USB connectors are rated at 500ma tops and the wire in their cables accordingly.
I Googled around and found that you can power the Raspberry Pi via the GPIO as an alternative. It would seem that putting 1amp of power into a 500ma cable and 500ma USB port is asking for problems. And the Version B board is rated at 700ma.
Why modify a 500ma rated device and hope for more? The wire resistance may be a negative factor.
So I would probably put 1000ma or more into it via the GPIO.
USB connectors are rated at 500ma tops and the wire in their cables accordingly.
Quick look on farnell site, HiRose ZX micro-USB connector range rated at 1A on data pins, 1.8A on power pins, 30mOhm contact resistance, gold plated. That'll take 700mA fine.
Don't confuse micro USB with mini USB connectors - the micro ones are mechanically stronger, electrically better and smaller... [ according to some article I read about micro-USB becoming the mobile-phone standard power socket, IIRC ]
Well, I am happy the USB connector for the Raspberry Pi is rated far beyond 500ma - it needs to be.
But if someone salvages a USB cable, the wire in the cable may bottleneck the whole set up. After all, to a cable manufacturer with a USB specification of 500ma, there is no reason to provide copper wire for more than 500ma.
Added to that is the problem that keyboard manufacturers may be sloppy in rating the actual power draw of their keyboards as they previously worked fine on a 500ma USB port. They may just pick a lower number for sales advantage as there are an excess of keyboard makers at the low end of the market.
@tingo
Thanks for an objective report of the cost.
BTW, I don't think putting a VOM on a salvaged USB cable is going to provide a useful reading. My own VOM is rather useless below 100ohms.
It looks like the USB spec specifies 28-20 AWG wire for the cable. This would provide a curretn capacity ranging from 830ma to 7.5 amps according to the wire gauge tables for ampacity that I could find quickly.
Good, but specs are specs. Manufacturers have real out of pocket costs. A penny saved is a penny earned. And that is all copper.
My attitude is brought on by the fact that this is a Parallax hosted forum and the 2800 views of this thread have long touted this non-Parallax product to be far superior in aspects that cannot possibly be true.
Several people are very quick to refute everything I say with cleverness, but not with pragmatic engineering experience. As long as the thread keeps taking advantage of Parallax's hospitality, I suspect you will hear from me from time to time.
Look at the Raspberry Pi forums and all the threads that are dismayed with the keyboard interface not working. I gave you a simple, direct solution and you argue on. Use the GPIO interface and make sure your wire is thick enough.
Of course, if the actual power traces on the Raspberry Pi board are inadequate, the whole project may be moving on to Board C and another price increase.
No idea how it goes with the raspi but my IGEPv2 ARM boards from ISEE only have USB v2.0 support. So USB v1.0/1 keyboards and mice do not work. Hence the need for a USB v2.0 HUB to act as a converter.
And WOOHO! Today Farnell tells me "your Raspberry Pi will be dispatched the week commencing09/07/2012" and RS have promised delivery of one within 10 weeks!
I was just commenting on what the spec said - yes, specs are specs and the real world is driven by pennies. I'm not refuting/arguing/defending or debating anything although I do try to be clever!
I've ordered a RPi out of curiousity, probably to sit next to my BeagleBone to see what it can do with a Propeller at some point. I don't think my daughter will become a genius computer engineer because I have one and let her use it. I don't I'll revolutionize the STEM programs in our local school district if I show it around the Superintendent's office. It does make you curious though and interesting to see where it goes.
I have an August 16th ship date according to Element-14. I'm looking forward to Propeller Pi too (mostly because it sounds better than Propeller Bone or Beagle Props)!
I've ordered a Trust ClassicLine keyboard, which is supposed to be one which works OK, from Rapid Electronics. They are very cheap. I should get it tomorrow.
I realize that the Raspberry Pi looks fun, and may really be fun for someone that already knows Linux in depth.
These days, it is nearly impossible to buy a 15amp 120VAC extension cord with 15amps worth of copper in it.
These days, a 20watt stereo boom box is never really 20 watts.
These days, a Euro isn't worth what it was last year.
