[introduction] the switch movement of MOSFET produces alternant cycle, among them electric current flows into inductor above all, next inductor discharge. This meeting brings about big DI/dt and big voltage aiguille. We expect this kind of noise. The problem is LC filter there is many in the others field that avoids these big voltage aiguille to transmit circuit effective.

With respect to its property character, the output end of NSMPS can have noise of a few switch. After all, their design is used at using pulse width modulation (or pulse frequency modulation) the electric current that signal switch comes from taller direct current source, use next 2 extremely LC filter has filter wave to its.

The switch movement of MOSFET produces alternant cycle, among them electric current flows into inductor above all, next inductor discharge. This meeting brings about big DI/dt and big voltage aiguille. We expect this kind of noise. The problem is LC filter there is many in the others field that avoids these big voltage aiguille to transmit circuit effective.

The typical output voltage of SMPS will be in indication ripple in switch frequency. A main index is when manostat does not have load and the typical and laden resistor in applying when its to load from time to tome how many ripple.

Measure the noise in switch mode power source

I have application of a low noise nearly, I want to try to use cost low-down 3.3 V SMPS; Need the load current of 50 MA only. I have to evaluate board, I join my power source of type of 5 V wall uses a simple 10 × probe to measure output. If the graph is shown 1 times,my measurement is configured.

Graph 1. Use 10 × probe to measure output voltage course.

DC n is firm when 3.3 V just. Depend on ability of 12 resolution of oscillograph of my Teledyne LeCroy HDO 8108 and big deflection, I can offset this voltage, so that I am OK,enlarge ripple noise is searched slow drift of fast signal dc. Graph 2 showed the voltage noise that measures on 10 MV/div scale.

Graph 2. The SMPS that uses 10 times probe to be measured on the scale of 10 MV/div outputs noise.

Switch implement 20 small seconds are periodic (corresponding the switch frequency at 50 KHz) very apparent. Triangular pulse predicts those who come from inductor electric current to charge with discharge cycle. But, besides the feature that anticipates besides this, still have the high frequency noise of two kinds of types. Even area has the noise of peak peak value of 10 MV, acerb aiguille noise can rise to the peak peak value of 60 MV sometimes.

High frequency noise and acerb noise aiguille are disturb. This is done not have by 2 extremely LC filter filters. If I use this power source, how does my general ensure my circuit board the function that can you still keep enough below all circumstances of these noise?

However, the fact proves, this kind of noise is not the voltage noise that power source output carries actually. This is the radio frequency signal that my probe receives.

Divisional voltage noise and RF collect

The big DI/dt of the inductor in passing LC filter can be brought about big magnetic field produces around SMPS. Any annulus ways that have low inductance way can produce magnetic induction electricity, these electric current can produce the tension that we measure with oscillograph.

I made even the 10 × probe that receives SMPS down-lead an annular antenna, can receive these aiguille signal. Your think of a way is possible be, is of 10 × probe the resistance that there is Ω of a 9 M? Is this one can prevent to give the big impedance of any alternating current in the induction in annulus road?

Have resistor of Ω of a 9 M, but also have shunt-wound capacitor of a 10 PF, high frequency electric current sheds the one part of the equalizer circuit of classics. When 100 MHz, the impedance of 10 PF capacitor is 160 Ω only, low breathtaking.

To check noise of wh some of which it is the RF collect in probe actually the idea that is not the actual noise on power source course, the output that I receive connector solder of a small-sized SMA circuit board is carried in order to reduce annular antenna area and to radiant susceptibility domain. In addition, I added probe of another 10 × around the probe that measures SMPS to output voltage, but of the 2nd probe with ground wire short circuit. This kind of setting makes I can use 10 × probe to measure the output course, output course that passes SMA connector and noise of this locality RF at the same time (by the probe collect with ground wire short circuit) . If pursue,3 are shown.

Graph 3. Use probe of two 10 × and coaxial the voltage noise that 1 × join will come to to measure SMPS output to go up.

