Вставка
- Опубліковано 5 січ 2018
- Are my USB cables any good? Or will they vaporize if used?
I will answer this questions for my cables and you will learn how to test yours (tutorial) to avoid surprises in the future.
We use USB cables for many of our projects and other devices like smartphones. Sometimes our projects do not work, and we are not sure if a crappy USB cable is a reason. When I recently showed a batch of USB cables in my mailbag video, a viewer wrote, that they are not of good quality. This is why I will have a closer look at my USB cables. The focus is on providing power to our Raspberries or other small boards, not on data transfer or Smartphone charging.
Links:
Good long cable: s.click.aliexpress.com/e/IqrNVrF
Bad colored cables: s.click.aliexpress.com/e/UJYZzvr
USB adapter set: s.click.aliexpress.com/e/vjaUvF2
USB Connector set: s.click.aliexpress.com/e/6iuZNbE
Simple electronic load: bit.ly/2F9DcYe
Good Electronic load: bit.ly/2gLz7Sj
Expensive Electronic Load: s.click.aliexpress.com/e/R3jMzj6
I do no more remember where I got the other cables from.
Supporting Material and Blog Page: www.sensorsiot.org
Github: www.github.com/sensorsiot
My Patreon Page: / andreasspiess
If you want to support the channel and buy from Banggood use this link to start your shopping: bit.ly/2jAQEf4 (no additional charges for you)
Official Wemos Store: s.click.aliexpress.com/e/jUzBiIq
profile.php?...
/ spiessa
www.instructables.com/member/...
Please do not try to Email me or invite me on LinkedIn. These communication channels are reserved for my main job - Наука та технологія
@10:06 I Learned to combine physics with engineering.
Now nothing works, and I don't know why.
:-))
Peter Lekkerkerker xD
😂 I can feel it. I can feel it.. I can feel it... 🤖 🤣🤣🤣
you made my day hahaha xD
😂😂😆😂😂😆😆😆
Andreas, Thanks for doing this. Going back to fundamentals keeps us honest. We all skip the basics sometimes especially when working at < 5 V and it is good that you have taken the time to analyse what we all know, but sometimes don't put our minds to, e.g. the higher intrinsic resistance of micro USB which can only increase with mechanical wear.
Please continue to do more of these 'Fundamentals' videos
I will do one from time to time
I learned this the hard way on a project were I "got smart" and used something similar to PoE to also power a sensor. Of course the power drop over the longer Ethernet wire out the sensor put it in _occasional_ brownout condition. I highlighted occasional as it is the worst kind of failure for engineers - intermittent failure.
I would suggest that makers construct their own USB power only cables using easily available connectors. You get better conductors and newer makers get in a little soldering practice in a reasonably forgiving space (connector headers). Just be sure to safely test each made wire to insure no polarity mistakes were made.
PoE usually uses higher voltages. Like that they can lose quite a lot of voltage without any problems.
Very nice video, I appreciate the methodical analysis technique. Regard the "mini" USB cables - while they would be superior in terms of the power handling capability, their big flaw (from my view) is that the the "springy" component of the mini USB connector is on the equipment-side, while the fixed contacts are in the cable. This is the reverse of the micro USB connectors, where as the connection arrangement ages and wears, it's the cheap cable that wears out, which is easy to replace, rather the connector on the device. The reverse is the case for the mini-USB connector arrangement, where the device's pins will wear and now you have a more serious problem if there are too many connector mating cycles.
I really enjoy your videos, thanks for sharing with us.
Thank you for your information. I did not know that. But it is for sure better to have the wear and tear on the cables instead of the expensive devices. I wonder how long my iPhone connectors will survive...
This is a great video about a pertinent topic related to a project I'm working on. Thank you for sharing!
You are welcome!
It was good to see the comparison among a large sample of different cables. Great job on a good topic, thanks for the insight.
You are welcome!
Very good as usual. Thanks for your work Andreas. Just a little advise if I may, please add the units where ever it is. I.e in your table, I would expect "Voltage drop @ 2Adc [Vdc]" or Pj [W]
I was not aware I forgot them. Thanks.
When you have a good video it's often a great one. This was great :) I don't use the micros you do so quite a few of your videos aren't to my taste, but the more general ones are very handy. Clear, concise and with lots of little things you do that help - like reminding me to use four wire test setups.
Thank you for your nice words. I try to cover different things, but I have to admit, my heart is close to micros, wireless communication, and sensors. But still, you find sometimes other things on this channel.
Thank you for a sorely needed tutorial.
You are welcome!
I've been doing cabletest for some time now.
Nice to see someone who really knows what he's talking about explain it :)
Thank you for your nice words!
