Do ESC's produce RF noise? RCHacker #69

[@joe0932]

Great video. Can't wait to see more of your crazy 12 motor multirotor! Keep the good work up :)

[@nrdesign1991]

Did you take a look at the entire spectrum? Is it just harmonics of a square wave? Maybe the processor is configured to output its clock on a port pin which is a common setting for AVR controllers. It can be disabled in the fuse bytes.

[@BMSWEB]

Another fantastic test! Thanks for sharing

[@Rchacker]

Thanks!

[@JohnMiltonOlivaresWalteros]

THAT FRAME LOOKS VERY COOL

[@GrantleyDesignJaguar]

Thanks Mark. Twisted and shielded cables... does it reduce the cable's RF noise out as well as in? Would it be worth wrapping them in foil?

[@hiwayM9]

Maybe a manufacturer needs to begin offering shielded servo wire. I am still too new to the flight portion and LRS tech, but my years of working coaxial rf and performing CLI leakage calls for FAA/FCC compliance in the cable tv business taught me that hasty practices in cable making at the fittings was the most common issue. Some micro coaxial, if it could be made at the proper impedance would be a boon... unless it exists and I am unaware of it. I am a big fan of your vids and efforts- many thanks.

[@Rchacker]

Thanks, trouble is they make these as cheap as possible. Sometimes that involves taking a good design then removing as many components as possible before it stops working. Although these do have an external oscillator.

[@CrashPilot1000]

Those internal Audiowires in your PC/Computer coming from the CDrom are shielded and very thin. So maybe they would reduce the RF noise a great deal? Just a thought, I have no equipment to check that out.

[@hiwayM9]

CrashPilot1000 nice lateral thinking... I have a resource I can ask who will test it if he can. Thank you.

[@CrashPilot1000]

***** Hi Hiway! Don't want to bother you but did you had a chance to check that out?

[@hiwayM9]

CrashPilot1000 I completely forgot- I need to follow up- it is winter but the guy I lean on for counsel travels a lot and we speak infrequently. I am putting a giant sticky on my board here in the office as a reminder. My apologies!

[@GreyhatInfiltrator]

Would you recommend shielding the RX with an aluminium tape of something?

[@RCShufty]

That's a pretty interesting dodecacopter setup you have there :)

[@ME-zu5kg]

Mark many thanks again more information for our builds. Please excuse the ignorance in the following question.....If I tie my servo cables with a cable tie and shorten them by rolling them up round my fingers then squash the centre and cable tie to reduce there size what type of RF noise should I expect.

[@Rchacker]

This is what I mean by twisting: www.milehighrc.com/images/4seHD_Twist.jpg And without proper test equipment is your best bet.

[@JohnFHendry]

RCHacker Just shows setting up FPV equipment really requires the correct RF equipment but this is the best advice on the subject IMO, thanks. Really surprised the noise is so localized in the ESC even with a linear BEC. Nothing like adding batteries to cut down on the noise and eliminate "clocks" the system hear "tic". What I'm wondering (as often asked with many "answers") when using a KK2.1* FCB with 4 ESC's with switching BECs is it better to pull the black ground wire with the red on the 3 ESC that's BEC are not used to avoid a second ground reference or keep it twisted with the signal wire. Seems using a separate ground wire in reference with the ESC's main supply ground twisted with the signal lead would be the best noise solution.

[@petleh82]

Have you checked the 5v / 3.3v regulator. perhaps try to desolder it and power the esc from an external 5v / 3.3v source to see if that is the cause of the noise.. Nice videos !

[@Rchacker]

It is a pretty common linear regulator on there. I assume that they do not make that type of noise.

[@Rchacker]

RCHacker Unless they have the filter cap values wrong and it is ringing, worth a look.

[@4mulas]

it's RF harmonic and may come from the camera or florescent light or other device around

[@scottthompson2481]

I love your videos mate , and am quet new to the hobby, so I am trying to learn , but it seems like a mine field of information and very involved electronics . Your knowledge is vast and very helpful , how would you advise me to learn the basics of the electronics and computer usage in the hobby , many thanks Scott

[@Rchacker]

Thanks Scott. If you want to learn computers and electronics, Arduino is a great place to start. Those skills will help with the modern flight controllers, Multiwii for instance is Arduino based. If you want the deep end: programming, start with these books "A book on C' or "the C programming language". and read the ATmega328 datasheet. Basics of the hobby, Flitetest have some great beginner videos. Experienced people will pick up great tips from them too.

