The Eagle design criteria doesn't permit single-point failures in the analog-mode transponder. Hardware is designed to eliminate that possibility. If the idea is that the signal-processing software has 100,000 instructions in it so each could fail, it doesn't work that way. Software is tested to elimiate potential failure paths. A software-defined transponder has already been demonstrated at the Dayton Hamvention with real users.
73,
John KD6OZH
----- Original Message ----- From: G0MRF@aol.com To: bill@hsmicrowave.com; brobertson@mta.ca; amsat-bb@amsat.org Sent: Thursday, September 14, 2006 23:59 UTC Subject: [amsat-bb] Re: The Mode B tradition
In a message dated 9/14/2006 10:39:26 PM GMT Standard Time, bill@hsmicrowave.com writes:
Hi David,
Help me out with the 100,000 single point failures. Are you referring to the complexity of digital gates etc., in the DSP.
Isn't P3E going with a digital IF or are they using an analog backup or vice versa? That's the SDX thingy isn't it?
Regards...Bill - N6GHz
Hello Bruce / Dom.
I agree, the UV on Eagle will be apparently analog (ue) to the casual user.
But be aware Bruce that it's not "primarily" a software driven transponder.
It is in fact TOTALLY software driven. There is no linear IF, just DSP in the core. Or from another viewpoint 100,000 single point failures just waiting to happen.
73
David G0MRF
Hello Bill.
As I understand it, and perhaps I'm being a little over conservative in my concern, The Eagle design has shifted with respect to the caution of engineering that was shown just 14 months ago. My reference to 100,000 single point failures just waiting to happen was aimed at the UV SDX transponders total reliance on some very impressive but completely untried in space digital electronics. From the various messages on the BB in the past 2 weeks I understand that the line up of the transponder is: U band receiver to 10.7MHz - DSP based core processing the signal at baseband - Then an upconversion process that goes to 10.7MHz then 145MHz. This would appear to place the DSP it's software and its control electronics within a critical path which has large potential for a single failure. For example if one of the junctions in the DSP is 'hit', then the transponder stops working. - There is no plan B. OK, Yes it will be radiation tested, but as I pointed out in a previous mail, it will be launched at the peak of the sunspot cycle and orbital manouvers can go wrong leaving the satellite in the wrong part of the Van Allen belts. Let's look at our own history: AO-10 was placed in a high radiation environment and within a short period the IHU died. However, the RF components in the transponder continued to function and the satellite provided a limited service. Why did it continue to function? Well, I can only guess. But how about the possibility that RF devices with large junctions are less vunerable to radiation than sub 1 micron devices in LSI logic. In AO-7, the RF components are still working 20 years on. I have suggested that a simple signal path be designed around the SDX core. Signals appear to exist at 10.7MHz and adding just a few extra components would allow a malfunctioning SDX to be bypassed. What is really curious is that these fears have been voiced before and reassurances were given by Rick W2GPS. However for some reason, what seems to be a prudent, inexpensive and simple option has been dropped in favour of increased trust in radiation testing or by providing an equally vunerable duplicate DSP unit.
Please see original question asked by Drew KO4MA with reply from W2GPS dated July 2005.
==============================================================
Drew,
See the Eagle specification document at _http://www.amsat.org/amsat-new/eagle/Eagle_Fall_2004.php_ (http://www.amsat.org/amsat-new/eagle/Eagle_Fall_2004.php)
It says "To provide a safe backup system in case of a failure in the digital implementation of the linear transponders there will be two S-Band transmitters and either of them will be capable of being driven by the SDR driver or by a traditional analog linear driver. One or more of the receivers will also be capable of analog operation."
Rick W2GPS AMSAT LM 2232
-----Original Message----- From: owner-Eagle@AMSAT.Org [_mailto:owner-Eagle@AMSAT.Org_ (mailto:owner-Eagle@AMSAT.Org) ] On Behalf Of Andrew Glasbrenner Sent: Sunday, July 24, 2005 7:45 PM To: Robert McGwier; Amsat-Bb Cc: Jim Sanford Subject: [eagle] Re: [amsat-bb] LONG and long overdue report
Hi Bob,
Thank you very much for the update on the Eagle transponder progress. The SDR based transponders sound very cutting edge and extremely adaptible to AMSAT's needs. I very much look forward to hearing them at the symposium in a few months. I'll have my satellite backpack station that was used for the Dayton demos, and will be glad to offer it up for demonstrating the new transponders. CC Rider is also very exciting and sure to open new realms of small stations and portable operation.
As someone primarily looking in from a user/member point of view I would like to offer two comments. First, I'd hate to think we are going to hang our entire mission on just the SDR transponders. Wouldn't it be prudent to include at least a bare bones analog backup transponder? Please tell me this is the plan. Second, I'd like to encourage that software development not lag too far behind the hardware. I'm not a big fan of the idea of writing the code once it's in orbit. One does not have to look too far to find where AMSAT has oversold or overpromised features based on writing the software after launch. Please take these comments in the spirit with which they were offered, with only my desire to see Eagle as succesful as possible.
73 and thanks for your efforts, Drew KO4MA AMSAT #33438
Sent via AMSAT-BB@amsat.org. Opinions expressed are those of the author. Not an AMSAT-NA member? Join now to support the amateur satellite program! Subscription settings: http://amsat.org/mailman/listinfo/amsat-bb