I looked at your results. You did a great job! And the way you do it is maybe a little less precise than mine, but as a way to compare cables it is precise and your results for sure are valid because you always use the same setup. Like that, you have always the same errors, if any.
I ordered also some cables recommended by my viewers and will test them. And now I can also compare it to your table if we have the same.
Thanks for this video. I didn't realize the cable quality is that important!
You are welcome!
That was super informative! Thanks! Happy New Year from British Columbia, Canada, Andreas.
Thank you and also Happy New Year!
Wonderful story. Worse than an engineer, I'm an apathetic engineer: I don't even care why it works, but how it works is dearly important. I love the useful information. Thank you once again, cool and hot!.
You are welcome, apathetic engineer ;-)
Thanks for another great video. My practical problems mostly arise from the mechanics of the micro USB. There's a distinct difference in the way one or the other 'locks' in place when inserted and it seems this also influences the quality of the connection in terms of resistance. Also, my kids tend to 'abuse' the cables so mechanical strength is also a factor to consider for the long run. On the other hand, the cable should always be weaker than the socket in your devices to prevent that getting broken. Anyway, for most people a cable tester would be the best way to distinguish I guess.
As I mentioned at the beginning of the video I did not focus on charging Smartphones. But for sure, mechanics of the micro USB connectors is a big issue, especially if they are not protected like on our boards. So most of us know how to deal with it (because probably most of us broke one or two off the PCB).
For that reason, type B connectors have been much more robust, but for many applications too big (and maybe also too expensive)
Sehr gutes Video Andreas, und frohes neues Jahr.
Ich wünsche dir ebenfalls ein glückliches neues Jahr!
What about when one does not know why it works and cannot do it either? That is the definition of a manager.
:-))
Haha :)
A very informative video, the best on the subject.
Thanks!
Again & again thank you....🙏 for this another nice video...
I understand and learn something in this video ... That are confusing me lot before...
Difference between physicists and engineers .... Lines are super...
You are welcome!
Great video as always.
:-)
It's a side note to the subject of your video, but I hate! Hate! HATE! USB cables that have only power and no data lines. Why do I hate them so much? Because you can't tell by looking at them which kind they are. If they had a different symbol or something, or maybe an attached tag that said "I'm sorry", it would be better. I know it saves the manufacturer some money (multiplied by a million cables), but can't they take my feelings into account? Every time I discover one in my big-box-of-USB-cables, I tie a simple knot in it so I will recognize it the next time I think of using it.
You know what: During my video making I though all of these cables away. Because I think like you!
I used to throw them away, too, before I thought of tying a knot in them. Sometimes I do need just a power cable. Now, I will throw them away if they have crappy electrical characteristics. :-)
Nice work Andreas, would love you electronic load. Just finished building the Gearbest Z5. Demo to come. Where did you get the infrared camera mate. Still hot downunder 42 degrees yesterday.
The camera is from an online store here in Switzerland. Not cheap :-( Here it is also quite warm for the season: 4C
So finally your cables turned out as bad as mine. It is also possible to use so-called USB Testers to check current and voltage. One of them (white one with short cable) is also perfect for checking quality of the cable because has extra microUSB connector.
But still, most of them are ok for ESP projects.
I ordered a few new ones from the supplier of the good long one to compare.
very nice video. a big thank you to you sir.
You are welcome!
This is Good knowledge!
Thank you!
You are welcome!
Could you test Supra USB cable with jitterbug attached? I have that kind of setup to connect my Scarlett and im curious if that reduces the current or voltage.
I looked at the homepage of jitterbug ( www.audioquest.com/jitterbug/jitterbug ). They claim to improve "digital noise on the data lines" which is not the topic of this video. I am also not in high-end audio. So I will not spend the 50$ for a jitterbug.
The same applies to the Supra USB cable. It is focussed on high-end audio. Looking at the picture about the wires it seems that they are thick enough. But the price is even "thicker". I will not spend 3 times the money for a USB cable than for my microprocessor. This might be different if your HiFi rig costs several 1000 $.
Some of us are both engineers and physicists, and yes, most of the jokes below are good ;~}. FWIW, copper has a tempco, so as your cheap cable heats up say 10C, the resistance also goes up - so your statement about it being the same at all currents isn't quite true (but almost). Copper resistance will be 1.0399 times as high when it's 10C hotter. Maybe not important to this, but good to know if using copper for say, a current measuring shunt, or on the edge of burning up.
Another viewer also pointed me to this fact. Of course, I knew the effect, but thought, it is neglectable. So I learned something...
Great job Andreas. Clearly there is an opportunity for some manufacturer to produce a line of larger gauge cables for better power handling.
I will order a few other cables suggested by viewers. Maybe they already exist, but I do not have them. Maybe stuff for another video.
