Searching \ for '[PIC]: Getting started with USB' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: techref.massmind.org/techref/microchip/begin.htm?key=pic
Search entire site for: 'Getting started with USB'.

Exact match. Not showing close matches.
PICList Thread
'[PIC]: Getting started with USB'
2006\02\10@030700 by Philip Pemberton

face picon face
Hi,
 I've just ordered a pair of PIC18F4550s and I've realised that I don't know
the first thing about interfacing with the USB bus... (a bit silly to order
the chips before learning about the hardware, but that's me in a nutshell -
"a bit silly").

 Does anyone know of any decent tutorials on USB, how it works and how it
can be used with the newer USB PICs?

 I eventually want to throw together a USB-to-GPIB bridge for my laptop.
I've seen the one that was published in EDN, but I want to do away with the
FTDI chip and hopefully speed up the interface as well.

Thanks.
--
Phil.                         | Kitsune: Acorn RiscPC SA202 64M+6G ViewFinder
spam_OUTphilpemTakeThisOuTspamdsl.pipex.com         | Cheetah: Athlon64 3200+ A8VDeluxeV2 512M+100G
http://www.philpem.me.uk/     | Tiger: Toshiba SatPro4600 Celeron700 256M+40G
... BREAKFAST.COM Halted... Cereal Port Not Responding.

2006\02\10@032900 by Chen Xiao Fan

face
flavicon
face

> -----Original Message-----
> From: .....piclist-bouncesKILLspamspam@spam@mit.edu
> [piclist-bouncesspamKILLspammit.edu] On Behalf Of Philip Pemberton
> Sent: Friday, February 10, 2006 4:09 PM
> To: .....piclistKILLspamspam.....mit.edu
> Subject: [PIC]: Getting started with USB
>

Here are some of the links.

http://www.lvr.com one of the best resource with USB
http://www.usb.org check out the documents and the forum
http://www.microchip.com/usb download the USB firmware
http://forum.microchip.com check out the USB section
http://pe.ece.olin.edu/ece/ assmebly and C18 based USB firmware
http://pic18fusb.online.fr/wiki/ 18F USB PIC wiki


Regards,
Xiaofan

2006\02\10@045540 by Jan-Erik Soderholm

face picon face
> http://www.lvr.com one of the best resource with USB

And don't forget the book !
I have the 2'nd edition (now 3'rd) and it's realy good.

Jan-Erik.



2006\02\10@062935 by Alan B. Pearce

face picon face
>I eventually want to throw together a USB-to-GPIB bridge
>for my laptop. I've seen the one that was published in EDN,
>but I want to do away with the FTDI chip and hopefully
>speed up the interface as well.

Ditto.

I have had a look at the C source (did you download that?) and pondered
wether it would compile in C18 with minimal errors.

The hardest part would then be to make the 18F4550 look like an FTDI 245,
unless one felt like writing the driver as well.

2006\02\10@070005 by Xiaofan Chen

face picon face
On 2/10/06, Chen Xiao Fan <EraseMExiaofanspam_OUTspamTakeThisOuTsg.pepperl-fuchs.com> wrote:

> Here are some of the links.
>
> http://www.lvr.com one of the best resource with USB
> http://www.usb.org check out the documents and the forum
> http://www.microchip.com/usb download the USB firmware
> http://forum.microchip.com check out the USB section
> http://pe.ece.olin.edu/ece/ assmebly and C18 based USB firmware
> http://pic18fusb.online.fr/wiki/ 18F USB PIC wiki
>

Here are many USB PIC related links.

http://forum.microchip.com/tm.aspx?m=123533

Regards,
Xiaofan

2006\02\10@070317 by Xiaofan Chen

face picon face
On 2/10/06, Alan B. Pearce <A.B.Pearcespamspam_OUTrl.ac.uk> wrote:
> >I eventually want to throw together a USB-to-GPIB bridge
> >for my laptop. I've seen the one that was published in EDN,
> >but I want to do away with the FTDI chip and hopefully
> >speed up the interface as well.
>
> Ditto.
>
> I have had a look at the C source (did you download that?) and pondered
> wether it would compile in C18 with minimal errors.
>
> The hardest part would then be to make the 18F4550 look like an FTDI 245,
> unless one felt like writing the driver as well.
>

The open source libusb and libusb-win32 might be of some help to solve
the driver issue. Microchip has a generic driver as well.