[@scottthompson2481]

RCHacker Thanks a lot Mark , really appreciate your help , I shall let you now how I get on many thanks Scott .

[@Jakeacotter]

Spin the leads up to help the frequency to collapse on itself

[@sirglider]

Thanks for the tests! As always interesting and educating :) My question about such noise: would for example an aluminum foil around the wires shield them and stop the RF noise to go further? Would be a cheap way to build shielding...

[@Rchacker]

Your welcome, It should make a difference. I'm starting to feel the need for a follow-up video here...

[@planemadman1]

is that an a canon eos M you have? what lens is it that you are using , vintage but what? is that what you shot your wildlife footage on your other channel with

[@Rchacker]

Yes, eos-m with a Minolta MD 50mm f3.5 Macro. I have quite a few old minolta lenses that I collected on ebay etc. The other camera was another eos-m with the 18-55mm kit lens on it.

[@planemadman1]

i was just wondering as i have an eos m and use some old olympus lenses and wondered what was working well for you, thanks.

[@Rchacker]

planemadman1 Tragic lantern or magic lantern, and the various focus aids is pretty essential with the legacy lenses. The adapters are cheap. EOS-M to Minolta MD adapter review and comparison. Marc Griffith #4 My favourite, the 200mm f2.8, 24mm 2.8, 50mmPG and the macro lenses. They all have similar colour and bokeh qualities which makes editing easy. Added bonus I can keep using them when I upgrade to full frame.

[@planemadman1]

yes I have a memory card with magic lantern on but i always forget to use it. I suppose its because i use them more on the spur of the moment than carefully planned. I normally go to photo mode and use the 10x crop for focus then switch to video and not change the focus, it works ok but I really should be more prepared! unfortunately on my budget I won't be able to upgrade to FF any time soon so at least its one thing less to worry about!

[@Rchacker]

planemadman1 I normally configure the half press for 10x zoom. Makes focusing really easy and manual focus on old lenses is a dream.

[@Rchacker]

+Spoif, I cannot reply to your thread. G+ or something? Yes you might be right this needs a closer look.

[@Spoif]

I'm sorry, I didn't realise there was a problem. I will look into why my profile doesn't allow reply. Thanks for letting me know. I would be very interested in your findings (as I'm sure others would) with regard to the ESC if you make any progress. Good luck. PS: It might be something related to the BEC switched mode regulator (or a harmonic thereof); if one is present on the ESC.

[@RobB_VK6ES]

I am still getting over the hair massacre Marc. :) Try a small cap from sig to gnd @ the ESC board pads. A SMD cap preferably. Possibly sig to +ve as well depending in which line has least impedance. At the RF frequencies the cap(s) will apear as a short but at the PWM servo frequency it will have no effect. Cheers Rob

[@Rchacker]

Thanks, I was thinking that, I wonder if there are any ESCs that have a small RC filter built in. Testing the various solutions might be worth another video.

[@johneh87]

RCHacker Could be that the signal wire is picking up RF noise from the UBEC. try desoldering the 5v wire instead.

[@TheIan350z]

johneh87 I would also like to see you try different configurations with the included toroid, like 5 turns with all 3 wires close to ESC... 5 turns with just signal wire, and maybe 5 turns with the 5v wire if that's also emitting noise. Wrapping all three conductors should be less effective than just the one which is emitting noise. And only one or two turns probably won't be enough to make much difference. An smd cap would be optimal from a weight perspective but using the toroid which is free and doesn't involve tiny soldering would appeal to a lot more people.

[@Spoif]

The ATMEGA8 ADC does not perform super-speed ADC as it uses the primary clock. Conversion occurs when ADSC is set and completed when the ADC interface sets ADIF. This requires multiple ADC clock cycles; and would be a magnitude of kHz, as opposed to 433MHz. I would be tempted to probe the board with a scope or spectrum analyser around the discrete ESC components.