I used to build homes as a carpenter and the 100' 15amp extensions cords had to be able to provide a real 15 amps or you spent all day at the circuit breaker box. One buys quality and enjoys the durability. That is the reason that the Propeller has 40ma i/o, though everyone else provides nearer to half that. It is so that the fun keeps on happening, even when the user goofs.
Superlatives have everything to do with sales and hype, very little to do with education.
...this thread have long touted this non-Parallax product to be far superior in aspects that cannot possibly be true.
There are no such statements in this thread. Except perhaps a debate over if using a full up Linux system or a simpler micro-controller is better as an educational tool for an introduction to programming, electronics etc. I say lets have both!
...As long as the thread keeps taking advantage of Parallax's hospitality
I will certainly take advantage of Parallax's hospitality because I have a plan...The "Propeller Pi". What better place to discuss it?
Look at the Raspberry Pi forums and all the threads that are dismayed with the keyboard interface not working.
Firstly I want to see all the nice new opensource Propeller tools up and running on the raspi. propgcc, the SimpleIDE, the new opensource Spin compiler etc.
Next I want to see a board of Raspberry pi form factor that will stack on top (or below I guess) with one or more Props on it, as a minimum all the Prop pins should be available from there.
Combined we have a tiny computer with lots of general compute power, ethernet, USB, files store etc from the Pi and a whole bunch of real time I/O and processing capability via the Prop(s)
AND it's a stand alone unit on which all it's software can be developed.
I think one of Jazzed's TetraProp boards re-footprinted to fit on top of the Raspberry Pi or Beagle bone would be a formidable pairing. There's a lot of Prop goodness with 4 on a Propeller Platform footprint currently and a bunch of exposed I/O pins. They clock nicely at 100MHz, too.
Once we get GO and/or Forth up and running on the RPi, the world will be a better place!!
I'd be happy with just all the Propeller tools running on the Pi and other ARM devices. The formidable pairing of ARM and Prop has been on my mind for ages.
As the good book says, we should GO Forth and multiply:)
A GPIO interface to a Propeller driven board would definitely enhance both.
The Propeller would have a LAN interface, and off loading the video and keyboard would allow all the space on the Propeller to be dedicated to parallel tasking other stuff - like motors and sensors. Plus, the HMDI video would be more sophisticated than the VGA.
But what I tend to love about Parallax products is that it is possible to comprehensively know all the code that goes into something without making a lifetime career out of it.
I am still learning Linux and loving it, but it demands keeping up on a rather wide variety of topics.
I agree, the ability to know what every byte of code in your machine is for is wonderful, still we need dev tools and they run on PC where you will only ever know a vanishingly small part of what is going on. Might as well run them on a small and cheap ARM board.
For starters it isn't from the Raspberry site but from one the distributors - RS. And yes it's not a Raspberry. If the Raspberry is alive a year from now I probably buy one. At least by then there will be software for it.
That is my stance, too. The RasPi seems to be a cool thing, and I'm not bothered by its discussion here. But I'm more than happy to let others pave the way. I'd like to see where time and talent take the project. Perhaps by next year I'll actually have a use for such a board.
Edit: after seeing this link of Heater's, Propeller Pi is beginning to make a lot of sense.
The "other" chip is a LAN controller. Otherwise it's a SOC computer.
The only whoops in this is, that production and distribution are a ongoing train wreck and probably costing Raspberry a lot of potential customers, but that's what you get when you don't involve people with that expertise. Heck they can't even get out a simple I/O board after promising to release one at the end of 2011, they are just releasing prototypes now. The board is so simple that almost any hobbyist with a PCB program could design it and send it off to Itead in a week or two. Not six-twelve months.
Rod1963,
If such a simple board for 30 odd dollars is so easy to create and market how come they are the first ones to do it?
There are of course similarly capable ARM boards like my IGEPv2 from ISEE or the Beagle board but at many times the price.
Don't forget they are a charity doing this on a shoe string budget not a huge company like TI.
For sure they were not expecting such an overwhelming demand for the first run, if ever. They have had to learn how to cope with this as they go along and it seems they have actually learned well.