Regards,
Xiaofan

2006\02\10@070432 by Jan-Erik Soderholm

face picon face
Alan B. Pearce wrote :

> Ditto.
>
> I have had a look at the C source (did you download that?)
> and pondered wether it would compile in C18 with minimal
> errors.

I'm not sure what source you are refering to, but
I built the "mouse" example in C18 (student edition)
without any problem.

> The hardest part would then be to make the 18F4550 look like
> an FTDI 245, unless one felt like writing the driver as well.

For the "serial" example, Microchip supplies Windows drivers, not ?

And the mouse example uses the standard HID driver,
if I'm not wrong.

Why do you need to make the PIC look like a
FTDI chip ?

Jan-Erik.



2006\02\10@085130 by Alan B. Pearce

face picon face
{Quote hidden}

Jan-Erik
You didn't read the original bit - the bit I quoted.

The original project as written up in a Design Ideas section of EDN magazine
uses an AT90S6515 micro with an FTDI245BM to interface a GPIB bus to USB.
Hence why I said "ditto" to the bit I quoted, as I have also been looking at
doing this as well.

It would be nice to use a USB PIC in place of the FTDI245 and AT90S8515. The
source files for the AT90S8515 are available through a link given in the
article. The source is written in C.

Anyone interested should look at
http://www.edn.com/article/CA6290453.html?spacedesc=designideas where there
is an HTML version, and a PDF version available for download. The PDF
version has all the design ideas from that issue, and this is the second or
third. Use the html version for the download link for the source files
(called complete.zip). It also has the FTDI driver, and Labview interface
module.

2006\02\10@091319 by Tomas Larsson

flavicon
face
The FDTI chip normaly emulates a comport, and Mchip has a Comport driver and
FW as well.
I think however that a HID driver and FW is a better choice.


With best regards

Tomas Larsson
Sweden

Verus Amicus Est Tamquam Alter Idem

> {Original Message removed}

2006\02\10@093915 by Alan B. Pearce

face picon face
>The FDTI chip normaly emulates a comport,

While I agree this is true for the FT232 serial chips, I am not sure about
the FT245. I was assuming it looked like a parallel printer port to the OS.

>and Mchip has a Comport driver and FW as well.
>I think however that a HID driver and FW is a better choice.

Well, I was figuring on emulating the FT245, to save driver development. I
do not know (and so far haven't got the time to investigate) precisely how
the Labview driver handles the FTDI dll. It may be all much easier than I
suppose, as the original article does seem to say it is reasonably simple in
that area.

2006\02\10@100308 by Robert Young

picon face

> -----Original Message-----
> From: @spam@piclist-bouncesKILLspamspammit.edu
> [KILLspampiclist-bouncesKILLspamspammit.edu] On Behalf Of Alan B. Pearce
> Sent: Friday, February 10, 2006 8:39 AM
> To: Microcontroller discussion list - Public.
> Subject: Re: [PIC]: Getting started with USB
>
>
> >The FDTI chip normaly emulates a comport,
>
> While I agree this is true for the FT232 serial chips, I am
> not sure about the FT245. I was assuming it looked like a
> parallel printer port to the OS.

No.  There are two drivers available from FTDI, the VCP (Virtual Comm
Port) and D2XX native mode drivers.

All FTDI chips when used with the VCP drivers look like a serial port to
the PC.  It is also possible to make the device look like a serial port
using the D2XX drivers too but not nearly as simple.

Depending on the chip, the other side will be the usual serial port
lines including control lines or an 8-bit parallel port with 4 FIFO
(different depths, depending on exact chip chosen) handshaking lines.
In the case of the 8-bit parallel port I/O, data is de-serialized and
presented on the 8 I/O lines.  There is also a "bit-bang" mode available
on all chips that allows the PC to toggle lines remotely.