[@Rchacker]

Funny that now I can reply, youtube is still a bit buggy sometimes since the G+ fiasco.

[@Spoif]

RCHacker Apologies, it was my fault. I seemed to have acquired some unusual G+ settings. Thanks for letting me know. In regard to your ESC issue, I'd be interested to hear if you make any progress. I wonder if it is related to BEC circuitry? I am not familiar with the BEC switching frequencies in an ESC, but in my experience buck DC-DC switch mode regulators, they generally operate the in the 100kHz < 1MHz range.

[@marclazarusgriffith]

***** I assumed this has a linear regulator on it, I will have a closer look and do a follow up video.

[@Spoif]

Marc Griffith If you were flying 4S and used a 7805 linear regulator for conversion to 5V at 1A to drive flight gear, you'd be looking at dissipating >9W in heat. That's a lot of wasted energy. ESCs usually have switch mode regulators in the BEC which are more efficient.

[@JohnFHendry]

Thanks for sharing. This subject dealing with electrons and Ohm's simple law seen within all major equations continues to make my head spin... But this much I know: We are dealing with clocks and time dilation separating them with {a} shown and described here increasing. To make further progress in electronics a true inertial frame of reference needs to be established so the graviton or "hole" as it called observed in semiconductors due to it's reverse oscillation phase arrow can have it's phase timing separated from the electron's forward arrow and both be set relative to the master clock within the weak force. Because we see it backwards (electrons @ - to +) logic is hidden but it's there in reverse. You may find this interesting: E=m{a}c2. (Shows value of E is relative to time) The speed of light is 186282 miles per second: IE 186282 miles = 1 second. This is the Master clock cycle with two arrows creating a pendulum that is slightly longer on the following oscillation phase always: e{a}/t=hv. (Shows fine structure constant maintaining E as a constant using time dilation, it's a worm hole where more time is measured going back in than came out because time did not stop during the oscillation cycle and time is what we use to measure/observe with. It shows gravity is cause by the subtrahend of the two arrows connecting gravity to the weak force as half of EM... or a tad more we notice as gravity. CERN's muon phase timed neutrinos were announced Worldwide in 2011 @ (v-c)/c=2.48e-5 sec in 453.6 miles. That is an exact match with SLAC's E158 asymmetry of the weak force data @ 2.48e-5 sec added to the speed of light traveling 453.6 miles creating an asymmetry in time of 20e-5 sec, what we call gravity. After the data was shown to match the E158 ratio SLAC went back over their BaBar data and confirmed the asymmetry in time exposed a year later giving it a sigma 14 level of certainty. There was a write up on it on Wikipedia but that's been pulled off now but was copied and noted in other areas of the Internet. Here is the simple math that proves it. If you open up a calculator and follow the flow it's easier to see doing it yourself. Weak Force Asymmetry {a} adds 1hour/3600 seconds every 1000 years: Re: SLAC E158 data "using clocks". Muon neutrinos do not exceed speed of light as CERN said "if the data is correct". They are created at the speed of light adding to the speed of light measurement just like the photon "tricks" Summerfeld detected in 1948. (Because their phase angle is @ theta 23 unlike electron neutrinos but you don't need to understand that to see the data match.) 3600sec / 1000 years = 3.6 seconds of weak force asymmetry time added in a year, the distance light travels in 3.6 seconds. 3.6 sec x 186282 (speed of light) = 670615.2 miles added to the speed of light in one year 670615.2 / 365.2425 year = 1836.082055072999 miles added to speed of light in one day 1836.082055072999 / 24 hours = 76.50341896137498 miles added to speed of light in one hour 76.50341896137498 / 60 minutes = 1.275056982689583 miles added to speed of light in one minute 1.275056982689583 / 60 seconds = 0.021250949711493 of a mile added to speed of light in one second Now that we know the know the {a} asy time/distance added in a second to the speed of light (0.021 mile) we need to know the fraction of a second speed of light distance it took for light to travel only 453.6 miles which is the distance CERN's neutrinos traveled to know how much {a} asy time to add to the speed of light traveling 453.6 miles and see if the {a} weak force asymmetry gain in time matches CERN's 2.48e-5 sec neutrino gain in time exactly as I said it would to create the force of gravity. So we simply divide 186282 miles (which equals one second) by 453.6 miles (the fraction of one second distance the neutrinos traveled) to acquire the weak force asymmetry value added to the speed of light. 186282/453.6 miles = 410.6746031746032 (or 453.6 x 410.6746031746032 =186282 SOL) Now we add the weak force asymmetry time gained to the fraction of one second value of 410.67 0.021250949711493 / 410.6746031746032 = 5.174644243208279 e-5 This shows the total forward/back total neutrino oscillation time gained in 453.6 miles caused by the asymmetry of the weak force. Now notice the total asymmetry time gained of 5.17 e-5 sec is almost exactly double CERN's 2.48 e-5 sec past the speed of light. The next two simple calculations change physics by adding a 4th oscillation phase to the atom (graviton) and a second reverse arrow to time needed to support the "missing" graviton misnamed the hole opposite the electron and use CERN's data to show it's correct length relative to the electron and prove time has asymmetry that supports the force of gravity exactly as I had been saying for years never expecting this to happen and be so easy to prove and show that E=T, and time like energy must be conserved. We add a second reverse arrow (required for time to have asymmetry) for time by simply dividing 5.174644243208279 e-5 in half. 5.174644243208279 e-5 / 2 = 2.58732212160414 e-5 sec So we have 2.58732212160414 