Great to see. Interesting to hear about making your own cables or buying _really_ good Micro USB cables :-)
I will order cables from tips given by my viewers in this video and test them in an upcoming video
Happy NY Andreas.
When I started, I also had issues with a cable. I assumed all cables had power and data lines.
I quickly learned this was a big mistake, all cables are definitely not equal.
I enjoyed the way an engineer broke down the problem.
I will be testing my own cables today. Just too be sure.
I did laugh a lot at Mrs Spiess’S input. Colour and length 😂😮😂
Thank you, I laughed a lot.
Happy New Year to you, too. And happy testing!
As always, usefull and interesting ;-)
Thank you!
Excellent!
Thanks!
Some USB cables sold (on eBay) have their wire thicknesses listed for the power wires. 20AWG are pretty good for 2A and fast charging loads. I would be interested in seeing those compared to the ones you tested here.
Maybe you send me a link?
They use AWG (American Wire Gauge) in their titles. I searched for "usb cable 18awg" and other gauges to find them. They start at $2.50.
So like eBay items ( I don't know if UA-cam will block a post if I use links, so I will use the item numbers instead):
171767082377 for 18AWG,
292360980058 for 19AWG and
173040671166 for 20AWG.
After getting a batch of USB cables that wouldn't run my SBCs, I specifically ordered the ones that listed their gauge and haven't had a problem since. I also use USB supplies that run at 5.2V to reliably power things like RPI3 boards.
Thanks for your info. I ordered now a cable with a 18AWG in the title and we will see...
WOW! Even more to consider.
Not a lot. Only a one-time effort ;-)
Andreas, have you heard of a CT2105 a 8 pin voltage regulator type for Li-on? I had one short 5v in and cannot find on Google. Also, are there fuses inline in some micro USB cables? After my short, cable was dead.
I do not know the CT2105. And crappy USB cables can work as fuses because the weakest part breaks (either cable or solder point)
ok thx
Good advice as usual. Thank you.
P.S. Where did you get those usb A to micro adapter from?
You are welcome! I now added the links to the things I used in my test to the description.
Thanks for this episode. Andreas, being in cable and power issues, it might me interesting to comparize mains (or 12v car) to USB chargers. The issue is that mobile phones use data D+ and D- lines to indicate how much current the charging circuit can draw from charger. I was caught on this issue several times when my phone does not charge when I borrowed it from others.
You are right, but I am not a specialist in Smartphones and cannot add a lot there. Therefore I did not mention all the issues around this signaling and QC.
Happy New Year Andreas!
Thanks for another very informative video.
I tested many USB cables from different sources and found Blitzwolf cables to be very good.
I bought my cables from Banggood, but this brand is also available from other vendors.
These cables are not very cheap, but you surely won’t regret the extra money you pay. :-)
Happy New Year to you, too and thank you for your tip!
In my own tests i used a battery with usb a to microusb. I used a meter between the battery and the cable that conntects to my raspberry pi 3. What i found is that the mechanical connection to the meter had an impact on voltage drop. If I physically moved the cable the voltage would stabilize and maintain the required 5 volts at the expected amperage, less ideal movements of the cable resulted in the aforementioned drops. This was tested with the cables that have an inline switch, so they were already asking for trouble to begin with. There was more resistance at the solder joints on the switch, and from the switch itself. Albeit negligible, these impacts lead me to want to solder directly from the cable to the battery itself and do away woth the usb a connectors altogether... This will eliminate the mechanical problems, and since this projects battery is dedicated to this project there are not really any downsides by removing that feature. Probably adding risk but i beleive this is acceptable. Anyway, thank you for sharing your findings. I think this is doing everyone a huge favor in their selection for a reliable power source in their projects.
Thank you for sharing your experience!
Love your explain at 10:14 "Physicist know why but not doit, Engineer doit but do not know why !! " just in nail !!
:-)
Very nice and informative video. Thank you.
btw the links to the simple and good electronic load are the same.
Thank you! I hope the link works now.
Thanks for sharing :-) Inetesting test, Thanks :-)
You are welcome. And for your loyalty to the channel!
For RaspberryPi 3, I had very good results with '15cm 0.5FT Micro B USB Fast Quick Charge Cable 2.1A Android SmartPhone 18AWG' from eBay (no data leads, from South Korea). They are not cheap, but really good. When 15 cm is too short, there's also a 30 cm version. With that cable, I can even power a RaspberryPi 3 with a small 1 TByte USB harddisk, which requires 500 mA on its own. No yellow lighting bolts in the top right of the screen. Yes, it is possible to use a Y cable with that disk, but that causes unstable power caused by back-feed, and you can't afford the voltage drop in diodes that would prevent the back-feed. In my opinion the capacitors in the Pi's power lines should have a higher value than they currently have.