The maximum claimed throughput for the VCP drivers is 300kbyte/sec, for
the D2XX drivers it is 1Mbyte/sec.  In practice I have never been able
to SUSTAIN 1Mbyte/sec with the chip, something always gets bound up.
For a sustained transfer (ie hours at a time) I've gotten 880kbyte/sec.
I have seen it burst out at 1Mbyte/sec.  Some dependence on the PC and
quality of the application software.

Rob

2006\02\10@101150 by Tomas Larsson

flavicon
face
The 245 is described as a "FIFO" and the VCP driver is a VirtualComPort
driver, whilst the 232 is "UART".
Don't know the actual difference as such, but the 245 seems to be able to
handle higher speed.

Correction, there is a 245 that is paralell as well.
So they have two different 245, 245R, paralell FIFO and 245BM/BL/BQ SERIEL
FIFO

With best regards

Tomas Larsson
Sweden

Verus Amicus Est Tamquam Alter Idem

> {Original Message removed}

2006\02\10@101347 by Bob Axtell

face picon face
Tomas Larsson wrote:

>The FDTI chip normaly emulates a comport, and Mchip has a Comport driver and
>FW as well.
>I think however that a HID driver and FW is a better choice.
>
>
>With best regards
>
>Tomas Larsson
>Sweden
>
>  
>
MChip's comport driver doesn't seem to work well with the FT232BM, but
FTDI provides
their own driver that works swimmingly.

BTW, the new FTDI FT232R s an awesome new chip that eliminates an
oscillator and
is in TSSOP28. It also has a unique internal Serial Number, VERY
helpful. Again, FTDI
is supplying drivers for it.

--Bob

--
Note: To protect our network,
attachments must be sent to
RemoveMEattachTakeThisOuTspamengineer.cotse.net .
1-520-850-1673 USA/Canada
http://beam.to/azengineer

2006\02\10@101409 by Tomas Larsson

flavicon
face
Ignore parts of my previous post, The 245 IS usb-Paralell, didn't read all
the text.

With best regards

Tomas Larsson
Sweden

Verus Amicus Est Tamquam Alter Idem

> {Original Message removed}

2006\02\10@101837 by Jan-Erik Soderholm

face picon face
Alan B. Pearce wrote :

> While I agree this is true for the FT232 serial chips, I am
> not sure about the FT245. I was assuming it looked like a
> parallel printer port to the OS.

On the page you linked to they wrote :

" The controller communicates with the host computer
 through a logical serial interface that enables use of
 the controller with any programming language that
 supports serial-port communications."

So maybe one have to check what layers and comms
protocols they are using before deciding on how to make
an PIC implemention.

Jan-Erik.



2006\02\10@103005 by Tomas Larsson

flavicon
face
{Quote hidden}

Don't think that Labview is able to understad the DLL, since the calls has
to be compiled in.

As far as I understand it, the DLL is for new projects where you build the
SW to communicate with the FDTI-device and don't need Com-port emulation (it
is not neccesary), the com-port driver is there to emulate the SW which
normaly do com through either serial or LPT. Not knowing how labiew I
supposed to work, but I imagine it is supposed to connect through LPT, then
you have to emulate it with a virtual driver.


With best regards

Tomas Larsson
Sweden
http://www.naks.mine.nu for downloads etc.
ftp://ktl.mine.nu for uploads. Or use the free http://www.yousendit.com service.

Verus Amicus Est Tamquam Alter Idem




2006\02\10@103322 by Tomas Larsson

flavicon
face
{Quote hidden}

Mchips drivers are obviously designed for their own USB-Solution with A PIC
and their FW.
I am surprised if it would work with FTDI chips at all, or vice verca.

With best regards

Tomas Larsson
Sweden
http://www.naks.mine.nu for downloads etc.
ftp://ktl.mine.nu for uploads. Or use the free http://www.yousendit.com service.

Verus Amicus Est Tamquam Alter Idem
>


2006\02\10@103421 by Dave Tweed

face
flavicon
face
Alan B. Pearce <EraseMEA.B.Pearcespamrl.ac.uk> wrote:
> > The FDTI chip normaly emulates a comport,
>
> While I agree this is true for the FT232 serial chips, I am not sure about
> the FT245. I was assuming it looked like a parallel printer port to the OS.

