On Fri, 13 Jun 2008 03:53:32 -0400, CBFalconer <c...@yahoo.com>
wrote:

>CBFalconer wrote:
>> Tomás Ó hÉilidhe wrote:
>>
>> ... snip ...
>>
>>> Can anyone suggest a USB to RS232 adapter that does its job
>>> perfectly 100% of the time under Microsoft Windows, i.e. it
>>> trully behaves exactly just like any other serial port?
>> 
>> Impossible.  USB is not a serial port.
>
>I'm hearing some objections to my statement above.  My point is
>that USB is a shared system, and that it queues up traffic for
>transmission at intervals.  You can't use the adaptor to respond to
>an input event in microseconds, as you can with the original port. 
>What you can do is asynchronous transmission and reception, which
>is most people's objective.
>
>For example I believe that the original X-modem protocol will fail
>miserably.  That requires responding to a transmission with an ACK
>(or NAK) within a very short time.  Z-modem will probably work.

I also very much doubt that you could use it with an external
RS-232/485 converter with data direction control using the RTS line.
Doing it properly is very hard even on any 14550 style UART,
especially on any multitasking OS. 

Also implementing Modbus RTU with proper timing is going to be a
challenge :-).

Paul
 

Paul Keinanen wrote:
> On Fri, 13 Jun 2008 03:53:32 -0400, CBFalconer <c...@yahoo.com>
> wrote:
> 
>> CBFalconer wrote:
>>> Tomás Ó hÉilidhe wrote:
>>>
>>> ... snip ...
>>>
>>>> Can anyone suggest a USB to RS232 adapter that does its job
>>>> perfectly 100% of the time under Microsoft Windows, i.e. it
>>>> trully behaves exactly just like any other serial port?
>>> Impossible.  USB is not a serial port.
>> I'm hearing some objections to my statement above.  My point is
>> that USB is a shared system, and that it queues up traffic for
>> transmission at intervals.  You can't use the adaptor to respond to
>> an input event in microseconds, as you can with the original port. 
>> What you can do is asynchronous transmission and reception, which
>> is most people's objective.
>>
>> For example I believe that the original X-modem protocol will fail
>> miserably.  That requires responding to a transmission with an ACK
>> (or NAK) within a very short time.  Z-modem will probably work.
> 
> I also very much doubt that you could use it with an external
> RS-232/485 converter with data direction control using the RTS line.
> Doing it properly is very hard even on any 14550 style UART,
> especially on any multitasking OS. 
> 

You can do RTS control of RS-485 drivers, but you have even less 
accurate timing with USB than with a standard UART.

An alternative is to use a feature of the FTDI chips - they can drive a 
RS-485 direction line directly whenever they are transmitting.

> Also implementing Modbus RTU with proper timing is going to be a
> challenge :-).
> 
> Paul
>  

In article <9bd63d75-3975-47d3-810a-361318b93f1f@
34g2000hsf.googlegroups.com>, t...@lavabit.com says...
> 
> I've purchased one already made by "Cypress Semiconductor" but it
> doesn't work. The VID is 04b4 and the PID is 5500.
> 
> Can anyone suggest a USB to RS232 adapter that does its job perfectly
> 100% of the time under Microsoft Windows, i.e. it trully behaves
> exactly just like any other serial port?
> 
Keeping in mind the other posts in this thread AND that your mileage may 
vary; we use, and have never had a SINGLE problem with, the ATEN 
adapters.  It may just be luck, or I just haven't tried the magic combo 
that blows it up, but those units have been rock solid performers for 
us.  I don't know what chipset the units use and really don't care, 
because they work.  They are a bit pricier than most of the crap out 
there.

Just google or search on amazon:

ATEN UC-232A 

and you will get plenty of info.

I need an RS232 port for the following two reasons:
    * For use with the GT ROM program for re-flashing firmware
    * For programming PIC chips

The device that Don suggested looks very attractive but I'd just like
ask one more question:
    My laptop has an "express card" slot. Would I be better off
getting an RS232 adapter that goes into the express card slot, or
should I go with the EasySync USB adapter?

"Tomás Ó hÉilidhe" <t...@lavabit.com> wrote in message 
news:9...@34g2000hsf.googlegroups.com...
>
> I've purchased one already made by "Cypress Semiconductor" but it
> doesn't work. The VID is 04b4 and the PID is 5500.
>
> Can anyone suggest a USB to RS232 adapter that does its job perfectly
> 100% of the time under Microsoft Windows, i.e. it trully behaves
> exactly just like any other serial port?

I have had the best results with Keyspan.  The p/n on the one I use says 
USA-19HS.  It is the most reliable that I have used.  It has a TI USB3410 
inside with a MAX3243 that I believe is just the RS232 level translator.

Scott

Tomás Ó hÉilidhe wrote:
> I need an RS232 port for the following two reasons:
>     * For use with the GT ROM program for re-flashing firmware
>     * For programming PIC chips
> 
> The device that Don suggested looks very attractive but I'd just like
> ask one more question:
>     My laptop has an "express card" slot. Would I be better off
> getting an RS232 adapter that goes into the express card slot, or
> should I go with the EasySync USB adapter?


