EEVblog #594 - How To Measure Power Supply Ripple & Noise

Fundamentals Friday
Dave explains what the ripple and noise specifications on a power supply is and how to measure it using different methods on both analog and digital oscilloscopes. From bad techniques through to good, showing the effect of each one. Traps for young players aplenty in this one.
How do you detect common mode noise issues and ensure that the signal you are measuring is really coming from your device under test?
Single ended & differential measurement, DIY coax solutions, termination, analog vs digital oscilloscopes, bandwidth limiting, and even oscilloscope probe coax construction issues. It's all here.
Mysteries of X1 oscilloscope probes revealed:
How to track down common mode noise:
Opamp Noise voltage tutorial:

EEVblog Main Web Site:
EEVblog Amazon Store:
Donations:
Projects:
Electronics Info Wiki:

[andymouse]

6 years on and your 'deep dive' stuff is still pure gold, I wonder if the UNI's around the world use this in their curriculum..they should..cheers.

[jarrodhroberson]

Ignore anyone complaining about content being to long. youtube is littered with videos that dont' actually explain anything in any useful detail. You rock! I am a software guy for 30 years, between you, Mike, Bruce Simpson, Jeri Ellsworth and the young guys like Kevin Darrah and Jeremy Blum I am actually understanding the hardware that powers what I have been doing for a living since the 80's!

[Sunnyskyguy]

Dave, you may think 20MHz is an arbitrary frequency, but it is also the point where poor probing methods with long inductive probe ground leads start to cause extra ringing on fast logic causing measurement errors.

So for clarity 20MHz is for convenience when risetime induces false ringing from poor probing skills and gets a better textbook looking trace.

For ideal trace remove probe tip and ground clip and only use tip and probe ring to a pair of test pins or wires of similar spacing for convenience or use the coiled ground spring accessory . This gives the best results using a 10:1 probe.

Generally signals much less than 30MHz are used for this mode but logic rise times faster than 100ns ( most are) will get an overshoot proportional to the ground probe length and transition frequency resulting in ringing..

Tony EE since 1975

[eded]

no book could have given such memorable clarity on these pitfalls.... excellent video Dave and the best part is its a free video )))

[JanPedersen]

Totally agree with Jens Andree. These kind of videos actually makes a difference in the world for a lot of people. Thx Dave....more of this stuff :)

[ftester]

For me one of the best videos from the EEVBlog series. Cannot really understand why two guys gave a thumbs down...

[BradMorrisKAYAN]

I've been wondering for months what a CRO probe was.  True, I didn't look it up, but still wondered.  Cathode Ray Oscilloscope (CRO) probe.

Brilliant!

[electronicinvestigations]

Thanks Dave, keep up the Fundamentals videos. I can't wait to get out to my lab and try this with my own scope..

[ronmuller]

Excellent video. Instrumentation and measurement and understanding the limits of your equipment and sources of errors are very important and often a poorly understood area. Nothing like a real demonstration to highlight the concepts and pitfalls. Good stuff Dave.

[INCYTER]

Dave, I really like your channel. Outstanding quality topics, content, judgement and sharing of your understanding and tricks. Thank you for sharing your excellence and passion.

[FractalY]

Wonder if you could do a video on ground loops? Thanks great channel!

[exce]

Thanks for doing the fundamentals. These are my favorite videos.

[roboknight]

Dave, really awesome.  Well worth the time.  Thanks for going through this one.  I can't tell you how much the EEVBlog comes up here at work lately.  Too many geeks with not enough to do? ;) ... Love it.

[discoHR]

Measuring the ripple is so much easier using a True RMS multimeter. Set it to AC voltage, read the result, done. However, using a scope is more fun.

[AllElectronicsGr]

Why the 50ohm terminators made so much diference? 

[oriole]

I got some Lambda LQ/LQD series linear bench PSUs recently, circa 1983 (warranty seals were unbroken haha!).. their ripple + noise specs are 0.15mV RMS, 1mV pkpk. Temp coeff. is ±(0.005% + 10μV) /°C. Line/load regulation is 0.005% + 0.5mV. Current regulation is 1mA/2.5mA (line/load), at full voltage range. Just fantastic. I verified those figures with my scope, still fully within specs. Zero overshoot with maximum load. They employ a custom Lambda branded IC for V/I regulation. I've sourced a replacement for one from China, gonna see if the specs change after I put it in.. I suspect it's not an original part from 31 years ago. :P

[richwiskin]

Brilliant video. So useful. Perhaps I shouldn't be so disappointed with my digital scope after all! Thank you.

[stevenhardy]

There is another aspect of power supply noise that is little thought of. I was involved with a huge project at a large computer site in 2010 that had to have the power supplies replaced on 5000 installed servers. The switching power supplies also  generate noise back onto the ac feed line. When you had 5000 servers with no isolation, all generating noise, the electrical noice would shut down the site UPS systems. The replacement power supplies, all 10, 000 of them fresh in from China, now had toroid coils on the ac line. Guess the server vendor didn't save 20 cents per server after all ?

[Switching-powerBlogspot]

Excellent because it shows the tips&tricks of ripple and noise measurement. 

[Ronb]

It's everyday friday here! thanks Dave for your lessons!