No. A parallel port includes a standard set of handshaking lines, in
addition to the (sort of) bidirectional 8-bit data path. The FTDI chips,
whether they physically implement a parallel or a serial interface, support
only the bidirectional data stream. Therefore, even the 245 is more like a
serial port than a parallel port as far as Windows is concerned.

-- Dave Tweed

2006\02\10@103805 by Tomas Larsson

flavicon
face
{Quote hidden}

Yeah, but the most common application seems to be USB-RS232, and to emulate
a com-port.
For those that don't have a comport.


One thing though, this type of emulation drives me nuts, I have a number of
commercial stuff based on the fdti-chip, emulating com-ports, and it seems
to change com-port each time it's used, its very frustrating.


2006\02\10@110156 by Alan B. Pearce

face picon face
>> Well, I was figuring on emulating the FT245, to save driver
>> development. I do not know (and so far haven't got the time
...
>
>Don't think that Labview is able to understad the DLL,
>since the calls has to be compiled in.
>
>As far as I understand it, the DLL is for new projects where
>you build the SW to communicate with the FDTI-device and don't
>need Com-port emulation (it is not neccesary), the com-port
>driver is there to emulate the SW which normaly do com through
>either serial or LPT. Not knowing how labiew I supposed to work,
>but I imagine it is supposed to connect through LPT, then
>you have to emulate it with a virtual driver.

Umm, there is a Labview unit supplied in the zip file of this particular
USB-GPIB project. So my thinking was to make the PIC hardware as compatible
to the project as possible to minimise playing with the drivers. It sounds
like this may all be a non-problem, from what others have said about the
FTDI chip.

I must admit that I had only skim read the article, and not digested it
deeply. I think I am going to have to dig deeper and get some sample chips.

Anyone know of an alternative to the 75160/75161 GPIB Bus Driver chips? They
do seem to chew a heap of current. They also seem to be on notice as going
out of production with "last buy" notices at TI and I think I found "not
available" at National Semiconductor.

2006\02\10@110707 by Alan B. Pearce

face picon face
>One thing though, this type of emulation drives me nuts,
>I have a number of commercial stuff based on the fdti-chip,
>emulating com-ports, and it seems to change com-port each
>time it's used, its very frustrating.

That occurs when there is no unique serial number in the port. That is why
the FT232A and B devices have an external EEPROM, to contain the OEM
configuration data.

It can also be a problem if you have plugged the device in once, and then
unplug it and plug into another USB port on the host - voila yet another COM
port, different number to where it was plugged before.

The new FT232R chips have unique serial numbers in them to stop this
problem.

2006\02\10@112157 by Robert Young

picon face
{Quote hidden}

There are a couple of things that could be happening.  First, if you own
more than one FTDI USB-RS232 dongle and they have been programmed with
unique serial numbers, they will appear as different COM ports if you
swap dongles.

I have also seen the COM port shuffle problem with a single device if it
is repeatedly installed and uninstalled.  Somehow not everything is
removed from the registry and eventually Windows thinks it is a unique
device everytime it is plugged in.

Not a complete fix but you should get the latest drivers from
http://www.ftdichip.com.  Before you do this, they have a utility somewhere on
their site that can work it way through the registry and remove all
references to FTDI devices with VID=0x0403.  

Rob

2006\02\10@112524 by Philip Pemberton

face picon face
In message <01d901c62e5b$4e6e9750$spamBeGonee7bdf682spamKILLspamspace.rl.ac.uk>
         "Alan B. Pearce" <.....A.B.Pearcespam_OUTspamrl.ac.uk> wrote:

> Anyone know of an alternative to the 75160/75161 GPIB Bus Driver chips? They
> do seem to chew a heap of current. They also seem to be on notice as going
> out of production with "last buy" notices at TI and I think I found "not
> available" at National Semiconductor.

TI have the SN75160B and SN75161B listed as "Active" near as I can tell.
Problem is, nobody seems to stock them on this side of the 'Pond. Farnell
still list the National DS75160 and DS75161 as "No longer manufactured",
which suggests they might still have some stock left...

--
Phil.                         | Kitsune: Acorn RiscPC SA202 64M+6G ViewFinder
TakeThisOuTphilpem.....spamTakeThisOuTdsl.pipex.com         | Cheetah: Athlon64 3200+ A8VDeluxeV2 512M+100G
http://www.philpem.me.uk/     | Tiger: Toshiba SatPro4600 Celeron700 256M+40G
... Is this yours?  Your dog left it on my lawn ...

2006\02\10@112602 by Wouter van Ooijen

face picon face
> One thing though, this type of emulation drives me nuts, I
> have a number of
> commercial stuff based on the fdti-chip, emulating com-ports,
> and it seems
> to change com-port each time it's used, its very frustrating.

try using each gadget with its own usb port (don't plug it into
different ports).

Wouter van Ooijen

-- -------------------------------------------
Van Ooijen Technische Informatica: http://www.voti.nl
consultancy, development, PICmicro products
docent Hogeschool van Utrecht: http://www.voti.nl/hvu


2006\02\10@112646 by Jan-Erik Soderholm

face picon face
Now, I'm a *little* guessing here...

Alan B. Pearce wrote :

> Umm, there is a Labview unit supplied in the zip file of this
> particular USB-GPIB project.

That piece of code probably calls the LabView API's
on one side and talkes to a serial (physical or emulated)
serial port on the other.

> So my thinking was to make the PIC hardware
> as compatible to the project as possible to minimise
> playing with the drivers. It sounds like this may all be
> a non-problem, from what others have said about the
> FTDI chip.

Yes, since the Microchip driver is also a serial port
emulator, it *could* be PnP... :-)

Jan-Erik.



2006\02\10@113029 by Philip Pemberton

face picon face
In message <00f601c62e35$42e3d650$TakeThisOuTe7bdf682KILLspamspamspamspace.rl.ac.uk>
         "Alan B. Pearce" <.....A.B.PearcespamRemoveMErl.ac.uk> wrote:

> I have had a look at the C source (did you download that?) and pondered
> wether it would compile in C18 with minimal errors.

Actually, I did. One monolithic blob of code. Ewww. I'm not sure if I was
looking at the right block of code, but the command interpreter seemed to be
a lump of strcmp()s. Double EWW! (I'd have used an FSM).

> The hardest part would then be to make the 18F4550 look like an FTDI 245,
> unless one felt like writing the driver as well.

Why not use Microchip's "CDC" driver and appnote code? Make the PIC look like
a COM port to the PC, then just tweak the LabView driver to look for the
Microchip driver instead of the FTDI one.

--
Phil.                         | Kitsune: Acorn RiscPC SA202 64M+6G ViewFinder
RemoveMEphilpemspamspamBeGonedsl.pipex.com         | Cheetah: Athlon64 3200+ A8VDeluxeV2 512M+100G
http://www.philpem.me.uk/     | Tiger: Toshiba SatPro4600 Celeron700 256M+40G
... M-O-O-N and that spells Corn Pops.

2006\02\10@113535 by Philip Pemberton

face picon face
In message <!~!UENERkVCMDkAAQACAAAAAAAAAAAAAAAAABgAAAAAAAAAluSn7JHVyUuKseEZdXamkMKAAAAQAAAAgeyy/spamBeGone770zkW8o0D8XzUhQgEAAAAA@spam@spamspam_OUTyahoo.se>
         "Tomas Larsson" <TakeThisOuTtomaslarssonsespamspamyahoo.se> wrote:

> The FDTI chip normaly emulates a comport, and Mchip has a Comport driver and
> FW as well.
> I think however that a HID driver and FW is a better choice.

That's what I was thinking about doing.. I've heard talk about HIDs being the
best part of the USB spec to learn with - apparently they're less finicky
than stuff like COM ports...

Must do some more reading up. And pop to Farnell tomorrow to pick up some USB
connectors, voltage regulators, crystals and USB PICs. Good fun, especially
now my laptop works again. Damn you Toshiba for not soldering down the
speaker filters (well, I think they were filters - square SMD block, about
5mm square, part designation "FL702") properly! Two pence worth of solder an
two hours spent stripping the thing to bits to get to the filters...

This is going to be a fun weekend!

