Jump to content

Characterizing interference from USB chargers


EricJ

Recommended Posts

Recently while flying right seat with my buddy in his Cherokee 180 we were having a ton of trouble with his comm radios.   The noise levels were high and we were having trouble understanding anything.  It stopped when we unplugged my USB charger that was connected to one of my tablets, the same one I've been using in my airplane for quite a while to power the tabs and the Stratux.   We plugged one of his USB chargers in and did not have any further issues with the radios.

I do radio comm for a living, so, naturally I had to nerd out on this and did some characterization with a spectrum analyzer:

http://ericjacobsen.org/Files/USB_Power_Supply_RF_analysis_2.pdf

Bottom line:   the RF output level varies significantly between different examples of generic chargers, and may vary significantly with an individual charger depending on input voltage and current load.   The quietest charger I tested was a cheapie give-away that has Wells Fargo branding on it.   So go figure, but if you're having radio trouble try unplugging your chargers and if they're problematic just try a different one, even a cheap one.

You probably already knew that, but thought I'd throw this out there, anyway.

  • Like 3
  • Thanks 1
Link to comment
Share on other sites

57 minutes ago, teejayevans said:

You can buy a large lithium battery with USB ports, they should be very quiet.

Give us an idea of sizes, costs, how long they would power a couple of IPads, and how you recharge them.

Link to comment
Share on other sites

You can buy ferrite chokes to clamp on one end of each wire and it will solve your problem.  I did this to solve RF problems you described.  

https://www.amazon.com/dp/B015RAZTIA/ref=cm_sw_r_cp_api_gO5dAbA0HA2XH

 

also, the usb cig plug adapter has to be of quality make    I bought an ANKER brand adapter and it solved issues.  This is a USA company and you can call them for all specs..  not cheap Chinese crap that is so easily available.  This one is two port, but I got one of their higher watt 4 port units and works great..  no noise  

https://www.amazon.com/dp/B00VH84L5E/ref=cm_sw_r_cp_api_yR5dAbY7CYQXY

 

For cables, get anker as well.  They use better shielding and higher gauge wires so you don’t get as much voltage drop.  Buy short lengths if possible for minimal voltage loss   

 

Link to comment
Share on other sites

EricJ, I am very impressed with the quality if your analysis. I was by formation trained with RF stuff and worked in that field early in my career. So I can still catch most of that stuff...I switched to software to make more money more than a decade ago.(Aviation is expensive!)

Do you believe that the interference travels electrically (via the power lines) or electromagnetically (airwaves)? Each way would require different approaches to diminish the interferences to appropriate levels. Adding filters to the power line would only fix one of them.

We could assume the noisy ones use the switching power supply regulation type. They probably did not care or have any EMC rules/limits to comply with. There is always the potential of major issue with the comms leading to an accident. This could lead to NTSB making recommendations to force having certified devices in the cockpit preventing shutting off comm reception.

Yves

Link to comment
Share on other sites

1 hour ago, Browncbr1 said:

also, the usb cig plug adapter has to be of quality make

I'd have thought that, but the quietest of the units I tested was the cheapest one.

 

28 minutes ago, yvesg said:

EricJ, I am very impressed with the quality if your analysis. I was by formation trained with RF stuff and worked in that field early in my career. So I can still catch most of that stuff...I switched to software to make more money more than a decade ago.(Aviation is expensive!)

Do you believe that the interference travels electrically (via the power lines) or electromagnetically (airwaves)? Each way would require different approaches to diminish the interferences to appropriate levels. Adding filters to the power line would only fix one of them.

We could assume the noisy ones use the switching power supply regulation type. They probably did not care or have any EMC rules/limits to comply with. There is always the potential of major issue with the comms leading to an accident. This could lead to NTSB making recommendations to force having certified devices in the cockpit preventing shutting off comm reception.

Thanks for the kind words.   You're right that it take a lot deeper analysis to sort out the path(s) through which an interferer actually causes interference to a particular system, and it could be different in each airplane since the avionics are generally of different makes and different installations and different conditions and the antennas could be placed differently and the cigar lighter placed differently as well.   Since my test fixture used an antenna for the sensor, I was really only measuring the radiated energy, so anything going back through the input power wouldn't have been accounted for except what was radiated through the wire.   If I really wanted to get crazy I could use one of my sniffer loops instead of the antenna and try to find the hot spots, but that was a bit beyond the time I was willing to put in on this.  ;) 

I think pretty much all of these types of devices these days use switching power supplies and regulators in the DC-DC converter, as linear regulators are way too inefficient.   So I think it comes down to how the switchers are made, how well they're shielded (if at all), and what frequencies they switch at.   It seems like the smaller/faster/cheaper/better they make them, the higher the switching frequency, which can move the spurious harmonics and radiation further up as well.   That's pretty much where the RF interference energy is likely to come from, and clearly some are just made differently than others.   The worst ones that I tested also have a numeric display and probably a little embedded processor to run it, so the clocks for those may be getting out as well.  They all have to pass FCC part 15 for unintentional radiation, which puts some basic limits on them but doesn't prevent a nearby radio or antenna from being susceptible, especially if it, or its antenna or cable, is nearby the interfering device.