Some are absolutely sold on the Raspberry Pi -- blinded by guts and glory. ( God save the Queen! Hail Britannia! and all that...........) {Thar be mad dogs and Englishmen walking amongst us.}
Broadcomm has a good chip and it certainly can easily support a PDA or an iPad type touch screen. It certainly is a rather complete Computer on a Chip.
Comparisons to the Propeller and Basic Stamp tend to presume the biggest, fastest in the tiniest of packages is the best for everything.
But Linux is based on a multi-user file system and has a lot more overhead than many microcontroller projects require - especially in an educational context where not having a large file system and library of support items can be quite useful. These system requirements tend to distract the student from what can truly be done with very little code.
In education, one has to crawl, then walk, then run. HDMI video is quite wonderful for those that know how to exploit it; but learning analog video and VGA on the Propeller are a much simpler presentation for the student. The Propeller actually presents the true simplicity of video generation in assembler - which is generally necessary to get the optimal speeds. Parallax also tends to do things in software that others do in silicon and thus eliminate one's dependency of buying a chip with all the hardware features that one requires and just writing code to particular registers to make it work. One learns actual algorithms for I2C, SPI, RS232, buffers, stacks, and so on.
Some of us believe 'everything in moderation' and ofter a contrasting view. Personally, I dislike the having to get on a list an wait - now at 14 weeks. BUT, I have registered as it may be pretty good in 14 weeks and people are now sorting out the real user problems. It is rather disappointing that one might have to use a powered USB hub to attach many useful USB devices as the board design tried to accommodate two USB ports on very little power.
It is also a problem that these two USB ports are running through the LAN chip and users are having trouble with LAN interface at this point. More shall be revealed.
Admittedly it is sometimes worthwhile to throw away a little money to learn what is going on. For the curious and more informed, the Raspberry Pi will provide a bit of educational contrast to other devices. At least, you don't have to hack an Apple iPad to explore this leading edge.
T... production and distribution are a ongoing train wreck and probably costing Raspberry a lot of potential customers,...
Sorry, but this is a silly assessment. Going from ten hand-built prototypes to hundreds of thousands sold and shipped in a few months by a group that "doesn't have that expertise" is a BIG deal. Pretty much shows the folks with "that expertise" how its done, I think. Just as the OLPC drew the the declaration that a sub $1000 laptop would be "impossible" yet started the whole trend towards the netbooks we see today; the Raspberry Pi is about to start a new paradigm shift in person computing devices, or new product development, or both.
But I couldn't find anyone to do a port of the prop GO CSP-Channels (GO-Channels and Propforth MCS) to Python (http://code.google.com/p/python-csp/). Python CSP and Prop MCS.
Maybe when enough folks have RPi, (and we get the GO Channels to propforth demo working on RPi) somebody will take interest in the python interface.
Comments
I Googled around and found that you can power the Raspberry Pi via the GPIO as an alternative. It would seem that putting 1amp of power into a 500ma cable and 500ma USB port is asking for problems. And the Version B board is rated at 700ma.
Why modify a 500ma rated device and hope for more? The wire resistance may be a negative factor.
So I would probably put 1000ma or more into it via the GPIO.
Raspberry Pi, model B: NOK 263.19
VAT and handling charges: NOK 192.-
usb charger: NOK 99.-
usb A to b micro cable: NOK 69.-
HDMI cable: NOK 150.-
a total of NOK 773.19. At the current exchange rate of 6.035, that is about USD 128.12
Quick look on farnell site, HiRose ZX micro-USB connector range rated at 1A on data pins, 1.8A on power pins, 30mOhm contact resistance, gold plated. That'll take 700mA fine.
Don't confuse micro USB with mini USB connectors - the micro ones are mechanically stronger, electrically better and smaller... [ according to some article I read about micro-USB becoming the mobile-phone standard power socket, IIRC ]
But if someone salvages a USB cable, the wire in the cable may bottleneck the whole set up. After all, to a cable manufacturer with a USB specification of 500ma, there is no reason to provide copper wire for more than 500ma.