Thanks for your tips. If you cables really have 18 AWG wires, then they have to be very good. These are already quite thick wires.
Thank You! Could not imagine such different behaviour in USB clables
I also did not know. So I measured...
Good video. You don't need to test each cable, just look at the writing on side of the cable for the size of the power conductors. Often on cheap cables both the data and power wires are 28AWG (or smaller) on better cables you will see the power wires are 24AWG or so in size. When buying new cables pay attention to this.
The moral of the story is that micro-USB connector are a terrible choice for 2+ amps of current. Not only are the connectors then selves not suitable but the USB cables themselves are questionable. RPI and others should move to a barrel type connector for these higher current requirements.
I agree with most of what you write. Still, I think, test them is not a big effort.
I think I have seen all of your Videos then I see this one that has been around for two years. Great video but now I have to take the time to evaluate all my cables. I probably have over 20 in my installation. I have one Rpi that dies every month or so. Interrupt the power and it runs fine for a a month or more. Tried everything (almost). Checking it first. Thanks for all you do.
Maybe power is the issue. RPis are quite picky...
The extended cable at 15:35 will be even better if built using a microUSB connector with the cable cut close to it and soldered to the thick cable. This will avoid more than half of the 0.37V drop through the adapter.
Excellent choice of theme for this video! I guess most viewers would not consider the losses using USB for powering devices. I'm pretty sure many will use much shorter cables to charge their devices too after seeing this. The power loss will make the charging time longer. Why? Well the charger is a constant voltage source not a constant current source.
Thanks for your tip. Also a very good way to solve the issue.
Thanks for this. I have an esp8266 project that isn’t working and I can’t figure out why. I think I’ll go check the cables.
Good luck!
I love your videos! You mentioned the Type A, micro, mini, and Type C connectors, but you left out type B! It's typically used on USB printers. My Arduino Mega 2560 has a USB type B connector on it.
You are right. The type B is still used in some setups. And I like it if there is enough space available.
I guess it would take an engineer to prove that bigger is NOT always better. Anyway, excellent explanation and demonstration. Thanks.
:-))
Very educative video! Thanks! I brought the same crappy usb cable You did (the black one). It really is crappy, my phone charges about 5 times longer with it. Actually I had this suspicion the first time I opened the package it was too thin and flimsy.
Thank you for your feedback!
Thankyou!
You are welcome!
Thanks for this video.
At the moment I have problems uploading my code to the ESP32 via a 2m cable (PC and ESP are on opposite sides of my room....). I have a 3/4 rate that there is a checksum error.... Is there a way to test cables for data loss? Or is there an indicator of data stability?
I never did that. Did yo test with a short cable?
Hi Andreas, thanks for the quick reply and sorry for my late response.
In a new setup I had to bridge about 4m from the computer to my video projector where another ESP32 lies, so I used the same cable with another USB cable that came with my wireless keyboard (the one where you're supposed to put the WiFi dongle in).
This time there are no problems uploading, so I guess it is not the cable. Maybe that other ESP32 or one of my USB ports is damaged or there are interferences with other hardware lying around (LED transformers) or even with the connected 433MHz receiver?
I was asking because there are other cables with visible shielding (that at least look more trustworthy). I'm not sure how good these longer cables actually are for transferring data, as these are sold as "charging cables". But they seem to work fine....
Thank you for the really interesting video.
I did the test with 2 amp / 5v (entry).
I was surprised to see a drop of 1.7v in about 50% of my cables (even the one that look good and expensive from Digitec...).
My best calbles are USB3 cables with almost no drop of voltage.
I had really good experience with some cables from Voxlink (2-3$/piece)
Thank you for your feedback!
Great Video thank you
:-)
It should be "Expensive Load" in the description! Good video.
Thanks. Corrected.
I remember the first bag of cheap esp8266s I ordered from china - for hours I thought "Can really all 10 of them be broken, !@#$%"? Then I switched out the USB cable and e vóila, everything worked fine. It's always the soft, squishy ones that try to look as "Designed in Cupertino", which fail fast, not the sturdier, but maybe not so cool looking ones, which have thicker copper wires in them
Thanks for sharing your experience!
tipp is a cable good then i can stick it in my kaybord and unstick (the usb on my keybord a havy)
So funny I'm encountering this problem with almost all types of USB cables with a Teensy 3.6 project. Just can't seem to catch a break, have used probably 50 different USB cables altogether though they all were just old ones I had around. Only a few were brand new. But I'm wondering where one would go for some quality cables that just plain work? Also, tests for power is fine but what about for data? Do the same principles apply or is data more advanced/involved? I imagine it would be...