  • Like 1
Link to comment
Share on other sites

11 hours ago, DonMuncy said:

You guys are amazing. I'm only smart enough to understand "buy this brand" or don't use that one.

I managed to find Eric’s web page (just on top of the report he published) and yes I can confirm he is a smart guy. BTW Eric, about your web page: I attempted to get to your linkedIn profile but the link on your page seems to be stale. I searched for your name and it seems there is a ton of guys with the same name.... no luck.

Yves

Link to comment
Share on other sites

Give us an idea of sizes, costs, how long they would power a couple of IPads, and how you recharge them.

Something like this:
6d929852df8a9d661c9bc1b2e506c6c3.jpg
A 10000mah should last about 4 hours under maximum load for 1 tablet (tablet uses as much as 2400mah. The above example has a 2.4 and 1.0 mah port, the 1.0mah should be ok for most phones. If you need 2 high capacity usb ports, you want a 15000mah or bigger. These are great if you have passengers and your charger is being used by the pilot.
Link to comment
Share on other sites

I too had trouble with noise from USB chargers in the cigar outlet.  The gadget also tended to back out of the hole and disconnect itself.  Annoying cheap stuff.  

So I created an “owner produced part.”  It contains a switch mode regulator operating near 1 MHz that drops the 28 volt power to 7.5 volts which then goes to a linear mode LDO regulator to create the 5 volt 2 amp source for the USB cord. The whole thing is in an aluminum die cast Bud box for shielding.  I put excessive ferrite and capacitive low pass filtering on the wires in and out.  

A classic of over-engineering.  But noise free.  

The battery pack approach is easier.  

  • Like 1
Link to comment
Share on other sites

29 minutes ago, Jerry 5TJ said:

I too had trouble with noise from USB chargers in the cigar outlet.  The gadget also tended to back out of the hole and disconnect itself.  Annoying cheap stuff.  

So I created an “owner produced part.”  It contains a switch mode regulator operating near 1 MHz that drops the 28 volt power to 7.5 volts which then goes to a linear mode LDO regulator to create the 5 volt 2 amp source for the USB cord. The whole thing is in an aluminum die cast Bud box for shielding.  I put excessive ferrite and capacitive low pass filtering on the wires in and out.  

A classic of over-engineering.  But noise free.  

The battery pack approach is easier.  

So where is the electrical diagram and photo of the finished product.

Link to comment
Share on other sites

On 11/18/2017 at 9:24 AM, yvesg said:

I managed to find Eric’s web page (just on top of the report he published) and yes I can confirm he is a smart guy. BTW Eric, about your web page: I attempted to get to your linkedIn profile but the link on your page seems to be stale. I searched for your name and it seems there is a ton of guys with the same name.... no luck.

Yves

There are!  I'm friends with a number of them.  It makes my FB page rather difficult to decipher sometimes.
http://ericjacobsen.org/whichej.htm

Warning:  there are a ton of stale links there, too, but you get the idea.

A not-infrequent happening on my FB page:

Eric_on_FB.jpg

And the correct LinkedIn page if it matters since I may not get around to fixing the link rot on my site for a long time:
https://www.linkedin.com/in/eric-jacobsen-b701503/

On 11/18/2017 at 10:12 AM, teejayevans said:


Something like this:
6d929852df8a9d661c9bc1b2e506c6c3.jpg
A 10000mah should last about 4 hours under maximum load for 1 tablet (tablet uses as much as 2400mah. The above example has a 2.4 and 1.0 mah port, the 1.0mah should be ok for most phones. If you need 2 high capacity usb ports, you want a 15000mah or bigger. These are great if you have passengers and your charger is being used by the pilot.

