Electronics tutorial - Filtering unwanted common mode noise from your oscilloscope measurements
Vložit
- čas přidán 5. 05. 2020
- #72 In this video I revisit one of the problems I have with my equipment, and that is the common mode noise coming from my bench supply. Previously I improved on this by hacking into the power supply, but today I will attempt to remove the common mode noise by an external method, so something that does not involve modifying the equipment.
Previous Power supply improvement video: • How to improve the out...
If you liked this video be sure to check out my other videos and you can also subscribe to be up to date with all the new ones!
If you want to support the creation of more and better videos you can at: / feszelectronics - Věda a technologie
You blew my mind when you demonstrated how shortening length of the wire from P.supply to Oscilloscope can reduce noise dramatically .... nice explanation. Thank you.
This is a great video on a topic few people know much about. Keep doing what you're doing.
Best info I've seen on common mode noise, great job, thanks.
Engineers should watch your videos more and gain more insight in electronics!!
Thanks, FesZ!
Thank you sir. Great simple demonstration.
Great explanation and clear demonstration. Thank you!
I have problem like this lingering to my measurements. Thank you very much for this technique.
This doesn't really eliminate the noise. To do that, connect the "hot" chan 2 input independently to the common at your circuit board, then display chan 1 minus chan 2. This eliminates all external noise, and a stronger ground connection isn't needed. Of course chan1 gain must exactly match chan 2 gain (CMRR > 100 dB). I do this with an external scope differential preamp. In almost all cases, this improves S/N ratio by a factor > 100x for sub-mV signals below ~ 1 MHz.
I didn't know about this technique. Thank you for sharing.
Stellar video! I'll try this immediately.
dont understand why you only have 5k views, should be at least 50k views
I love your videos! Got a suggestion for a future video. It would be great to learn how to measure the complex impedance of an HF common mode choke using both a VNA and, for those of us who don't have a VNA or just want to verify the VNA's calculations, using a modern oscilloscope with an accurate signal generator.
One of my top 3 youtube channels. I especially appreciate how you demonstrate clear modeling of physical setups using circuit theory. This is one of the trickiest skills that I feel is lacking in a lot of explanations and learning resources.
A big example is how you model common mode noise in your schematics/simulations. I have not found many quality breakdowns of common mode modeling.
Thank you and please continue making this electronics content.
Thank you for the kind words!
@@FesZElectronics It's so true. Until I saw the schematic I had no idea what was going on.
It's often caused by degradation of grounding wire of oscilloscope. I have old oscilloscopes with old probes on work. And it happens often. In the beginning sometimes this common noise appears but grounding wire is still connected ("ringing" when you test it with multimeter/tester), but some time pass and it even disconnects from DC tester 's point of view. So I have some spare probes for this case and replacement the probe is first thing I do when I see CM noise. Also in some cases using oscilloscope with isolated channels can help (for example Tektronix). And if it also does not help it is better to use differential probe.
Great solution. Thanks for sharing 👍
Dude! Awesome. Just awesome. I will remember this.
The fundamental difference between the operation of a nanocrystalline core and a ferrite core lies in the absence of a resonance characteristic of ferrites at frequencies of 0.5÷2 MHz, and work on the absorption of RF interference, in contrast to the reflection of interference characteristic of ferrites, which is associated with a change in the inductive nature of the ferrite choke to capacitive and can create problems in the work of the IIP.
Increase in impedance and attenuation by 2 - 6 times,
High resonant frequency and wide bandwidth (up to 5 times),
Absence of many resonances in the frequency response of the impedance,
Reducing the leakage inductance by 2.3 - 3 times,
High saturation induction (1.17 T),
Excellent temperature stability (in the range from -60°С to +155°С the change in permeability is less than 15%),
Low active losses in the wire,
Wide operating temperature range: from -60°С to +100°С,
Reduction of volume and weight by 2 - 3 times.
Very nice. I have been wrestling with just that kind of noise.
Also, I wonder if you might consider posting your LTSpice files alongside the video. I find your use of LTSpice one of the most helpful parts of your videos. I don't mind building the smaller circuits myself in LTSpice, but some, like the one in this video, are probably more than I can find the time to do. Thank you for considering this.
I second this request
You don't have 3 minutes of your time to assemble such an easy circuit?
I learned many years ago from a radar vet that everything returns to ground. All those paths need calculating. Star ground / Common ground is king. =Exercise >> put a bus bar on your desk and tack the rear IEC ground and the front probe grounds to the *same* bus bar and it should do the trick. You can work "lifted" and not use the ground of the building if you wish. If you do that be careful to not mix the wall ground and your new ground. Put a breaker or a fuse to mimic a standard home panel if u want to be nanny safe. Neutral bonding is a searchable term.
I'm not sure if I follow but, I did have both pieces of equipment connected to the same PE(IEC) ground all the time - I was not using different connections. The issue was the length of the connection - If I just rely on the cables that the equipment comes with, the wires have to much inductance and at high frequencies they no longer couple noise correctly. So the probe ground was connected to IEC ground by default, just the power supply GND was not (that's how the equipment came standard). Normally any proper supply cable that can be used to supply the 2 pieces of equipment from the mains will have more inductance than my short direct interconnection.
Anyway, for the moment this sort of short connection (I talked about in the video) is good enough for my needs but in the future I plan to upgrade the power supply. Its easier to just remove the noise source completely than just applying bandages to try and hide it.
