A aviation & planes forum. AviationBanter

If this is your first visit, be sure to check out the FAQ by clicking the link above. You may have to register before you can post: click the register link above to proceed. To start viewing messages, select the forum that you want to visit from the selection below.

Go Back   Home » AviationBanter forum » rec.aviation newsgroups » Military Aviation
Site Map Home Register Authors List Search Today's Posts Mark Forums Read Web Partners

Question about the F-22 and it's radar.



 
 
Thread Tools Display Modes
  #1  
Old April 3rd 04, 05:23 AM
Paul F Austin
external usenet poster
 
Posts: n/a
Default


"Scott Ferrin" wrote
"Tarver Engineering" wrote:


"Harry Andreas" wrote
John Cook wrote:


Just the official reports!!, Lockheed has only purchased enough
processors for 155 F-22's because there out of production, the demand
for Air to ground operations has increased the demand on processing
power, something the original processors are not quite upto hence the
_need_ for the 'upgrade'.

So the processors are obsolete, (too old)... the Avionic architecture
needs to be replaced before the F-22 can become the F/A-22 because

the
present system is based on the old processors and rewriting the code
is pointless on an obsolete system, that would only support half of
the F-22 fleet

Methinks there's some confusion there between processors, avionics
architecture, and software.
While it's true that Intel tried to shut down i960 production causing a
chinese fire drill, there are enough assets to get by until a new

processor is
ready.


Intel has agreed to provide mil-spec i960s, thanks to a very fat check

from
USAF. The new processor has already failed to be integrated, due to a

loss
of tracability. (ie scrap)

That has nothing to do with the avionics architecture, which is
not changing. Plus the whole point of writing all the OS and AS in Ada

was
to be as platform independent as possible, so that upgrades to the CIP
could be relatively painless and not force re-flight testing of the

A/C.
Ideally, one would not re-write the code, but re-compile the code for
the new platform, then do a LOT of integrity checks, and take it from

there...

They have to go with a more COTS based system (similar to, if not the
same as the JSF), which they are working on now, for fielding in

(very
optomisticlly) in 2007.

Other than using commercialy available processor chips, what is "COTS"
about it?
Hint - nothing.


Wrong. Name for us the one and only modern processor that is mil-spec,
Harry.



I don't know if it's milspec but ISTR reading that Intel donated the
Pentium 1 design to the US military to do with as it pleased. I also
remember reading an article on some Russian naval electronics in which
the advertiser was boasting that they were "Pentium" powered.


Intel did donate the rights to Pentium 1 to the USG and Sandia has been
working on producing a rad-hard flavor but it seems to have been overtaken
by events.

In general, MIPS and PowerPC processors are favored over Intel parts because
of power vs performance metrics. Intel parts deliver lots of performance but
put a heavy burden on the designers trying to get the heat out.

The main issues in using commercial parts in military applications are 1.
Quality level, 2. Operating temperature range and 3. Packaging.

As far as quality level is concerned, there are several MIPS and PowerPC
CPUs available screened to -883B and also to class S (space grade). Both IBM
and Moto PPC603Es and -750s of various flavors are available screened to MIL
standards. You have to buy upscreened parts (by second parties) but that's
the way it's done. Aeroflex sells a 600MIP MIPS processor that's also
available compliant with MIL standards. .A secondary problem is support
chips. That's most often done with IP hosted on FPGAs.

As for temperature, all the high performance CPUs operate over a restricted
temperature range smaller than the mil -55 to 125C. Instead, you have to
work within industrial temp range (-40 to 105C) but that just makes life
hard for the thermal designers.

Packaging can be tough. There are a few sources for hermetic, flat-pack high
performance CPUs (Aeroflex is one). Mostly though, we've had to learn to use
ball grid array parts, some of which are ceramic and others plastic.
Depending on the application, the plastic ones are used as is or repackaged
(which is expensive and risky). Either way, BGAs present major challenges in
avionics applications because of temperature cycling induced ball failures.
Each vendor is working to develop processes that will survive but right now,
it's a black art.



  #2  
Old April 3rd 04, 06:43 AM
Tarver Engineering
external usenet poster
 
Posts: n/a
Default


"Paul F Austin" wrote in message
...


In general, MIPS and PowerPC processors are favored over Intel parts

because
of power vs performance metrics. Intel parts deliver lots of performance

but
put a heavy burden on the designers trying to get the heat out.


Intel went out of the Mil-Spec processor business and Motorola kept making
them. The Mil-Spec components specifications were abandoned in place in
2000 and Intel had no incintive to continue to support a fantasy world.

The main issues in using commercial parts in military applications are 1.
Quality level, 2. Operating temperature range and 3. Packaging.


AKA the Rome data, as based on the RPL Model. RL has a pretty nice software
reliabilty model as well, but of course the F-22 was to early for COTS. I
am optimistic about the F-35, with it's injection of the RPL model.


 




Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

vB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Forum Jump


All times are GMT +1. The time now is 12:05 AM.


Powered by vBulletin® Version 3.6.4
Copyright ©2000 - 2025, Jelsoft Enterprises Ltd.
Copyright ©2004-2025 AviationBanter.
The comments are property of their posters.