Graph 4 showed the noise that uses these 3 kinds of methods to measure.

Graph 4.  The measurement on SMPS output voltage. All passageways all use same 10 MV/div scale.

Probe attenuation affects SNR

Have two important observation results. Above all, the general noise level of 1 × coaxial cable is far under 10 × probe. Because 10 × bougie is not 10 × bougie,this is actually, however 0.1 × bougie. It signal attenuation 10 times, its extent reduces 20 DB. When we measure small signal n, for example a few millivolt, measures voltage is very sensitive to the amplifier noise of oscillograph.

Most oscillograph is enough intelligence, can identify the link on the passageway probe of a 10 × . They are met the voltage scale that self-correcting shows shows tension with compensating decuple attenuation. Accordingly, when oscillograph shows signal with 10 MV/div scale, it is actually 1 MV/div scale is used on amplifier. We are in see almost noise of peak value of 10 MV peak is oscillograph amplifier actually on about noise of peak value of 1 MV peak.

Cable of coaxial of use SMA connective is probe of a 1 × actually. This trace also shows with the scale of 10 MV/div. Below this kind of circumstance, noise of amplifier of peak value of 1 MV peak includes the line in trace more or less wide inside.

This watch understands to measure practice importantly: Examine low extent signal when us (for example noise of power source course) when, probe of any 10 times attenuation can reduce our SNR 20 DB. Become when every DB is very important, do not use attenuation probe.

Coaxial join and oscillograph probe

The 2nd observation is coaxial cable as a result in join nonexistent big and acerb aiguille, but probe of two 10 × measures consist in in. As a result of among them a probe is outputted without course of bring into contact with even, this shows acerb aiguille noise is caused by RF collect strongly, is not the voltage noise on SMPS output.

This watch understands the 2nd to measure practice importantly: When measuring low extent signal, make exhaust the likelihood approachs measurement unit of coaxial cable connective, in order to reduce the annulus road area of probe and the effectiveness that its regard aerial as.

If we carry out these two to measure practice, we will obtain the noise of ripple of peak peak value of 30 MV in 3.3 V course. This is the ripple of 1% , to low cost SMPS perfectly. In addition, high frequency noise is reduced greatly, and short when twinkling changes (the form that actually with RF collect noise is not course voltage noise exists) show no longer for switch implement the one part that exports signal.

The noise in frequency domain

Want me to use ground level in the place of the power source that stands by me and signal method only, this is an important design practice, the signal that goes up by this the equipment of SMPS power supply and my board will see the harmonic switch power source of the 50 KHz ripple that arises by SMPS only.

Use connection of direct coaxial, low noise, I measured the audio spectrum of a confusion of voices on the power source course that comes from SMPS. Graph 5 showed a give a demonstration.

Graph 5. The audio spectrum of a confusion of voices on power source course. Coping is when chart of frequency conversion table, exceed 10 seconds, show very stable amplitude. On this scale, 0 DBmV is noise of 1 MV amplitude.

The peak value in spectrum is the 50 KHz harmonic of switch frequency. The extent of harmonic is 10 DBmV about, namely 3 MV. This is far the voltage of peak value of 30 MV peak that in be less than time domain, measures. Because,this is of ripple noise occupy sky comparing very low. There is wave of too much sine in pulse of the triangle when what harmonic is in is short. The peculiar form that many Gao Cixie wave demonstrates time domain is undee reachs his high frequency composition.

Be in about when 3 MHz above, the amplitude of all switch noise all under V of 10 μ . To my application, this is an acceptable noise level, actually it is low-down for the SMPS to a such low cost.

Conclusion

The article discussed the serious consideration factor of the voltage noise that produces actually about switch mode power source, put forward two kinds to measure practice, the oscillograph that can help you undertake to the output course of switch manostat is measured.

Avoid duty statement: The article is reprint an article, reprint this article purpose to depend on passing more information, the person that copyright puts in original work ‘s charge is all. If involve work copyright issue,article place uses video, picture, written language, contact please small make up undertake handling.