I did do a previous video in which I hacked into the power supply and another way to reduce the noise was to make all signal return to ground as you correctly put it - I added a bigger capacitor between the SMPS output ground and the IEC - that way the noise source was locally short circuited rather than closing trough other pieces of equipment.
Excellent.
This kind of stuff makes me happy going to bed.
Great channel and video - I found it very helpful. Many thanks for putting this together. Quick question: what is the name of the small blue two terminal connector that you use to connect the scope probe to the circuit board. I am looking for something similar to reduce the ground loops that arise from the small grounding loop?
Its nothing special, its a piece of 2.54mm 2pin mother connector (receptacle?) with a couple twisted wires going off it; the blue thing is just heat shrink tubing. The only thing noteworthy about the connector is that the contact is built like a tube, its not the more basic contact on the sides only; not sure how best to describe this.
Both your videos and your English are excellent. I would add that the word oscilloscope has the accent on the second syllable: os-CIL-lo-scope. It sounds like you are accenting the first or the third syllable and it took me a while to understand the word you meant. Thank you again for these excellent videos. I have watched several of them more than once.
nice one
To be clear, will this just reduce the probe noise or the noise from the psu ?
I am assuming it is to see clearly the actual noise in your circuit rather then the added noise from the probes ? Regardless of the psu.
Hello,
Can you make a video in which you try to explain and SIMULATE in LTspice the behaviour of a transformer?(i am talking about how we could see in the simulation when a transformer gets saturated and other effects that which we CAN NOT INTUIT or SEE EASILY in mind)
Thank you.
Hello @Bejan Cornel ! I do plan on doing a video about transformers at some point, but until then, you might find my video on inductors usefull; in it I also look at how saturation can be simulated - czcams.com/video/65fNDRows90/video.html
How about connecting PSU output negative directly to PE at PSU output?
Wow!
Would the issue of CM noise not be of any concern if the Load was "Floating" (battery pack) ? If i am probing a battery pack , can i be sure that i wont have any CM nose to worry about ?
Depends on how much CM noise is an issue. Any device will have some form o capacitive coupling to ground (walls, the floor etc). For a battery pack, you will have some capacity from the battery to ground (even if there is no obvious connection) and the device you are powering will have some capacity to ground, so it is possible to get CM noise. In any case, to have CM noise, you need a noise source - in my case its the switching converter. The chemical battery does not generate noise, so there should be no worry.
Are you happy with Hantek 5072?
what do you call those little sockets you can solder into the circuit to put your probe into? been looking for one for forever but don't know howto search for it!
Hello! It's nothing special really - a 2.54mm female pin array that has metallic sides. I found these in 2 versions, one that has a shallow and thin hole, its commonly used as an IC socket and the other one has a wider longer hole - the second is way better.
Something like: DS1002-03-1*10131
what happens if you isolate your oscilloscope?
my signal from IR sensor looks terrible! It looks like a cheese with all those holes :/ it might be from the OR gate maybe I used to add the signal and carrier hmm
Нанокристаллический материл лучше по сравнению с ферритами в разы. Принципиальное отличие работы нанокристаллического сердечника от ферритового заключается в отсутствии характерного для ферритов резонанса на частотах 0.5÷2 МГц, и работе на поглощение ВЧ помехи в отличие от характерного для ферритов отражения помехи, которое связано со сменой индуктивного характера ферритового дросселя на емкостной и может создавать проблемы в работе ИИП.
Рост импеданса и затухания в 2 - 6 раз,
Высокая резонансная частота и широкая полоса частот (до 5 раз),
Отсутствие множества резонансов на частотной характеристике импеданса,
Снижение индуктивности рассеяния в 2.3 - 3 раза,
Высокая индукция насыщения (1.17 Тл),
Отличная температурная стабильность (в диапазоне от -60°С до +155°С изменение проницаемости менее 15%),
Низкие активные потери в проводе,
Широкий диапазон рабочих температур: от -60°С до +100°С,
Снижение объема и веса в 2 - 3 раза.
Although it is severely pooh-poohed due to the potential case shock hazard, for 40 years I ripped out the grounding terminal on the plug-strip I plug the scope(s) into. That results in the scope ground being relative only to the probe ground.
I didn't know about the capacitor trick, I'll try that next!
-Matt
Just use an AC line filter?
If you put a choke in the ground line, won't that pass 50/60Hz (ELF) but block RF?
My ancient Phillips scope has only a two-wire plug to the mains. The whole chassis floats.
power line noise also
You are tying the Earth Ground of the Oscilloscope to the Floating Ground of the PSU. Doesn't this present the opportunity of bad, and/or dangerous ground loop problems? It seems to me that it has to be defeating the whole point of having the PSU ground floating.
Well the GND of the oscilloscope (all oscilloscopes that I checked - even the expensive ones) have the GND tied to EG. So when you make any measurement using the standard probe , you end up connect the oscilloscope EG to the circuit GND.
There are certain pieces of equipment that have the GND-EG connection like oscillocope and PC. I guess as long as you dong interconnect 2 of these equipment to form a proper EG loop its ok.
On the other hand, why is this done? I guess for the measurement equipment to have a stable reference (not a floating ground)
You can't cheap out on copper and get rid of noise on ground and ground loops.
try this one bro
czcams.com/video/xaELqAo4kkQ/video.html
👍👍
Your explanations are very good, but your manner to switch camera every 5-10 seconds makes me dizzy. If you'll make continuous scenes for 40 seconds or longer, it will really help. Especially when you are showing schematics as it needs time to read and understand. Also, you can try picture-in-picture (camera view in the corner of schematics).
Uh-sill-oh-scope