e-5 sec weak force asymmetry time gained in the forward arrow of time, and 2.58732212160414 e-5 sec weak force asymmetry time gained in the reverse arrow of time. We now add gravity to the second reverse arrow simply by using CERN's data which provides the correct forward arrow measurement of weak force asymmetry added in 453.6 miles by subtracting the forward arrow gain of 2.48e-5 sec from half of the ratio gained in 453.6 miles of 2.58 e-5 sec. 2.58 e-5 - 2.48 e-5 = 0.10 e-5 We now add it to the second reverse arrow 2.58 e-5 + 0.10e-5 = 2.68 e-5 sec So using SLAC's exceptionally specific but proven E158 WF Asy ratio obtained from adding the distance light travels in one hour to the distance light travels in one thousand years applied to 453.6 miles we can see using basic math that SLAC's E158 data matches CERN's neutrino data announced worldwide @ (v-c)/c=2.48e-5 sec in 453.6 miles with both sets of data adding 2.48 e-5 sec in the forward arrow of time and SLAC's data adding 2.68 e-5 sec to the rearward neutrino oscillation reverse arrow. This creates an asymmetry in time of 0.20e-5 sec (2.68 e-5 - 2.48 e-5 = 0.20 e-5) needed for gravity exactly as predicted. CERN Neutrinos @ (v-c)/c = 2.48 e-5 create a lesser diesis (Einstein's Comma) = 0.20 e-5 (0.10e-5 + 0.10e-5) by matching SLAC's E158 asymmetry of the weak force data in the forward arrow of time @ 2.48 e-5 sec in 453.6 miles. This is a game changer from dice to two sided cards because the data match shows time has a reverse Mass oscillation arrow that gravity needs to exist exactly as shown by the Unified Field Theory equation that first showed me and a few others the weak force has an asymmetry that allows our clocks to move independent of each other. SLAC's E158 ratio that creates the asymmetry of the weak force ratio is an incredibly specific ratio and I had been quoting the E158 data for years after finding it's value showing why it had to match like this creating 3 specific ratios out of one to add the asymmetry to time calling it a lesser diesis comma. A coincidence of this magnitude is simply not possible. (This was done long before SLAC went back over their BaBar data and confirmed the asymmetry in time giving it a sigma 14 level of certainty "far more than needed to declare a discovery".) There is a lot of money involved with Science so don't think scientists don't play games with the truth. That's why was I was forced to sue my neighbor, Dr. Mark Skinner, Boeing's Senior Scientist on Maui which isn't helping the truth be told due to what followed affecting over 100 KAPWC HOA members that tried to vote the good doctor and his friends off our HOA Board which we did with a Court Reporter present only to get sued in return with our own HOA money as they refused to step down even taking out money to pay their criminal defense fees exposing a corrupt legal system most Americans refuse to accept exists at the extreme level it does. But you can't have positive energy without negative energy and the larger the object the more obvious that becomes. To challenge CERN's (v-c)/c=2.48e-5 sec claim neutrinos (CERN used muon phased timed neutrinos) were exceeding the speed of light which they weren't just as the photon is mistaken to do too, same mistake from misunderstanding the data like dark matter caused by an octave level structure that fixes and puts the matter "back", many physicists said the data had to be in error because neutrinos (electron phase timed neutrinos) were measured traveling at the exact speed of light from ten's of thousands of light years away, not exceeding it. The problem with that "correction" is the fact that we know neutrinos oscillate and have mass and the same laws of Relativity that state no particle can exceed the speed of light ALSO state no particle with mass can travel at the speed of light. So the experts used one "impossibility" to dismiss another "impossibility" failing to see the gross error in logic and some tried to reduce the neutrino mass to say there must have been some slight delay but from ten's of thousands of light years away even a "slight" mass amount would measurable, not to mention CERN stated and an exact SOL measurement after they fixed the "loose cable". There is a reason for observing a difference in electron vs muon phase timed neutrinos but it is beyond the scope of this post's layman presentation. What we see happening in physics where data is "doctored" to make theories fit is what happens when you use "dice" to measure time with rather that listen to the greatest physicist that ever walked the face of Earth when he tells you that "God dose not play dice with the Universe". E=m{a}c2 adds the cosmological constant using data not known until 2004 and shows how much energy you have is relative to the value of weak force asymmetry in effect keeping E a constant and that value tells exactly what time it is as it establishes an inertial frame of reference to connect to time and space. You won't hear much from people who see understanding gravity connected to money I'm afraid as that's just not how science works anymore where long complex nondisclosure agreements are becoming the norm to get a paycheck in the new fields opening up. When you do the math yourself and realize the scale of vast distances involved showing how specific a ratio the E158 data creates it will make your hair stand on end. The greatest discovery in the History of mankind is being hidden from the general public by a handful of scientists but they can't hide it from SLAC and the people that created the new data and have been aware of it for many years. Time tells all and the end result takes them and their deception out of the effect of the new picture created just as Nature intended to protect Life's freewill e{a}/t=hv as Pauli noted is connected to the fine structure constant showing Freewill is a measurable 5th force in Nature. No Observer = no Universe. So it only makes sense that the observer would play the role of the last feather to land on the wheel and determine what card to pick...