On the low data speeds we use for our micros we should not get problems, at least not with relatively short cables. For higher speeds, data cables have to be built with knowledge. But resistance is not important for that reason
Nice work as usual Andreas, hank you for this video! I suspect that big voltage drops are very common and many people don’t understand why their Raspberry or ESP sometimes crash without obvious reason when the culprit is simply the bad USB cable. So I wonder if it is advisable to power the ESP or the Raspberry through the GPIO pins. The main problem is that these pins are unprotected (no fuse) and maybe unregulated. That means one has to build an input circuit with fuse and regulator. But maybe that would be worth trying. What do you think?
I would keep the Raspi as it is. For the ESP I often use a barebone ESP-12 and a voltage regulator.
Some good usb cables actually tell you the thickness of the internal power wires. The one I use to charge my phone says 26AWG. sadly I don't have the tools to measure its resistance.
You were lucky to find this kind of information. Usually, it is not available.
So after soldering barrel jack onto RPi3 it will be far more better than those small USB connectors.
If you find a place. Or just solder two cables with a Type A connector for the charger as another viewer suggested.
I see that some cell manufacturers actually use the resistance of the cable to limit current. I found this out when I replaced an "approved" cable with a "better" lower resistance cable and soon after that while my device charged much faster, it also destroyed two chargers due to overheating.
That is well possible, but would not considered to be „good practice“ if you are allowed to change the cables. If connected to the charger I would accept it.
Ya .. bad design ... and a sad reality! (as i found out). I guess that the chargers do not include a current protection mechanism ... and interestingly the little manual that comes with the charger says to use only "approved" products. So much for standards.
i have been using AMS1117-3.3 voltage regulator but its always trouble but alternate i have is MIC5219-3.3BM SOT-23-5 package with 500mA peak output The MIC5219 is designed to provide 200mA of continuous current. if AMS1117-3 is bad choice then what other options we can have ?
I often used the HT7333
@@AndreasSpiess HT7333 is good choice but it has Maximum output current:250mA, is there any voltage regulator with 600ma to 1A
Maybe the HT7833?
I'm starting to prefer color coded cables so I don't have to sift through looking for micro/mini/C ends. Right now mini is blue, micro is any other color, C is black.
Good idea. Too late for me :-)
Good Information
Thanks!
I like my USB C cables to be cloth covered, and the micros to be plastic. This allows me to charge my phone (USB C and my headphones (micro) in the dark by touch when my wife is already sleeping. So it isnt all aestetics, there are practical reasons for covering choices but this video narrows down my options even more.
There is an Android application called Ampere which dynamically reports current during charging or use from the point of view of the battery which I used to sort out the worst cables. I dont know how accurate it is.
Good idea. I try to buy the C cables in red... Does not work in the dark...
Can you do same with
1 car charger
2 power bank also portable battery
3 usb meter also usb tester
I do not think so, because these topics are already covered on many other channels.
Andreas Spiess please tell me which channel?
Some time ago, I bought a collection of cables that came in different colors (yay), but they also had a braided covering on the cable (double yay). I wanted that for my family members who don't always take good care of things. (That's why I have big buckets of USB cables, USB flash drives, small power banks, and so on. Somehow, they go away and never come back.) Alas, I discovered in practice that their electrical quality is not very good ... I figured this out because they were very slow to charge devices.
Now, from your video, I have some idea of how to test the electrical quality of the cables. But I have a naive question (I'm a software guy, so the answer is probably obvious to EEs). What is the difference in the method you used with a current load compared to just measuring the no-load resistance of the two power leads? In other words, plug both ends of the cable into a breakout board and then measure the resistance.
No difference, because I actually did nothing else. But many multimeters are not very accurate in the low ohm range. And you have to pay attention that you do not add any additional resistance between your multimeter and the cable. This is the reason for the 4 wire method.
I used my method because it is easier understandable and less abstract for many people. But you can do it much simpler if you do not need it very accurate. I might do an update about that as a „quickie“.
I use an app called "AccuBattery" with my phone which shows the current throught the phone while charging. Most usb cables can't actually supply 2A even though the packaging claims so.
I do not know these apps. But phones have some signaling possibilities to reduce current if they do not need it. So it might be, that you did not measure the cable or the charger but the actual "command" of the phone.
Andreas Spiess
Thank you for the reply.
Yes that's true, I notice after about 85% capacity the current starts to decline. I turn off the wifi and lower brightness to lowest settings and make sure the phone is less than 70% charged to a make an "acceptable" control environment. I switch the cables one after the other for comparison and find about 0.7 amp difference between high-quality and ordinary usb cables. But it's still just a casual test to see if the cable is very bad or not, only comparison.
At time stamp 14:03 - There are the following equations:
R = 0.55v/.7A
R = 0.78 Ohms
R or 0.78 Ohms + 50% of 0.78 or 0.39 = 0.78 + 0.39 = 1.18 Ohms
I must have missed a step. How was a value of 0.52 Ohms reached?
It is the other way round. I assume the connectors „gain“ resistance with age. So the new ones should have a smaller resistance. But maybe I made a calculation error with the percentages.
Just wondering have you tried UM24/UM25 USB voltage testers? I know, that with your equipment its just an expensive toy, but for hobbyist like me it is great thing how to measure cable resistance with almost no hassle.
Of course the problem is it has only type A and MicroUSB connectors...
I have one of these and it is great. I never used it to test cables as a wanted to go to the basics. I am sure you can get reasonable results if you know what you do.
I am trying to know what I am doing :-) After last 2-3 years I got to know things about electricity a lot, but as much I keep getting to know, the less I know :-(
I am grateful for your videos, back then I bought ESP8266 chips I know they are pretty good ones, but had not too much time to test and play with them. Now, when I go through your videos it seems it is all too much simple, but without proper knowledge it is hard. Thank you and keep going.
:-)
Für den Schweizer eine Anregung auf deutsch:
Ich weiß nicht, wie Sie Ihre Videos erstellen, aber falls Sie die Tabellen irgendwie mit Excel oder so generieren, dann fände ich bei den Zahlenreihen eine "Bedingte Formatierung" mit Datenbalken, Farbskalen oder Symbolen hilfreich. Das würde die Sache viel übersichtlicher machen (für visuelle Leute wie mich).
An dieser Stelle aber mal ein DICKES DANKE für die ganzen Videos, da habe ich (als Physiker ;-) ) schon eine Menge gelernt.
Danke für die Anregung. Ich werde sdehen, was sich machen lässt. Ich bin allerdings nicht ein Fan von farbenfrohen Tabellen. Wenn ich Farben verwende, dann zum Hervorheben von z.B. Ausreissern. Mein Weg ist eher die Animation um die Aufmerksamkeit des Zuschauers zu lenken. Die Geschmäcker sind eben verschieden...
It's a pity that the USB standard didn't include a 5th wire to allow for the power supply to measure and adjust the destination voltage.
The equipment needed on both sides was probably too expensive in these days
Hello, I know I'm very late to comment this but was just rewatching and noticed (what appears to me anyways, please someone correct me if I'm actually wrong here) a mistake. @12:50 you state that any cable with less then 1.5Ω would be acceptable then when you went to the table you highlighted the acceptable and unacceptable wires but based it off of the voltage drop instead of the resistance value (the last acceptable wire had a voltage drop of 1.34V with 0.67Ω but based on your measurement all wires were technically acceptable, with the last one being 1.21Ω) I know this won't matter much since the use case described is very limited and also the cable obviously being quite crappy quality doesn't hurt discarding it. I'm not sure if this was already pointed out by anyone, I couldn't be bothered to read through all the comments😅, but thought I'd point it out since I noticed the small mistake. Love your videos sir thank you for all you do for the community!
You are right, I was not very precise.
When I plug in my USB cables, I fail constantly at the first attempt. Why? Because of the ports resistance is too high. ;-)
Nowadays many USB 3 cables come with external hard disks, card readers etc. I would be interested if these cables are better than the old USB 2 ones?
For the tests I did only the power wires are important and to my knowledge they are not related to the data transmission standards. But I did not test this aspect.
Thanks for this contribution of this very important Topic.
You could do these tests ver easily and automated using the ZKE electronic load you introduced in Episode #148.
Because of this crappy USB cables I am used to solder the power supply directly to the board. Even on my Orange Pi Plus 2e which has a strange barrel connector. But also this barrel connector has a voltage drop of several hundred mV.
Normally I am using cheap but certified (VDE or UL listed) power supplies of a final sale. E.g. 5V 1.5A for 2.50 EUR or router supplies 15V 1.66A for 1.75 EUR.
The 5V can be connected directly and for the 15V (or the 24V from my battery backed home power line ) I am using the tiny and cheap buck converters with MP2315. They are cheap (0.50 EUR) very versatile, "mini" and claim to have an efficiency of 97.5%.
Concerning the ZKE load. You are right. I just did not know where exactly it measures voltage. With my setup I was able to control this part. Concerning soldering: This is of course the best solution as it removes most of the resistances. And the MP2315, especially with the USB connectors, are a good Idea. Best would be if somebody would produce one with a microUSB connector...
The MP2315 ones with USB connector have an additional 5.6V Zener at the output. This is an advantage. But also here I do not use the USB connector at the output but soldering. Soldering ist fast, cheap, reliable and low Ohm.
Mostly I use the "adjustable mini 3A 97.5" from ali. You can use the ugly pot or one of the preinstalled voltage dividers, which I prefer.
