| Electronics > Projects, Designs, and Technical Stuff |
| Evolution of usb connector? |
| << < (3/18) > >> |
| coppercone2:
its small compared to a gpib or something sure but i am surprised they did not just try to shrink it all the way on the first iteration i have seen many engineered products that they make absolutely tiny first generation for chic reasons then they get bigger because its too difficult to assemble and stuff. I usually do it too, try to make it as small as possible when I make it but then I end up with a bunch of engineering design unknowns that just make me make it bigger to reduce the number crunch and parts sourcing (i.e. how much do you know about tiny rf connector reliability, durability ,etc.. too much research for me, so you just go with a SMA rather then one of those microscopic ones), or like durability of tiny capacitors etc.. you end up with a shit load of research if you go to small with electronics. is it like that with connector pins and crimps and stuff too? Like a different materials behavior regime is entered? like i wanted to go with a bunch of 0201 parts and stuff before but I started getting paranoid about the manufacturing process. I want to know why they might have done what I do if thats how it went down. you make it sound like someone just said yea its like a order of magnitude smaller good enough (and they knew exactly what they needed and how it would behave if it was made smaller and it was a cost decision). It sounds like its a function decision based on what james_s said but I still think those microusb fit better but I never broke one before. all the big USB ones feel pretty sloppy (but I think the original barn house shape one seems to wiggle less ) . I guess the old ones are just better from a gross physical abuse standpoint (I am pretty careful with my connectors). But I did notice with the normal rectangle USB connectors before, you can break the plastic tab out inside of them so you just have a bunch of floating pins in the connector, it still works but you can plug in something backwards when this happens, the new shape prevents this unless you use pliers to deform it like crazy, since its a metal on metal interface rather then plastic tab being a direction control (it seems careless because a DSUB is by default protected by this, the USB was a step down because of materials choice and stress points until it got shrunk where the stress points were moved to a chassis exterior so it was under expansion strain rather then torsion or whatever)on the plastic tongue. |
| Mr. Scram:
That's why I asked how old you are. It seems you don't have a concept of the demands of that era. Printers were connected by huge connectors. Hard drives and disk drives were connected with what are essentially regular flat cables with the 2,54 mm pin spacing we're familiar with in electronics. There was no push or reason to shrink the connectors down that much. The technology around simply didn't require it and it didn't line up with the existing production capabilities. Devices weren't portable or tiny. Why would you need a tiny connector for your printer? I'd even ask you the reverse. Why do you think they'd push for connectors that tiny? The connector they designed for mice and other peripherals is still used and dominant today. That's a well chosen and engineered solution if you ask me. Why would they have pushed for more until ever shrinking mobile devices started demanding it? A good engineer makes it as good as it needs to be, and no better. Better yet, if you look at changes in data and power capabilities, the current iteration of USB has been shaped by iteratively fulfilling the demands of the time. |
| ejeffrey:
Lithium ion batteries weren't really a thing for consumer electronics when USB was standardized. Nor was flash storage. The early use cases were bulky external drives (including ZIP drives if you remember those), printers, modems, and that sorts of thing. For the most part they weren't intended to be super portable -- they were intended to replace the zoo of existing connectors on desktop PCs -- plugged in once and left connected or moved only occasionally. The chunky nature and relatively strong retention force were considered positives for that application. Low-speed devices like computer mice and most other input devices were required to have captive cables so they don't have a B connector at all rendering the connector size moot. Even when mini-B was created for things like cameras, plugging it in wasn't a daily activity for most people (pros used firewire). It wasn't until the iPod where Apple really pushed the model of the lithium ion battery recharged in the device with a single cable for charge and data that really made the case for micro-B. I still like to see full-size B connectors on devices like printers that have the space. They have much higher pull strength and take a lot more abuse. USB-C is considerably better than micro or mini-B, but the original full size B is a pretty nice connector. I imagine that USB-C will continue to displace A and B even in applications where size isn't a premium, and I guess that is OK, but kind of disappointing. As to why they didn't make a cylindrical symmetric connector like TRS jacks: those are horrible for hot plugging or power delivery as you will short signals together while inserting and removing. That means connecting power pins to data pins or +5V to ground. It is possible to design that to work, but much simpler to go with a keyed connector that only makes contact the right way. Also, the goal was to be A) universal, and B) incompatible with any existing standard. Looking too much like an audio jack or DC barrel jack would have been considered a big disadvantage. |
| coppercone2:
i posted the question you are asking me in my original post, it is based solely on the specification or have connector manufacturers learned anything new since the field testing of regular USB for a decade or something that was required for further miniaturization. would they have pulled it off knowing what they knew about connector physics behavior using the engineering methods at the time? or does it require some kind of new methods of finite element analysis etc to do what we have now. i want to know if it was 'my boss told me to do this' or 'we told the boss this is what we can do without serious development research'. based on the answers its based on what the boss thought reasonable not that they had some kind of question that was answered (like in semiconductors they have a ton of questions relating to miniaturization of processes). I don't know much about the behavior of small stampings and plating on deformable surfaces etc. I thought maybe they needed emperical test data and r&d time to make something smaller thats good for these frequencies. And I don't know much about plastics either. like in the 1920's they could match the geometry of modern cutting tools but without knowledge of advanced plating processes like special nitride coatings and thermal analysis they probobly could not solve it easily.. there is some magic to mechanical development in this way where manufacturing processes allow certain directions of thought to seem rational. it would be a los alamos effort vs seeing 'yea we can figure this out with the new solid works and a few experiments'. Did they need to see USB 1.0 work for a decade before they realized it would be reliable to make it smaller? I think thats a very different line of thought then 'there is no point to doing it another way at this time but we certainly can'. I just want to know about the evolution of the 'physical limits' (within mass production) in the connector field. And if there were any other lines of feedback other then specification (politics vs physics/economics). Was it like 'cutting edge stuff' that they did not really know the behavior of exactly when they first made the USB? I notice its much different then most electrical connectors because they use very thin little fingers/tabs/stampings/cuttings instead of the simple cylindrical pin structure thats been much well studied over the last 100 years. And the interface between the solder and the planar tab vs solder and a pin, ( or crimp). Does it tie into the evolution of crimp connectors? Are the behaviors known good enough now that the interface physics are becoming a dead, uninteresting field to a researcher? Kinda like would Subway restaurants be successful without kickass developments in ovens (that thing is a beast). |
| james_s:
Why are you so obsessed with smaller connectors? USB connectors were some of the smallest connectors you'd find on any computer equipment of the era, replacing many 25 pin and larger connectors. The standard centronics 50 pin scsi connector makes GPIB look small. The original size A and B connectors are still pretty standard on stuff like printers, optical drives, scanners and 3.5" hard drives. They're much more robust than the mini and micro stuff and easier to handle. They're superior in every way except for portability, so why on earth would they have made smaller USB connectors years before tiny portable devices existed? USB had been around more than a decade before I ever saw it used solely to charge or power a device other than gimmicky lights and fans. |
| Navigation |
| Message Index |
| Next page |
| Previous page |