[@vazza69]

HK blue series ESC?

[@Rchacker]

Yep.

[@SynergyhubOrg-WA]

Great video... I keep harping at people that all of our RC electronics can create interference and it should be checked. Good abatement practices like twisting the wires, shielding, and power filtering are all critical.

[@Rchacker]

Thanks, I was really expecting to see some noise from the wires driving the motor. I was quite surprised to see the noise around 433 even when it was idle.

[@SynergyhubOrg-WA]

RCHacker Try looking at BEC's (switching types a.k.a. UBEC), they do it as well. I tend to like the ones by Dr. Mad Thrust since they have a shield on them. Anything with a clock running the circuit is prone to spitting out noise. We just hope it is not on our frequency :)

[@MrHolozip]

that was great, thank you!

[@solutionclinic]

So 15minutes without any result, left with more "maybes" than before... What was the purpose of this video again?

[@paulvs8525]

Hai If you have noise on 433 MHz you should not use ferrite material for LF you have to use ferrite material for VHF or UHF type I use FT50/43 30-600MHz and on top of it a T50/2 then you're ready to cancel once from 1 to 600MHz You can found the mini ring core calculator the for triodes on paul.van.strien.org/software/minirk12_install.exe on www.pi4raz.nl/index.php?option=com_content&task=view&id=1040&Itemid=32 I have more useful programs I hope this will help you to solve your noise problems

[@pampaflyer8762]

i bet lots of rf noise around 430Mhz is coming from your usb extension cable from your computer. Check there. Turn off you computer and check it.

[@Rchacker]

Yes the computer spews it, but I did initial test with it off also. Good point on the extension cable that was sloppy on my part.