Added to that is the problem that keyboard manufacturers may be sloppy in rating the actual power draw of their keyboards as they previously worked fine on a 500ma USB port. They may just pick a lower number for sales advantage as there are an excess of keyboard makers at the low end of the market.
@tingo
Thanks for an objective report of the cost.
BTW, I don't think putting a VOM on a salvaged USB cable is going to provide a useful reading. My own VOM is rather useless below 100ohms.
My attitude is brought on by the fact that this is a Parallax hosted forum and the 2800 views of this thread have long touted this non-Parallax product to be far superior in aspects that cannot possibly be true.
Several people are very quick to refute everything I say with cleverness, but not with pragmatic engineering experience. As long as the thread keeps taking advantage of Parallax's hospitality, I suspect you will hear from me from time to time.
Look at the Raspberry Pi forums and all the threads that are dismayed with the keyboard interface not working. I gave you a simple, direct solution and you argue on. Use the GPIO interface and make sure your wire is thick enough.
Of course, if the actual power traces on the Raspberry Pi board are inadequate, the whole project may be moving on to Board C and another price increase.
No idea how it goes with the raspi but my IGEPv2 ARM boards from ISEE only have USB v2.0 support. So USB v1.0/1 keyboards and mice do not work. Hence the need for a USB v2.0 HUB to act as a converter.
And WOOHO! Today Farnell tells me "your Raspberry Pi will be dispatched the week commencing09/07/2012" and RS have promised delivery of one within 10 weeks!
Then it's Propeller Pi time:)
I've ordered a RPi out of curiousity, probably to sit next to my BeagleBone to see what it can do with a Propeller at some point. I don't think my daughter will become a genius computer engineer because I have one and let her use it. I don't I'll revolutionize the STEM programs in our local school district if I show it around the Superintendent's office. It does make you curious though and interesting to see where it goes.
These days, it is nearly impossible to buy a 15amp 120VAC extension cord with 15amps worth of copper in it.
These days, a 20watt stereo boom box is never really 20 watts.
These days, a Euro isn't worth what it was last year.
I used to build homes as a carpenter and the 100' 15amp extensions cords had to be able to provide a real 15 amps or you spent all day at the circuit breaker box. One buys quality and enjoys the durability. That is the reason that the Propeller has 40ma i/o, though everyone else provides nearer to half that. It is so that the fun keeps on happening, even when the user goofs.
Superlatives have everything to do with sales and hype, very little to do with education.
The Plan:
Firstly I want to see all the nice new opensource Propeller tools up and running on the raspi. propgcc, the SimpleIDE, the new opensource Spin compiler etc.
Next I want to see a board of Raspberry pi form factor that will stack on top (or below I guess) with one or more Props on it, as a minimum all the Prop pins should be available from there.
Combined we have a tiny computer with lots of general compute power, ethernet, USB, files store etc from the Pi and a whole bunch of real time I/O and processing capability via the Prop(s)
AND it's a stand alone unit on which all it's software can be developed.
Propeller Pi.
I think one of Jazzed's TetraProp boards re-footprinted to fit on top of the Raspberry Pi or Beagle bone would be a formidable pairing. There's a lot of Prop goodness with 4 on a Propeller Platform footprint currently and a bunch of exposed I/O pins. They clock nicely at 100MHz, too.
Once we get GO and/or Forth up and running on the RPi, the world will be a better place!!
I'd be happy with just all the Propeller tools running on the Pi and other ARM devices. The formidable pairing of ARM and Prop has been on my mind for ages.
As the good book says, we should GO Forth and multiply:)
The Propeller would have a LAN interface, and off loading the video and keyboard would allow all the space on the Propeller to be dedicated to parallel tasking other stuff - like motors and sensors. Plus, the HMDI video would be more sophisticated than the VGA.
But what I tend to love about Parallax products is that it is possible to comprehensively know all the code that goes into something without making a lifetime career out of it.
I am still learning Linux and loving it, but it demands keeping up on a rather wide variety of topics.