--
Phil.                         | Kitsune: Acorn RiscPC SA202 64M+6G ViewFinder
philpemEraseMEspamdsl.pipex.com         | Cheetah: Athlon64 3200+ A8VDeluxeV2 512M+100G
http://www.philpem.me.uk/     | Tiger: Toshiba SatPro4600 Celeron700 256M+40G
... Doing my part to preserve order in the universe

2006\02\10@114223 by Jan-Erik Soderholm

face picon face
Philip Pemberton wrote :

> TI have the SN75160B and SN75161B listed as "Active" near as
> I can tell. Problem is, nobody seems to stock them on this
> side of the 'Pond.

Both are marked "Free samples" in PDIP and "Purchase samples"
on SOIC. The PDIP in quantity 1-10 each.

And the samples service at TI is very good and fast (a couple
of days delivery).

Jan-Erik.



2006\02\10@114857 by Bob Axtell

face picon face
Tomas Larsson wrote:

{Quote hidden}

I can explain this.The USB port can't tolerate omething hanging on the
bus that does not
respond to an enquiry. This will happen if power is removed from a
USB-based product BEFORE
the USB cable is removed. After several tries, the driver removes that
channel from the list,
so that NEXT time a new channel has to be setup. That drove me crazy for
a while.

]--Bob

>
>  
>


--
Note: To protect our network,
attachments must be sent to
spamBeGoneattachEraseMEspamengineer.cotse.net .
1-520-850-1673 USA/Canada
http://beam.to/azengineer

2006\02\10@115818 by Alan B. Pearce

face picon face
>TI have the SN75160B and SN75161B listed as "Active"
>near as I can tell. Problem is, nobody seems to stock
>them on this side of the 'Pond. Farnell still list the
>National DS75160 and DS75161 as "No longer manufactured",
>which suggests they might still have some stock left...

EBV seem to be the favoured dealers in the UK for TI. I can buy tube lots
(25 off) of each device from them for what look like reasonable prices.

But at TI, when you dig down through the layers they are actually talking
about "last buy" orders for both of them, even though they are "active".

2006\02\10@115923 by Alan B. Pearce

face picon face
>Yes, since the Microchip driver is also a serial port
>emulator, it *could* be PnP... :-)

Thats what I am hoping.

2006\02\10@120351 by Alan B. Pearce

face picon face
>> I have had a look at the C source (did you download that?)
>> and pondered wether it would compile in C18 with minimal errors.
>
>Actually, I did. One monolithic blob of code. Ewww. I'm not
>sure if I was looking at the right block of code, but the
>command interpreter seemed to be a lump of strcmp()s.
>Double EWW! (I'd have used an FSM).

<VBG> I have only scanned it by opening in an editor. I couldn't easily see
where the entry point was, so appreciate your reaction.

>> The hardest part would then be to make the 18F4550 look
>> like an FTDI 245, unless one felt like writing the driver as well.
>
>Why not use Microchip's "CDC" driver and appnote code?
>Make the PIC look like a COM port to the PC, then just
>tweak the LabView driver to look for the Microchip driver
>instead of the FTDI one.

Yes that is looking more what will happen, as people have explained the
innards of USB in this thread. 'Twill be my first foray into USB (apart from
using some FTDI chips in Wouters kits).

2006\02\10@120543 by Alan B. Pearce

face picon face
>And the samples service at TI is very good and fast
>(a couple of days delivery).

OK, I might go that route instead of buying a tube from EBV. Just means
registering on yet another site that will send at least one email a week on
anything you don't want to know.

2006\02\10@124646 by Philip Pemberton

face picon face
In message <002701c62e63$2f594f60$e7bdf682spamBeGonespamspace.rl.ac.uk>
         "Alan B. Pearce" <RemoveMEA.B.Pearce@spam@spamspamBeGonerl.ac.uk> wrote:

> EBV seem to be the favoured dealers in the UK for TI. I can buy tube lots
> (25 off) of each device from them for what look like reasonable prices.

Well, if you want to split a tube, I might be interested in buying a few
(depending on price).
I've never had much luck dealing with distributors TBH. The conversation
usually ended with words to the effect of "Come back when you want 1,000 of
them". Last time I made the mistake of mentioning I was a student - the
response was (to paraphrase) "Go away, we're not going to be interested in
orders from the likes of you".

