High current USB cables from Daiso

Comments

• jmg Posts: 15,146

  2020-09-13 09:24 edited 2020-09-13 09:25

  Those seem quite high voltage drops ?
  Google finds this table, where 55~70mΩ is at the 'good' end, and 255mΩ is at the 'poor' end, tested at 2A currents.
  
  http://budgetlightforum.com/comment/1164554#comment-1164554
  
  eBay seems to have many 3A 'Charging & Data Sync Cables' with magnetic heads sub US$2, but no resistance specs given.
  Some claim a 540° head, and I see one with 7 contacts, which just means redundant pairs on 3 of the 4 cables. That should help the resistance.
• Peter Jakacki Posts: 10,193

  2020-09-14 01:52 edited 2020-09-14 02:14

  jmg wrote: »

  Those seem quite high voltage drops ?
  Google finds this table, where 55~70mΩ is at the 'good' end, and 255mΩ is at the 'poor' end, tested at 2A currents.
  
  http://budgetlightforum.com/comment/1164554#comment-1164554
  
  eBay seems to have many 3A 'Charging & Data Sync Cables' with magnetic heads sub US$2, but no resistance specs given.
  Some claim a 540° head, and I see one with 7 contacts, which just means redundant pairs on 3 of the 4 cables. That should help the resistance.

  
  Ooops! The +5V wire drop didn't seem right. My bad, I inadvertently measured the +5V voltage through a Schottky on a P2D2 board. Make that a 138mv drop for the +5V wire itself on the short white cable that Parallax use.
  
  The fact is that the Daiso cable has the same current capacity despite being 3 times longer and flexible. Overall these cables will drop 300mv at 1.1A which leaves plenty of headroom for the P2 switchers although they will draw more current at lower input voltages.
  If 1.8V draws 1A and 3.3V draws 1A, which are rather high figures then the total power is 5.1W times loss at say 90% efficiency is around 5.66W or 1.13A at 5V which is very close to my test conditions but not totally, because I measure 4.72V at the microUSB end but even before the Schottky drop to the regulator input.
  I will do some more comprehensive measurements sometime with maybe the load resistors across the 1.8V and 3.3V outputs instead.
  
  In summary: use these cheap Daiso cables with the P2.
  
  Measurements taken sequentially with 4ohm load and PSU adjusted to ensure 5V at USB source.
  
  DAISO #3003CA microUSB 100cm metalflex
  SOURCE: 5V
  LOAD: 1.1A
  GND DROP: 137mV
  +5 DROP: 155mV
  FINAL: 4.71V
  
  PARALLAX P2 microUSB short 35cm white
  SOURCE: 5V
  LOAD: 1.1A
  GND DROP: 122mV
  +5 DROP: 156mV
  FINAL: 4.72V
  
  P.S. The original USB spec was drawn up by Intel and Microsoft and others and specifed 5V as the source voltage. Having worked with supplying power over buses I already knew that this was a big mistake unless you have heavy and very short cables. They should always have gone with +12V as even then it was very easy to switch down, but not as easy to switch up at around the same voltages. At 5V the current draw is much higher and so the losses are much higher than compared to +12V. At a higher voltage you can carry more power with less losses or you can at least get away with cheap cables. That is why AC is distributed over long distances at high voltages. Now the USB-C spec allows for higher voltages, although through a complex arrangement after starting at 5V. Such shortsightedness.
• Cluso99 Posts: 18,069

  2020-09-14 03:20 edited 2020-09-14 03:20

  Peter Jakacki wrote: »

  P.S. The original USB spec was drawn up by Intel and Microsoft and others and specifed 5V as the source voltage. Having worked with supplying power over buses I already knew that this was a big mistake unless you have heavy and very short cables. They should always have gone with +12V as even then it was very easy to switch down, but not as easy to switch up at around the same voltages. At 5V the current draw is much higher and so the losses are much higher than compared to +12V. At a higher voltage you can carry more power with less losses or you can at least get away with cheap cables. That is why AC is distributed over long distances at high voltages. Now the USB-C spec allows for higher voltages, although through a complex arrangement after starting at 5V. Such shortsightedness.

  That's what you get from non-engineering companies. Intel can design state-of-the-art chips, but not so good at other electronics engineering. MS, well they haven't been very successful in most of their hardware ventures.
  <waits for the flack>
  
• evanh Posts: 15,187

  2020-09-14 04:18

  I'm the first to criticise USB - usually on reliability and protocol complexity - but there's no way that anything other than 5 Volts was going to be the power rail voltage. DC/DC converters were not considered a cheap solution at the time. If USB had been introduced a little latter then the voltage might well have been 3.3 Volts instead. Phone charging and power delivery to anything other than mouse and keyboard was not on check list of requirements. 0.5 Amp was considered a large amount.
  
  
• Peter Jakacki Posts: 10,193

  2020-09-14 04:41

  evanh wrote: »

  I'm the first to criticise USB - usually on reliability and protocol complexity - but there's no way that anything other than 5 Volts was going to be the power rail voltage. DC/DC converters were not considered a cheap solution at the time. If USB had been introduced a little latter then the voltage might well have been 3.3 Volts instead. Phone charging and power delivery to anything other than mouse and keyboard was not on check list of requirements. 0.5 Amp was considered a large amount.
  

  As I said, shortsightedness. I was using switch-mode supplies since forever and many cheap solutions such as the MC34063 were available at the time. However for stuff like mice and keyboards which were relatively low current you only had to use a linear regulator. Dropping up to 7V at 50ma is not a problem. For a full 2.5W load (as per the 5V @500ms) this is only around 200ma at 12V and many devices could use 12V directly, but nonetheless for these higher power devices it was worth it to switch down if necessary. In which case the standard 500ma port would be be able to supply 6W, not 2.5W.
  
  A 100mohm cable will drop 200mV over both ground and power wires at 500ma and that's already 4% of the 5V supply whereas the same cable and power would only drop 42mv using a 12V supply which is just 0.35% of the supply.
  
• evanh Posts: 15,187

  2020-09-14 04:49

  Wasn't short sighted at all. USB has morphed, and rather stupidly so. Expectations are now ridiculous.