That looks like a nice one!   I've been using these for a while:

https://www.amazon.com/gp/product/B00ZWUZG70/ref=oh_aui_detailpage_o03_s01?ie=UTF8&psc=1

My cross countries tend to be long-ish, e.g., flew back from KGGG in a day, 7.5 hours of flying.   Trips to SoDak and back are typically 6-6.5 hours each way.    Going to see the eclipse was 7+ hours each way.   I don't want to be caught with an electrical failure with a low battery in either tablet, and don't want to lose the Stratux in a failure, either, if I don't have to.    The Stratux is always plugged into the USB battery I linked and ship's power keeps that battery charged.   Tablets get plugged into the charger as needed to keep them above 50% or so.   There's usually at least one tablet plugged into the charger.    So if/when I lose power to the charger, hopefully one less problem I'll have is losing either tablet or the Stratux before I can get somewhere suitable.   I carry a hand-held comm radio for the same reason.   

Keeping the Stratux powered by the USB battery also solved a problem I was having of the flash card getting corrupted during power up/down.   I think the power relays on my airplane are bouncy, i.e., they chatter when turned on/off.

 

 

Edited by EricJ
  • Thanks 1
Link to comment
Share on other sites

Thanks Eric for the links. I will connect through LinkedIn when I have a chance. I am also a consultant but as stated changed for software along the path. I did teach at college level telecom stuff before I made the switch. All this because I love flying. It went pretty close for me to become involved with Software Defined radio stuff. I got an offer to work for Zeligsoft (that I rejected) when I was with IBM. Several guys there (when they were triving) came from Rational / IBM like I did. I am causing a thread shift now so let’s talk more about airplane stuff:-)

Jerry’s device is the proof that when you put your mind into it that you can devise some device that will be EMC. It will cost more but seems to be much more reliable. Jerry, did you check your thing with a spectrum analyzer? (or check all receivable channels) It is always possible that your invention does generate stuff but you did not run into it yet.

Yves

  • Like 1
Link to comment
Share on other sites

All of the USB battery packs that I have saw have a switching boost regulator inside them so they are going to be producing some level of noise also. The actual battery voltage in them is usually 3.6 volts.

Attached is a picture of our anechoic chamber at work we use for all our radiated emission testing. The part under test sets on a table that rotates while the antenna is raised and lowered. We have close to $1M invested in it. It's not a fun place to spend the afternoon alone. The reference antenna on the end of the boom is over $10k. It got a small ding in it last year and had to be replaced. 

image.jpg

  • Like 3
Link to comment
Share on other sites

Ah....  those were the days.  HP 8510 C. 100 dB across the shield (30 MHz - 3 GHz).  Finger stock on the doors that needed replacing every four months.

FCC Part 15 and three letter agencies.  Followed by testing at the OAS.  

Long time ago.  

 

 

Link to comment
Share on other sites

On 11/19/2017 at 2:50 PM, Jerry 5TJ said:

So I created an “owner produced part.”  It contains a switch mode regulator operating near 1 MHz that drops the 28 volt power to 7.5 volts which then goes to a linear mode LDO regulator to create the 5 volt 2 amp source for the USB cord. The whole thing is in an aluminum die cast Bud box for shielding.  I put excessive ferrite and capacitive low pass filtering on the wires in and out.  

A classic of over-engineering.  But noise free.  

HUH?  :-) :-)    Way beyond my pay grade even when I was working :-)  I can grasp the battery source though. 

Link to comment
Share on other sites

On 11/20/2017 at 6:36 AM, yvesg said:

.....Jerry, did you check your thing with a spectrum analyzer? (or check all receivable channels) It is always possible that your invention does generate stuff but you did not run into it yet.

I hooked the USB output thru a dc block to the HP8595E in a sort-of CE test. Pretty quiet to 6 GHz. I can see the switcher noise but it is way down:  AM broadcast stations leaking into test set is louder. 

 

 

Link to comment
Share on other sites

  • 1 year later...

@EricJ  nice thread and I enjoyed your pdf article.  Do you think a SDR setup could be used to do the spectrum analysis as you did, or would it be too noisy?   It seems that if the antenna is as far way from the computer as possible and near the USB adapter it might work?  Thx.

Link to comment
Share on other sites

The USB adapters employ switching regulators to increase the dynamic range and reduce heat. You can reduce the noise they generate by putting them in the linear mode by adding a resistor in series at the cigarette lighter input. The resistor would be about 27 ohms/1 watt but it may vary depending on the load and whether your plane is 12v or 24v.

Interference is sometimes coupled through the power bus. For this case try a 1000mfd/50v capacitor connected in parallel to the cigarette lighter input. 

Edited by Gagarin
  • Confused 2
Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...

Important Information

We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue.