Evolution of usb connector?

[dmills]

Part of it is probably that nobody at the time would have taken a micro USB connector seriously... I mean printers were usually on 25 pin centronics (about the size of a GPIB connector), usually cabled to a D25 on the PC end, keyboards were 5 pin DIN or mini DIN, mice were D9 serial.

The real mystery with USB, is did nobody actually try plugging in a USB-A before standardising the thing? The number of the stupid things I have tried to stuff in upside down over the years (It always seems to take three tries!). 

Bear in mind that USB was really seen as a sort of more standard (and slightly faster, but filling the same niche) version of the Apple Desktop Bus that used to be the way you plugged keyboards, mice and laserwriters into mackintoshes, portable devices as we now understand them were not even a consideration, and the phones of the time all used gratuitously different connectors anyway.



Regards, Dan.

[Bassman59]

its kind of weird because you would think something like an audio jack is bomb proof because its a giant cylinder

You don't spend any time around professional audio set-ups, do you? 1/4" TRS plugs get broken all the time. Sometimes you gotta pull the parts out of the jack with thin needle-nose pliers. Never mind the fact that the "hot" contact on the plug is briefly shorted to ground on insertion and removal.

[dmills]

Ah, you are talking about the infinitely inferior "A Gauge" audio jack (Fit only for headphones), a proper "B gauge" part in a proper long frame socket is surprisingly reliable (needs occasional cleaning of the normalling contacts, but at least you CAN clean the normalling contacts).

Regards, Dan.
 

[james_s]

Part of it is probably that nobody at the time would have taken a micro USB connector seriously... I mean printers were usually on 25 pin centronics (about the size of a GPIB connector), usually cabled to a D25 on the PC end, keyboards were 5 pin DIN or mini DIN, mice were D9 serial.

The real mystery with USB, is did nobody actually try plugging in a USB-A before standardising the thing? The number of the stupid things I have tried to stuff in upside down over the years (It always seems to take three tries!). 

Bear in mind that USB was really seen as a sort of more standard (and slightly faster, but filling the same niche) version of the Apple Desktop Bus that used to be the way you plugged keyboards, mice and laserwriters into mackintoshes, portable devices as we now understand them were not even a consideration, and the phones of the time all used gratuitously different connectors anyway.


Regards, Dan.

It is slightly less than ideal, if I were designing it from the start I would have given it angled edges like HDMI or some other standardized and distinctive marking or indentation.

[westfw]

did nobody actually try plugging in a USB-A before standardising the thing?

There is a substantial chance that there is plastic molding and/or other indicators, and/or a different vision of type-A ports, than exists in the current realm of non-official cables and "many" USB ports on a computer...

how come the reliability of the mini was bad?

[micro-usb is] more robust than any USB connector version previously made. More mating cycles and mechanically stronger.

So the vendors claim;  I actually talked to one of the big connector vendors at a trade show about this - they claim that one of the big changes is that the spring-loaded part of the connector is now in the (easily replaceable) cable, rather than the PCB-side connector.

Micro USB connectors just don't break.

That statement does not match my personal experience.  They may resist breakage in exactly the ways that they were intended to resist breakage, but other "system-level" issues seem to come up that have been the death of several (otherwise undamaged) products I've owned.   Including some expensive cell phones.  The cables break even more frequently, too (apparently usually due to shoddy strain relief, fine wires, and poor soldering/assembly technique.)

Only variants with no through-hole mounting pins are subject for breaking off from PCB

Unfortunately, the "no through-hole" versions are by far the most common, and in between the lack of TH, the smaller overall area of the pads, and the thinner PCBs, the "break off of PCB" failure mode is very common :-(

[wraper]

Micro USB connectors just don't break.

That statement does not match my personal experience.  They may resist breakage in exactly the ways that they were intended to resist breakage, but other "system-level" issues seem to come up that have been the death of several (otherwise undamaged) products I've owned.   Including some expensive cell phones.  The cables break even more frequently, too (apparently usually due to shoddy strain relief, fine wires, and poor soldering/assembly technique.)

Don't forget those connectors and cables experience daily use and abuse which usually does not happen with full size connectors. There are not that many devices which you drop or pull with cable plugged in. USB-A connectors in laptops get damaged just as well, likely even more often.

Unfortunately, the "no through-hole" versions are by far the most common, and in between the lack of TH, the smaller overall area of the pads, and the thinner PCBs, the "break off of PCB" failure mode is very common :-(

I repaired thousands of phones in the past. Non TH connectors were used a lot by Nokia. Later they started using mostly connectors with TH pins in symbian smartphones. Those were extremely reliable. I no longer repair phones as my daily job, but phones which I occasionally repaired in the last 5 years used SMT+TH connectors.

[stratcat]

USB ports are proof that there are 4 dimensions - try to plug, won't go.
Turn it over, won't go.
Turn it over again, Success! 
    

[james_s]

The micro connectors are ok for phones, because phones are small and light and generally do not exert a great deal of force on the cable. Put a micro USB connector on the back of a 15 pound laser printer and then trip over the cord and I bet it snaps off the plug, rips the connector off the PCB or both. Try the same with a fullsized USB connector and there's a pretty good chance it will spin the printer around and yank the plug out of the socket without damage.

The micro and mini connectors are small out of necessity because they're used on small devices, they aren't used on bulky heavy peripherals because they're not as sturdy.

[Mr. Scram]

you should post this in other peoples threads repeatedly

Thanks for the advice, but I don't think proper forum form is really your area of expertise. It'd still be good to know what you did to investigate this yourself. What did you find about the history of USB? There's probably a lot of interesting stories you've found you could tell us. Don't tell us you didn't investigate?

[coppercone2]

NOTE: This message has been deleted by the forum moderator Simon for being against the forum rules and/or at the discretion of the moderator as being in the best interests of the forum community and the nature of the thread.
If you believe this to be in error, please contact the moderator involved.
An optional additional explanation is:

[wraper]

The micro connectors are ok for phones, because phones are small and light and generally do not exert a great deal of force on the cable. Put a micro USB connector on the back of a 15 pound laser printer and then trip over the cord and I bet it snaps off the plug, rips the connector off the PCB or both. Try the same with a fullsized USB connector and there's a pretty good chance it will spin the printer around and yank the plug out of the socket without damage.

Really? USB-B is among the worst ones. Only two mounting legs located closer to the rear side. Very easy to rip off. Actually I've done just that a few days ago on some defective boards I was going to throw out. Plugged cable in and easily ripped off USB-B connectors. And weight of a phone does not matter much when you grab the phone and pull it to yourself.
The only variant which is significantly stronger than micro USB with TH is USB A with multiple sockets stacked. Those have 4 mounting legs and are very sturdy. BTW metal thickness of micro USB and full size connectors is about the same. Also one important thing is how connector is surrounded/supported by enclosure. It can add a lot of strength.

[wraper]

Also how sturdy do you think are these? Simple school level physics about lever which work against such implementations.

[james_s]

Those PCB mount plugs are pretty bad, but a micro version wouldn't be any better, I don't think they even make PCB mount male micro USB.

The fullsized B connectors normally have the far end supported by the opening in the housing it goes through. You may have had one break but I never have. They're bigger and heavier, with large through hole posts. Has anyone else here ever had one break? Can anyone point me to a printer, scanner, 3.5" hard drive or other bulky USB peripheral that has a micro USB port? Anyone?

[Gyro]

The fullsized B connectors normally have the far end supported by the opening in the housing it goes through. You may have had one break but I never have. They're bigger and heavier, with large through hole posts. Has anyone else here ever had one break? Can anyone point me to a printer, scanner, 3.5" hard drive or other bulky USB peripheral that has a micro USB port? Anyone?

Loss of retention pressure (contact pressure?) is the only thing I've ever seen with them - insertion gets loose.

[wraper]

The fullsized B connectors normally have the far end supported by the opening in the housing it goes through

So do micro USB.

You may have had one break but I never have.

I've seen USB-B broken in a device worth more than $100k. Mounting legs just snapped off.

3.5" hard drive or other bulky USB peripheral that has a micro USB port?

As of current year most external HDD, even 3.5" don't have full sized USB port.

[fsr]

I remember when USB was appearing on different motherboards, and you looked at the computer stores and there were like no devices to plug there! Mostly mouses, and that's it. Well, at least that's what happened here. A bunch of years ago.

And yeah, the rectangular connector of USB is ridiculous, as it's very hard to see in what orientation must the connector go, something that is made still worse, as many connectors are in the back of the PC, and you cannot really look at them very well most of the time.

Currently, another problem is that many connectors have similar dimensions. I think that you can probably force a USB A into the wrong socket.

[james_s]

That wasn't really an issue when USB was new though, well the orientation was always trickier than it ought to be although really no worse than the mini-DIN used for keyboards and mice at the time. When USB was a new thing, it was far smaller and a completely different shape from all the other connectors used on computers. It was very distinctive and there was no confusing it.

Some of the earliest things I remember using it for besides mice were scanners and printers where it replaced the gigantic Centronics-50 SCSI connector and slightly smaller parallel printer cables.

https://goo.gl/images/ShoLVu

https://goo.gl/images/kGHUXa

At the time it was marvelously tiny, not only the connector but the cable too. The thickest USB cables are less than half the diameter of the thinnest printer and SCSI cables I can recall seeing. And you could connect them to a hub too! It may be hard for younger people to comprehend just what a change it was. Suddenly we went from a bunch of different standards using similar looking but incompatible big bulky connectors and thick cables to one slender cable with a single type of tiny connector for almost everything. Then with things like SCSI you had to set a unique ID on each device, fuss around with terminators and make sacrifices to the computer gods in hopes that it would all actually work.

It reminds me of the first time I saw a 2.5" hard drive, it was so small I could hardly believe it held 20MB. It never occurred to me at the time that hard drives could be made yet smaller, and I'd never have believed it possible to fit tends of GB onto a fingernail sized micro SD card as we can today.

[tooki]

IIRC the original USB connectors were based on the connector used for linking the Nintendo Game Boy handheld systems, which had proven it to be a robust design even in the hands of clumsy or rambunctious kids.

The 6-pin FireWire 400 cable is more like the old gameboy connector.

Yep. It is FireWire 400, not USB, that was directly inspired by the GameBoy connector. (I don’t think Nintendo ever got enough credit for pioneering that type of pinless connector. Before it, only large plugs like Centronics used anything like that arrangement, where it’s basically a plastic card-edge connector instead of actual pins.)

Ultimately, if we were going to invent the USB connector today, we’d do all sorts of things differently: it’d be smaller, reversible, have rounded leading edges to go in more easily, and be designed to stand up to the rough handling in mobile devices.

Oh wait, we just did that, it’s called USB-C. (And indeed it’s like a reversible, metal-sheathed mini version of the GameBoy connector.)

[tooki]

i wanna know the engineering process at this point I don't care the politics behind the spec

What you casually dismiss as “politics” ARE the engineering requirements that drove the spec!

[coppercone2]

I should focus down to the metallurgical interface properties and the dielectric durability or whatever else may effect it. Was some kind of novel thing discovered from the first USB connector between the current connector. From a manufacturing process prospective.

All the stuff in this thread is pretty hard but it seems like you don't need to know anything special if you already know about the first USB connector design other then to work out the new mechanics from a structural engineering view point.

I don't know how to put this. is there an interesting secret sauce ? process step? materials choice? finishing steps? extra quality checks? more advanced adhesives? something (physics) weird that came up?

example: with cables something I considered to qualify for the above criterion is the kevlar fiber reinforced USB cables that are for sale now. They use a special weave method (if you trust the package and its too expensive for me to destroy) to make it more durable. IMO thats a innovation rather then a refinement (but its extremely well hidden unless you advertise it). I don't know if it was just borrowed from another industry or came out because of the portable electronics revolution but its pretty cool. Or magnetic connectors, a whole new fairly complicated mechanical design aspect was introduced to connector manufacturing with the magnetic mate. Or those little crimps (molex I think) that crimp on both the wire and insulation with separate wings.

[tooki]

Christ on a cracker, man...

What produced the modern USB connectors (culminating in USB-C) isn’t magic new materials. It’s simply ordinary engineering done to improve upon the shortcomings of what came before. The number of devices with USB today is probably orders of magnitude larger than all the computers in the world in 1995, and the use cases cover things never anticipated before. At this kind of scale, you discover issues that you wouldn’t have otherwise. And the economies of scale involved mean that you can amortize much more expensive R&D than you could have before, so we could spend more money designing better connectors to improve upon the last. And at this kind of scale, manufacturers can afford the more expensive tooling for higher-precision parts.