Their loss I suppose.

> But at TI, when you dig down through the layers they are actually talking
> about "last buy" orders for both of them, even though they are "active".

"Get 'em while they're hot!"

Thanks,
--
Phil.                         | Kitsune: Acorn RiscPC SA202 64M+6G ViewFinder
.....philpem@spam@spamEraseMEdsl.pipex.com         | Cheetah: Athlon64 3200+ A8VDeluxeV2 512M+100G
http://www.philpem.me.uk/     | Tiger: Toshiba SatPro4600 Celeron700 256M+40G
... Dumb luck beats sound planning every time.  Trust me.

2006\02\11@041615 by Jinx

face picon face
> I've just ordered a pair of PIC18F4550s and I've realised that
> I don't know the first thing about interfacing with the USB bus

Well, gotta get marks for spelling it right

I've a 452 product that needs USB, and rather than go for the
FTDI solution, the 4550 seems the likely replacement stock. I've
not had a good look into USB yet, but found this link at Dontronics
to Jan Axelson's "USB Central"

http://www.lvr.com/usb.htm

2006\02\11@162147 by Jinx

face picon face
en.wikipedia.org/wiki/Usb

Not PIC specific, plenty of background and industry links


2006\02\11@180217 by Philip Pemberton

face picon face
In message <.....7D4AB72251D4D949AB2732ABEABDA54F130054RemoveMEspamPFSG-MX1.ap.p-f.biz>>          "Chen Xiao Fan" <.....xiaofanSTOPspamspam@spam@sg.pepperl-fuchs.com> wrote:

> http://www.usb.org check out the documents and the forum

Eeek.. That's one long specification document...

I've been scanning through <http://www.beyondlogic.org/usbnutshell/> - "USB
in a Nutshell: Making Sense of the USB Standard".
I've also been playing around with SnoopyPro (a USB bus-sniffer) - available
for free from <http://usbsnoop.sourceforge.net>. It's neat seeing what
happens when a device appears, right down to packet level.

> http://www.microchip.com/usb download the USB firmware

Got them - thanks. I'm looking at the PICDEM FS-USB demoboard code - I've
flashed a chip with the bootloader, but the demo code keeps dropping off the
USB and then reconnecting. I suspect that's got something to do with the fact
I've torn the FS-USB down to the bare minimum of hardware necessary to get
USB comms going. In other words, an 18F4550, a 20MHz crystal, a B-type USB
connector, two switches, PURs and some decoupling caps.

> http://forum.microchip.com check out the USB section

Now THAT is useful. Or at least it would be if they didn't delete attachments
on what appears to be a weekly cycle...

> http://pe.ece.olin.edu/ece/ assmebly and C18 based USB firmware
> http://pic18fusb.online.fr/wiki/ 18F USB PIC wiki

That last one is a fantastic "get going FAST" type site - my current "USB
devboard" is a breadboarded version of the stripped-down PICDEM FS-USB on the
18FUSB Wiki.

OK, next steps:
1)  Pick apart the 18F USB bootloader, learn how it works, change stuff, see
    what happens, try not to confuse the Win2k USB stack (too) badly :)
2a) Wire up an LED, write a simple "X sent over USB turns LED on / Y sent
    over USB turns LED off" firmware app and use LibUSB to test it.
2b) Make the LED example use interrupts instead of polling
3)  Try and get some form of primitive GPIB stack up and running
4)  Buy a block of VIDs/PIDs from VOTI
5)  Debug, tweak and perfect the GPIB stack
6)  Release it on an unsuspecting world :P
   
Thanks.
--
Phil.                         | Kitsune: Acorn RiscPC SA202 64M+6G ViewFinder
philpemEraseMEspam@spam@dsl.pipex.com         | Cheetah: Athlon64 3200+ A8VDeluxeV2 512M+100G
http://www.philpem.me.uk/     | Tiger: Toshiba SatPro4600 Celeron700 256M+40G
... Pets just die on you, where's the fun in that?

More... (looser matching)
- Last day of these posts
- In 2006 , 2007 only
- Today
- New search...