Great, sounds like we are winning you over:)
I agree, the ability to know what every byte of code in your machine is for is wonderful, still we need dev tools and they run on PC where you will only ever know a vanishingly small part of what is going on. Might as well run them on a small and cheap ARM board.
Ah yes, except perhaps......
Are you winning me over, not really. I see now that the LAN interface is having trouble. I'll just wait and see.
The first problem is it is all in English. The second is the link to register is broken. And.... isn't that a BeagleBoard?
Edit: after seeing this link of Heater's, Propeller Pi is beginning to make a lot of sense.
well whats the problem with only one other chip, can you connect sensors xbee etc to it ? Whoops caps.
The only whoops in this is, that production and distribution are a ongoing train wreck and probably costing Raspberry a lot of potential customers, but that's what you get when you don't involve people with that expertise. Heck they can't even get out a simple I/O board after promising to release one at the end of 2011, they are just releasing prototypes now. The board is so simple that almost any hobbyist with a PCB program could design it and send it off to Itead in a week or two. Not six-twelve months.
If such a simple board for 30 odd dollars is so easy to create and market how come they are the first ones to do it?
There are of course similarly capable ARM boards like my IGEPv2 from ISEE or the Beagle board but at many times the price.
Don't forget they are a charity doing this on a shoe string budget not a huge company like TI.
For sure they were not expecting such an overwhelming demand for the first run, if ever. They have had to learn how to cope with this as they go along and it seems they have actually learned well.
Broadcomm has a good chip and it certainly can easily support a PDA or an iPad type touch screen. It certainly is a rather complete Computer on a Chip.
Comparisons to the Propeller and Basic Stamp tend to presume the biggest, fastest in the tiniest of packages is the best for everything.
But Linux is based on a multi-user file system and has a lot more overhead than many microcontroller projects require - especially in an educational context where not having a large file system and library of support items can be quite useful. These system requirements tend to distract the student from what can truly be done with very little code.
In education, one has to crawl, then walk, then run. HDMI video is quite wonderful for those that know how to exploit it; but learning analog video and VGA on the Propeller are a much simpler presentation for the student. The Propeller actually presents the true simplicity of video generation in assembler - which is generally necessary to get the optimal speeds. Parallax also tends to do things in software that others do in silicon and thus eliminate one's dependency of buying a chip with all the hardware features that one requires and just writing code to particular registers to make it work. One learns actual algorithms for I2C, SPI, RS232, buffers, stacks, and so on.
Some of us believe 'everything in moderation' and ofter a contrasting view. Personally, I dislike the having to get on a list an wait - now at 14 weeks. BUT, I have registered as it may be pretty good in 14 weeks and people are now sorting out the real user problems. It is rather disappointing that one might have to use a powered USB hub to attach many useful USB devices as the board design tried to accommodate two USB ports on very little power.
It is also a problem that these two USB ports are running through the LAN chip and users are having trouble with LAN interface at this point. More shall be revealed.
Admittedly it is sometimes worthwhile to throw away a little money to learn what is going on. For the curious and more informed, the Raspberry Pi will provide a bit of educational contrast to other devices. At least, you don't have to hack an Apple iPad to explore this leading edge.
Sorry, but this is a silly assessment. Going from ten hand-built prototypes to hundreds of thousands sold and shipped in a few months by a group that "doesn't have that expertise" is a BIG deal. Pretty much shows the folks with "that expertise" how its done, I think. Just as the OLPC drew the the declaration that a sub $1000 laptop would be "impossible" yet started the whole trend towards the netbooks we see today; the Raspberry Pi is about to start a new paradigm shift in person computing devices, or new product development, or both.
I know python is popular, and the Chicago hackspace PS-1 is even has classes going http://pumpingstationone.org/2011/03/introduction-to-programming-with-python/
But I couldn't find anyone to do a port of the prop GO CSP-Channels (GO-Channels and Propforth MCS) to Python (http://code.google.com/p/python-csp/). Python CSP and Prop MCS.
Maybe when enough folks have RPi, (and we get the GO Channels to propforth demo working on RPi) somebody will take interest in the python interface.
I like Python and always have. But I keep falling back to lower levels of programing for my own pursuits.