I just did a google, as I wasn't sure if they were readily available, 
but found plenty, pricey however. $90USD for this one:
http://www.usbgear.com/computer_cable_details.cfm?sku=SS-AS1RS&cats=496&catid=1302%2C496%2C585%2C538%2C464%2C468


And I wouldn't know what sort of chip set these or others use. Some 
people may be aware.

Now have a look at mine:
http://www.dontronics-shop.com/easysync-premium-gold-usb-rs232-adapter-cable-10cm-cable.html

Price includes world wide postage. It won't be the cheapest, but it 
works when most of the others fail. That is why I handle them. You can 
speak to the guys that designed the chipset, if you need in depth support.

Re-read the guarantee of the ability to return the goods if it doesn't 
work for your application. Our loss on postage, all you pay is return 
postage.

And re-read the customers feed back.
I think there are 18 responses from Feb 2002, to Dec 2006. I simply 
stopped adding them. Applications that range from Garmin GPS's to Pfaff 
sewing machines. I doubt if anywhere else on the web, you will find a 
report such as this in the way of genuine feed back, on a USB to RS-232 
converter. Please correct me if I am wrong.

Cheers Don...



-- 
Don McKenzie

Site Map:            http://www.dontronics.com/sitemap
E-Mail Contact Page: http://www.dontronics.com/email

Intelligent 2.83" AMOLED with touch screen for micros:
http://www.dontronics-shop.com/product.php?productid=16699

On 2008-06-13, Glenn M?ller-Holst <n...@xx.dk> wrote:
> CBFalconer wrote:
>> Tom?s ? h?ilidhe wrote:
>> ... snip ...
>>> Can anyone suggest a USB to RS232 adapter that does its job
>>> perfectly 100% of the time under Microsoft Windows, i.e. it
>>> trully behaves exactly just like any other serial port?
>> 
>> Impossible.  USB is not a serial port.

Actually it is if one is being pedanting, and this is Usenet so
the pedantic flag is set by default. ;)

> He properly means that he wants to have the RS232 tunnelled through USB 
> and showing itselves as a serial port in Windows.

None of the USB->serial adapters (FTDI, Prolific, or any other)
will behave "exactly just line any other serial port".  The
timings and latencies are going to be _way_ different.

-- 
Grant Edwards                   grante             Yow! Bo Derek ruined
                                  at               my life!
                               visi.com            

On 2008-06-13, Paul Keinanen <k...@sci.fi> wrote:

> I also very much doubt that you could use it with an external
> RS-232/485 converter with data direction control using the RTS
> line.

Unless the UART handles that itself (decent ones do).

> Doing it properly is very hard even on any 14550 style UART,
> especially on any multitasking OS. 
>
> Also implementing Modbus RTU with proper timing is going to be a
> challenge :-).

Ah, I think you misspelled "impossible".

-- 
Grant Edwards                   grante             Yow! I would like to
                                  at               urinate in an OVULAR,
                               visi.com            porcelain pool --

In article <4...@yahoo.com>, CBFalconer <c...@yahoo.com> writes:
> CBFalconer wrote:
>> 
>> Impossible.  USB is not a serial port.
> 
> I'm hearing some objections to my statement above.  My point is
> that USB is a shared system, and that it queues up traffic for
> transmission at intervals.  You can't use the adaptor to respond to
> an input event in microseconds, as you can with the original port. 
> What you can do is asynchronous transmission and reception, which
> is most people's objective.
> 

I'll give you another (and probably more relevant for this newsgroup)
example: the low cost serial port programmers that use the modem and
hardware flow control signals to drive a microcontroller's onboard
programming capabilities. See (for example) the serial port based AVR
programmers.

USB latency means that there's no way you are going to be able to run
those over a USB to serial adapter, even if the adapter does a 100%
accurate implementation of the modem and flow control signals.

Simon.

-- 
Simon Clubley, clubley@remove_me.eisner.decus.org-Earth.UFP
Microsoft: Bringing you 1980's technology to a 21st century world

In article <W...@posted.visi>, 
g...@visi.com says...
> On 2008-06-13, Glenn M?ller-Holst <n...@xx.dk> wrote:
> > CBFalconer wrote:
> >> Tom?s ? h?ilidhe wrote:
> >> ... snip ...
> >>> Can anyone suggest a USB to RS232 adapter that does its job
> >>> perfectly 100% of the time under Microsoft Windows, i.e. it
> >>> trully behaves exactly just like any other serial port?
> >> 
> >> Impossible.  USB is not a serial port.
> 
> Actually it is if one is being pedanting, and this is Usenet so
> the pedantic flag is set by default. ;)
How true.
'U'niverial 'S'erial 'B'us
Yep, definitely a serial port. :)

> 
> > He properly means that he wants to have the RS232 tunnelled through USB 
> > and showing itselves as a serial port in Windows.
> 
> None of the USB->serial adapters (FTDI, Prolific, or any other)
> will behave "exactly just line any other serial port".  The
> timings and latencies are going to be _way_ different.

The only BIG problem I have ever had was the fact that most of the 
adapters we tested had some strange glitches occurring on the outputs at 
very consistent intervals.  Like 50KHz and ~100KHz on two I recall.  I 
presume this was some artifact introduced by the chipset and the polling 
rate of the USB system/driver.  In some instances these spikes were 
interpreted as start bits and things didn't go well after that.

                               Jim