Concerning the ZKE load. I added some 2.5mm^2 15cm wires. Using Wago 221 (or 222 or soldering) I can bypass the USB input connector and measure what I want up to ca. 20V. Adding an micro-USB connector adaptor board you can easily and quickly measure the complete cable including the power supply.
Good idea. So far I did not do that because I have no need. I have to amortize my expensive IT8512A ;-). If you always use the same current you also can assume your voltage at the charger is constant. Or you can put a checking voltage meter there...
"I have to amortize my expensive IT8512A ;-)." Ok, I see: Noblesse oblige ;-)
The main problems I have with usb charging cables are not resistance but only connector stability - I have to throw them away so often, no idea why. (But I mostly speak about mobiles, tablets and powerbanks)
That is a main reason they changed hte concept for USB-C. They should be better...
Andreas Spiess don’t have usb c experience, but with TB it’s mostly not the contacts (unlike microusb) but the cable ends.
I was looking for a long decent cable. Now I know, I'll get that 1.5m blue cable, shouldn't be hard to find....
:-)
Im an engineer! :D I have wondered what a proper usb cable testing would look like. Darn, I dont have the proper equipment yet.
Maybe in the future !
ALi links are not working...
Well possible. After a while suppliers change their products and I cannot update links for 300 videos :-(
Why don't you just short the connectors on the micro USB end, and then directly measure the Ohms from the USB (charger) side? Is there a special reason why you recommend connecting it to a load and measuring the current and voltage instead?
Would give the same results, but I thought it is easier understandable for the general public if I show how much voltage is lost.
Very good analysis. Well presented.
I wonder if there is a market for a small tester (assuming is does not already exist)? Something that can just plug in the cables and present the results on a screen or on a phone via BT. Would really help the hobbyist especially during troubleshooting. Probably need someone like Dave to do that though and crowdfund.
Maybe I do a "Quickie" about that. I have an idea.
I googled that and found QualMeter. Seems to be what you are looking for, except it presents the results of the tests on LEDs, wich could be better for non tech people .
lunedemai2 I believe you found exactly what I was suggesting. I little better if it gave real readings and could do some transient drops (to pick up erratic behaviour) but it definitely coves the gist of the suggestion properly.
Undefined Lastname yes its "automating" the testing. If intermittent errors occurs on a circuit a piece of kit that can eliminate one possible cause it a good thing to have in the arsenal.
As far as I know an arduino is not capable of supplying the currents used in testing.
A static 2A test say is not really ideal. As the current flows it heats up the wire and changes the results. As noted in the video sometimes the peak current draw could change based on the usage. Supposed the peak draw is only for a fraction of a second. Would a cable pass or fail? Google the QualMeter lunedemai2 suggested. There is a video below in the campaign info that shows how expensive the cables can be. Correct selection can be worthwhile.
Undefined Lastname yes its doable. I understand your point but in the age of so many very very useful amateur level kits in the under USD$10 range it would have been nice to see something you could just pick up and run with. Something that's so cheap its not worth doing it yourself. Just yesterday I did an electric guitar amp (uploaded today if your interested) utilising a LM386 as the pre amp. I have all parts to make this myself but at US$1.15 delivered its just not worth it to do a layout and spend the time soldering.
DOOOOOOOOOOOOD! In US I just plug my USB devices directly into a 120V house current plug.
Do not forget to disconnect your Smartphone before ;-)
I'm not an engineer or a physicist, I'm a technician. I know how it works and how to do it but it doesn't work. It's all designed wrong. I'm so frustrated. Never-the-less, you'll probably find the cheap cables are using CCA instead of straight copper.
I grabbed the thickest usb cable i had the other day and stripped it to use it for something, it had the tiniest wires I have ever seen, so yeah you never know the actual conductor gauge.
This was also my impression. This was one of the reasons for a closer look...
Before that i thought it was a beefy cable, I was using it for 2A lol, oops
For years now I'm using a different and very quick test. On my Samsung phone I have an app called Galaxy Charge current. It shows max, ref en average current while charging.
play.google.com/store/apps/details?id=com.abmantis.galaxychargingcurrent.free&hl=nl
Knowing what my charge adapter can deliver it quickly shows the quality of the cable. Even longer cables can work fine, but are rare.
I can buy and test it outside the shop and go back in to buy more if they test for good current.
This is the way I can quickly sort my good, bad and ugly usb charge cables.
One thing you didn't mention in your test is the material the cables are made of. Even good cables, can become bad if the cables (very often) breaks at the bending points.
I mostly (for my phone) buy multiple cheap cables that can handle the current and toss them at the first sign of a disconnecting charge.