[coppercone2]

what does their R&D process look like?

[IDEngineer]

Desire to design a universal serial interface to replace the hodgepodge of connectors currently in use such as DB25 and DB9 serial and parallel ports, DB15 game ports, SCSI ports, PS/2 kb/mouse, bus mouse, etc while also providing power to the devices. The standard USB connectors were made to be robust and foolproof, easy to manufacture, sturdy, easy to insert and remove, durable enough to handle loads of insert/removal cycles, designed to connect power before data...

and:

What produced the modern USB connectors (culminating in USB-C) isn’t magic new materials. It’s simply ordinary engineering done to improve upon the shortcomings of what came before. The number of devices with USB today is probably orders of magnitude larger than all the computers in the world in 1995, and the use cases cover things never anticipated before. At this kind of scale, you discover issues that you wouldn’t have otherwise. And the economies of scale involved mean that you can amortize much more expensive R&D than you could have before, so we could spend more money designing better connectors to improve upon the last. And at this kind of scale, manufacturers can afford the more expensive tooling for higher-precision parts.

With these two excellent responses this thread should be done.

I will say, though, that as a hardware/firmware/software Engineer USB has definitely suffered from feature creep. The layers and layers of details necessary to properly understand and implement a simple device are a serious PITA for what I suspect are the majority of devices that just need a modest rate data path to/from another device. Used to be you could just include an RS-232C port, with its mind-numbingly simple hardware and software, and you had connectivity to 99% of the world's devices. With USB things aren't so simple anymore, and while it's nice to get power from that single cable it comes at the overhead of drivers and endpoint negotiation and much more - on both ends of the wire. YMMV.

[westfw]

Used to be you could just include an RS-232C port, with its mind-numbingly simple hardware and software, and you had connectivity to 99% of the world's devices.

I'm sorry... ARE YOU OUT OF YOUR MIND???!!!




Are you talking about the same +/-12V requiring "rs232" with a multitude of ambiguous, poorly defined, large and expensive connectors that used cables with a randomish number of conductors and signals that were NEVER "right", requiring the average IT person to have a drawer full of "gender changers", "null modems", DE9-to-DB25 adapters, test boxes that allowed you to see and jumper signals as needed, and additional random "improved" higher-density connectors that varied from vendor-to-vendor, that tortured the very souls of many an IT professional for the decades before USB came out?
The connector that Certain Major PC manufacturers decided to re-used for a parallel printer port that needed different signals and cables?




The specification defined specifically for a MODEM and a TERMINAL or mainframe, that made about as much sense as charging for CPU time in the days of personal computers with peripherals like mice and printers and such?   The one spec'ed at 19200bps max speed (over short distances, of course.)  With such poor data reliability that there were at least three different common schemes for providing flow control (one of which completely threw away portions of the "specification"?




Sure, a UART is conceptually simple.  A Serial card was a relatively inexpensive add-on for the original IBM-PC; only 12 chips or so.  And if you got one of those newer high-speed modems and wanted to run that new-fangled "windows" OS, they'd ship you a NEW card with a deeper-FIFO UART?   (For free, to stop customer complaints that the modems didn't work.)  (and custom software, because DOS and/or the PC BIOS had crap support for UARTs.)




Nowadays even very cheap microcontrollers come with built-in UART hardware.  But that UART will talk to a USB/Serial adapter just as easily as it will talk to rs232 drivers and  connectors, probably at lower cost, and certainly at less stress for both the PC user and the device user (and the support personnel.)




USB has become a bit cluttered with feature-itis, and maybe some cronyism. But in its original form at only 11Mbps and with the "huge" connectors, it was a pretty brilliant solution to real problems that were plaguing the industry.

[james_s]

USB is only simple because you can think of it as a black box, both from a hardware and software standpoint. A serial port is still simpler though, simple enough that it's easily understood at a very low level. Yes there are a lot of variations but in the most basic setup you only need rx, tx and ground. Whether you're wanting to talk to the console of a modern smartphone or a 1970s mainframe RS232 just works. No drivers, no handshaking, you can put a scope on it and see the bits, decode them by hand if you like. There's a reason that RS232 and the TTL variant are still ubiquitous. You may not find a DB25 or DB9 on the back of a modern PC but chances are there's a serial port in there somewhere, even if it's just some test pads on the PCB.