly, using long test leads may not be so easy.

In fact, you can connect the power supply in series with the leads to increase the voltage.

In the next few days, I will write a detailed article on the schematic website.

Hi, maybe someone needs this kind of measurement every day, in which case this might be a useful tool.

However, you can use any other capacitance meter to make the exact same measurement.

The only thing you need to do is to decouple the DC bias voltage from the measurement circuit.

Option 1: Connect two identical samples in series in the test circuit, and then apply a DC voltage to the intermediate node through a 10 MΩ resistor.

Option 2: If you only have one sample, connect a known second capacitor (or a capacitor with a capacity two or more orders of magnitude higher) and perform the same measurement.

Similarly, the C(V) characteristics encountered under certain conditions of specific ambient temperature and certain mechanical stress are almost useless in applications.

The most important thing in the application is to know the upper and lower boundary values ​​of the allowable deviation from the ideal C (V) characteristic curve. For this reason, if you do not recycle parts, you only need to read the data sheet.

I admit that in most cases, you can infer this characteristic from the measured curve

The capacitor under test can provide isolation. You need an isolated power supply, then lift the ground wire of the capacitor under test and apply a negative bias voltage to it. I only increase the voltage after connecting all the voltages to avoid sudden charging current flowing into the capacitor tester.

Then, the bias voltage is the power supply voltage and the voltage applied by the capacitor tester. In the case of the EDN circuit, it only applies about 600mV plus or minus about half a volt. Therefore, if you want to test under 5V, set the power supply voltage to 4.4V. If you are testing at a voltage of 50V or higher, the slight offset may be ignored.

"In addition, the C(V) characteristics encountered under certain environmental temperatures and certain mechanical stresses are almost useless in applications.

The most important thing in the application is to know the upper and lower boundary values ​​of the allowable deviation from the ideal C (V) characteristic curve. For this reason, if you do not recycle parts, you only need to read the data sheet. "


I should say that I assume that the short line from the power supply provides the bias voltage and the ESR capacitor in the power supply is low. Alternatively, a capacitor with a sufficiently high rated voltage, low ESR and at least 10 times the capacitance of the capacitor under test can be inserted in parallel next to the bias voltage and connected in parallel with the bias power supply.

I think it’s worth repeating that whenever discussing the failure of an old capacitor...

If you have not turned on the power of the old device for a long time, charge the capacitor slowly. Depending on the design of the device, it may be simple, just plug it into a minimal autotransformer and slowly increase it to the line voltage. Or... it may involve disassembling the device and individually powering up specific parts, whose starting voltage is much lower than its normal working voltage, and then slowly turning it on.

In doing so, many of these capacitors will be reassembled... they will not fail. Now, you can open it normally next time.

Say again...they are still veterans and their value may drop slightly. As an alternative method.. Just replace them all before powering up. This prevents accidental damage to other parts when these old capacitors become short circuits (or catch fire).

Either way...don't just plug it in and open it! ! !

Please be kind and respectful to help make the comment section great. (


The site uses Akismet to reduce spam.


By using our website and services, you expressly agree to our placement of performance, functionality and advertising cookies.