My raspberry and other SOC have fixed power cables I soldered myself or a good usb cable that I pin down.
Thanks for your tips. Using the Smartphone as an "electronic load" is an interesting concept. And cheap!
Concerning bending cables: I focussed on small devices and there, I think, the problem of bending cables is smaller than with Smartphones (and uneducated users)...
So, injecting a little physics into the results (I'm neither physicist, nor engineer), isn't it possible that the smaller connectors heat more quickly since they have less surface area to dissipate the heat? They may have similar resistance, but the larger connectors act as heat sinks to get rid of the heat.
At any rate, I'm off to test the resistance of all of my USB cables!
Your argument could come from an engineer! This is well possible. We also have to keep in mind that the cables are in plastic (which does not dissipate heat very well) and the connectors are of metal (good dissipator).
This is why we have to read the thermal pictures with caution and later test the hypothesis with other means (which I hopefully have done). But a general rule is: If no temperature increase there is not much heat produced and also not a lot of resistance. This way usually is helpful to avoid searching for causes with a low probability.
As an engineer... I was wondering if it would be possible to overcome the lame-cable loss with a 1-V step up booster-converter. The losses will be what they are as long as they don't melt the cable. Another idea: measure resistance directly by using a 1 ohm resistor in series and measure the voltage drop. Hmmm.... and that reminds me of the difference between an physicist and a failed politician.
There are USB micro A, and micro B, mini A and mini B, its good to let ppl know there such types exist
I saw this in some charts but was not sure if this is relevant because I never saw it in the shops
I used to think like that and blame cables. But USB spec says only 500mA should be supported with USB 2.0.
So it's not actually cables. It's stupid people like raspberry pi designers who use micro usb to power it up.
For raspberry pi, I got sick of it's connector and now powering it up using %V pin in interface.
I do not blame anybody. I just want that it works. But I am with you in not really loving these micro USB connectors.
cable plays a big role, fast charging for example
You are right.
Hello and 73 from Berlin, Germany
Good morning. I hope everything is ok on your side.
This why I build my own cabled, oversized wires, and nothing you can do about connectors. This way I can have 5m or longer usb cable (extender)
Thanks for your feedback!
I don't understand ? How can R+50% = 1.5Ohms ? R = 2.2 , .5 * 2.2 = 1.1, 2.2 + 1.1 = 3.3 ?
I do not exactly know what you refer to. My point was, that a new connector has to have 1.5 ohms. Because with age, it will increase its resistance to the maximum acceptable 2.2 ohms.
@@AndreasSpiess at 12:34 there is a formula that threw me. So in fact, it is 1.5 + .75 = 2.25 which is approximately 2.2.
For rpi3 I usually look at the red led, if it flashes it means that the voltage supplied wasn't enough.
Thanks for the tip.
Shadan Rikan but basically if you see the led solid red most of the time you are going to be fine, that was the idea.
Good vid Andreas. For some, here's an easy testing method. I had a problem with Samsung Galaxy Tablet not charging, and eventually found it was dependent on the cable. There's a program called "Galaxy Charging Current" (play.google.com/store/apps/details?id=com.abmantis.galaxychargingcurrent) that will show the charge current being delivered to the phone/tablet. Using this and the same Samsung 2.5A charger, I found that the charge current delivered to a tablet could vary from 0.28A to 1.4A simply by altering the cable, and for the 7" tablet, it simply didn't charge if the current was too low. As might be expected, the brand named cables were best, and it's well worth the time to test and label each cable using a black marker pen, or for black plugs, form a little 'flag' by folding a short piece of yellow or white insulation tape around the cable and write on that.
This is a good idea for people with Android phones. It does not work with my iPhone. And I am not sure if the "current signaling" used between the power supply and the phone influences your results.
No wonder that white cable was bad, it was full of pink goo! :)
:-)
Long Black < Tiny Black
(applies only for USB cables)
And what about the blondes?
👏👏👏👏👏
:-)
How are we to purchase good quality cable? The internal wire and connection resistance information isn't included in most cable information. I am sure some sellers have high quality cables. Maybe even the low price sellers have a higher quality option. Thank for the videos.
I am not sure if price and quality is in a relation. So far, only testing is possible. However I ordered a few short cables from the supplier of the good long one. I will test them and, if good, inform you.
Andreas Spiess Can you test quality cables such as Anker powerline?
+Silvio I plan to purchase cables suggested by my viewers and do another test. Your proposal is already ordered
Andreas Spiess thank you :) 180cm, 90cm or 30cm?
YMMV but not having to worry about dodgy usb cables is worth spending a bit more for me. No guarantee that more expensive cables are better, but "you get what you pay for" is usually a factor.
I use Anker Cables they seem to be the best both in Quality and Price....
Thank you for the tip!