ts-7000
[Top] [All Lists]

Re: [ts-7000] Re: Iridium Satellite modem + TS7260 timeouts on transmit

To: dazz_taz <>
Subject: Re: [ts-7000] Re: Iridium Satellite modem + TS7260 timeouts on transmit
From: Todd Valentic <>
Date: Thu, 08 Oct 2009 09:29:00 -0700
Hey Darren -

> Thanks for the information, I have seen the exact behaviour your have 
> described with the CTS control line.
 > Now that you describe it I can now see what is happening.  So am I to 
 > understand it that you dont 
use the
 > /dev/ttyAM ports at all and use a usb to serial adapter on port /dev/ttyUSB0?
 >
> I have done this and I have also set the terminal software to 19200 8N2.
 > I was not able to get 19200 8N2 to work with the /dev/ttyAM0 port, the modem 
 > would not
 > respond to commands with readable characters.

Ah - you're right about that one. I had forgotten about the AM* issues and two 
stop bits.
On my current systems, I roll my own 2.6 kernel and OS. The 2.4 kernel that TS 
ships with
the board has an issue with two stop bits on the internal AMBA based ports. Our 
systems
had a TS-SER1 board that we were using and that handles the two-stop bits just 
fine. The
new 2.6 kernels also work great. I don't think I tried COM3 with the default 
2.4 kernel.
That serial port is implemented in their FPGA and thus has a completely 
different driver.

Like yourself, I was in a bind with a set of experiments that were getting set 
for
deployment and ran into a wall with the lock up issue. Thinking it was a flow 
control
problem, we added a SER1 board that has a real 16550 UART (not emulated) and 
can do
flow control on chip. That still failed. On a lark, we tried some USB 
converters and
they worked. We have a combination of pl2303 and ftdi based devices. All of 
them worked
and solved the problem. I did have to backport (make that just copy over and 
compile)
the latest 2.4 versions of those drivers, but that wasn't too bad of a problem. 
I can
send you the binaries if you want/need them. The systems that we ended up 
shipping to
the field then all used the USB solution. I didn't like it, but we were stuck. 
I did
manage to find a set of cold temperature rated USB-to-serial adapters which 
made things
a bit more palatable.

The two-stop bit solution was discovered after those system were deployed. We 
also
saw lockup failures, although to a much less extent, with the TS7800 as well as 
some
486-based PC-104 boards. It is not limited to the 7260 by any means and is a 
problem
in the transceiver.

Some of my current projects require UVC based webcams onboard, which need the 
latest
2.6 kernels. I have a nice buildroot-based system for compiling everything, 
from the
cross-compilers, the kernel and on through the root file system. The new kernel 
fixes
the /dev/ttyAM0 problem with the two-stop bits and that is the serial port we 
are using
for the Iridium comms at this point.

> The USB port has yielded better results in that the connection does not lock 
> up.
 > However I seem to get a lot of CRC and timeouts which reduces the overall 
 > performance
 > to about 130cps rather than the maxmimum 300cps.   But at least now it does 
 > not just lock up and 
stop.

What data services are you using from Iridium? Is this a modem-to-modem dial up?
DirectIP? RUDICS?

Are you using commercial SIMs or the US DoD gateway?

Where is your antenna located? Does it have a clear view of the sky, 
horizon-to-horizon? If
part of it is blocked, then you'll be in a shadow as the satellites move across 
the sky. It
only takes about 10-minutes for them to pass overhead. We've seen lots of 
incidents where
you get drop outs due to the shadowing. Having said that, call drops are part 
of the nature
with Iridium. On average, in the polar regions, you should see calls nowadays 
last for about
20-30 minutes. That's up from what it was a few years ago.

What you shouldn't be getting are any errors. Once you've got a connection, 
bytes going into
the system are almost guaranteed to come out on the other side.

With our RUDICS connections, we can stream data at the full rate (about 
2500bps).

> It is not an aerial/reception problem as transmission works fine on a Windows 
> to Windows setup.
 > So it still must be something to do with teh TS-7260 serial interface.

What else is happening on the 7260? Writes to USB or SD? Our systems run off the
internal flash and consist of a lot of Python-based code. They are quite busy
with a number of different tasks, but we don't miss characters.

> Which model Iridiums are you using, we have a couple of 9255As but the rest 
> are 9255B series :
 > A3LA-XM.  I seem to get varing results depending on which modem I use.

You name it, we've got it. Our research group supports the communications
aspect of the NSF polar program in the Arctic. We have a couple hundred
handsets and transceivers.

The older A3LA-D and -I models were great for cold temperature operations.
They would run down to -50C. The newer -SA models only go to -30C, but can
vary widely in that regard. We are currently testing the newer -X models.
I actually was doing that yesterday and can confirm that they are working
for our setup. We've done cold temperature testing and the ones that we
have are working at -30C, but not any lower. If you can, you definitely
need to temperature test before deployment.

The -X models are new and the various groups that I know of are just
starting to get them working. One big change has been in the amount
of power they use - much more than the SA, which hurts in these low
power setups.

> Any thoughts, have I missed something obvious?

One other place to look at is the power supply for the modem itself.
We've seen problems where you have enough current to power the modem
and talk to it, but not enough to transmit (or transmit sporadically).
Since your Windows boxes work, that probably isn't the problem unless
you are powering them differently at that point.

If you don't mind, I'm going to cc the list with this response so
people searching in the future can find it.

Regards,
Todd

-- 
_______________________________________________________________________

Dr. Todd Valentic                                     SRI International
Senior Research Engineer                             333 Ravenswood Ave
Center for GeoSpace Studies                        Menlo Park, CA 94025

E-Mail :                    Office: (650) 859-4642
Website: isr.sri.com                             Fax   : (650) 322-2318
_______________________________________________________________________


------------------------------------

Yahoo! Groups Links

<*> To visit your group on the web, go to:
    http://groups.yahoo.com/group/ts-7000/

<*> Your email settings:
    Individual Email | Traditional

<*> To change settings online go to:
    http://groups.yahoo.com/group/ts-7000/join
    (Yahoo! ID required)

<*> To change settings via email:
     
    

<*> To unsubscribe from this group, send an email to:
    

<*> Your use of Yahoo! Groups is subject to:
    http://docs.yahoo.com/info/terms/

<Prev in Thread] Current Thread [Next in Thread>
  • Re: [ts-7000] Re: Iridium Satellite modem + TS7260 timeouts on transmit, Todd Valentic <=
Admin

Disclaimer: Neither Andrew Taylor nor the University of NSW School of Computer and Engineering take any responsibility for the contents of this archive. It is purely a compilation of material sent by many people to the birding-aus mailing list. It has not been checked for accuracy nor its content verified in any way. If you wish to get material removed from the archive or have other queries about the archive e-mail Andrew Taylor at this address: andrewt@cse.unsw.EDU.AU