McDonnell-Douglas F-15 Avionics, AN/APG-63/70 radar

Dilbert said:
I think that the animation could still appear the same, regardless whether the antenna is roll-stabilized or not. The antenna symbol in the HUD shows only its angular aiming direction, and the roll angle of the aircraft - not the antenna's own roll angle, independent of the aircraft roll angle.
if the antenna isn't roll-stabilazed, the "antenna direction symbol" moves as Lock On



You're right, my mistake - so now mrdetonator should check to see if MiG-29 can hold lock on targets, while flying upside-down for more than 12 seconds? :)
I think maximum 5-6 sec.

There's another problem when flying inverted for extended time - I think that the accumulator fuel tank in modern jets only works upside-down for some seconds, no?
I don't know

There are some other differences. If I understand correctly, in Russian aircraft the guard horn significantly decreases the maximum detection range, so there is a switch in the cockpit to use or not use it, as necessary. I never heard of such a switch in the Western cockpit.
A switch is used when you have a lot of jamming signals and not on all aircrafts ( Mig's ....SU-27 doesn't have it at all) And we don't know whether exectly this guard horn is disable.
 
Dilbert said:
Rather, I have difficulty to make the conclusion that this animation demonstrates antenna roll-stabilization during a lock. I think that the animation could still appear the same, regardless whether the antenna is roll-stabilized or not. The antenna symbol in the HUD shows only its angular aiming direction, and the roll angle of the aircraft - not the antenna's own roll angle, independent of the aircraft roll angle.
I`m thinking the same thing. I don`t see a reason why the antenna should be roll-stabilized during target track(lock-on). Actually I mean if roll-stabilized, it might run into problems earlier when aircraft rolled more than 120deg but less than 240deg, because the antenna can`t roll back to reacquire the target again.
Well,well, we will see....you said flying upside-down not more than 6sec ;D
 
OT
unfortunately roll-stabilized in STT it is not my theory-it is a fact.
6 sec. It is limitation of comp. softwear version for extrapolation target track :)
 
Just found this article from Kopp with some avionics details, though not much has been released:
The F-15A was unique in having several firsts in the area of avionics. The AN/APG-63 radar was one of these. Designed for use in a single seat aircraft this radar had to perform both as an air superiority sensor and air defence system, both missions having specific requirements. For the air defence role the radar employs its all aspect look-down/shoot-down capability, 130 nm+ range and track-while-scan (8 targets), range-while-search, AIM-7 Sparrow illumination modes and it couples to an IFF interrogator.

For its period the radar had excellent ECCM ability, though this was necessary as the synthetic computer generated symbology displayed on the head-down ANMI scope conceals a lot of the observable effects of jamming.

Targets are adaptively tracked in ground clutter with Kalman tracking filters and the APG-63 has been known to lock onto fast moving road vehicles, outside of legitimate bogies. Early models were sensitive to Jet-Engine-Modulation (JEM) of radar returns, which affected the tracking filters when a target was illuminated up its tailpipe.

As an air superiority radar the APG-63 represented a quantum leap with its semi-automatic dogfight modes. In Boresight mode the pilot flies the F-15's longitudinal axis onto the target and the radar locks on, in Supersearch the pilot manoeuvres the target into the 20 degree HUD field of view, which is scanned by the radar, after lock-on the radar places a square box over the target to facilitate visual acquisition. The radar supports the gun and Sparrow/Sidewinder launch in these modes, all of which are selectable by throttle and stick buttons.

The APG-63 is fully modular, uses an X-band planar array antenna and a gridded TWT transmitter, the whole system weighing in at 494 lb. The aircraft has a single IBM built 32-bit 'Central Computer', which controls the weapon system, displays and radar; for its period its 3.4x10^5 instr/sec / 16 kword RAM capability was considerable. The F-15 also had the first self contained internal electronic warfare system in an air-air fighter. The core element is the large Loral ALR-56 Radar Homing And Warning System (ESM), supported by the ALQ-128 launch warning system, ALQ-154/155 tail warning system and AAR-38 infra-red rear warning system. These were complemented by the ALQ-135 internal jammer, all EW data is displayed on the head-down Tactical EW System (TEWS) scope. Precision navigation reference is provided by the AN/ASN-109 inertial system, with 1.5 nm/hr drift and a TACAN receiver is fitted. The F-15 has a conventional head up display (HUD), which eases workload considerably.

Never heared about the ALQ-154/155 and AAR-38. So when assessing the TEWS we have:
- AN/ALR-56 RWR
- ALQ-128 missile launch warning system (probabley) detecting the CW illumination for SAHR guided missiles
- AN/ALE-45 chaff/flare dispensers
- AN/ALQ-135 ECM
- ALQ-154/155 tail warning system?
- AAR-38 IR rearward warning system

So there're the following questions:
1.) Doesn't the ALR-56 provide 360° azimuth coverage?
2.) Did the F-15 have a rearward missile approach warner (AAR-38)?
3.) Could it be that the 6-20 GHz frequency coverage was just achieved by multiple systems? (could also be an explaination for the "various" systems)?
 
When was the Kopp article originally published? Some of his stuff is quite old. The ALQ-154/155 is a tail warning radar also associated with the B-52H while the AAR-38 might be the same early IR MAWS also found on the F-111. AFAIK neither system is actually found on production F-15s, but they might have been intended to be installed at the time of writing.
 
The article is from 1984. I have indeed never heared about the ALQ-154/155 & AAR-38 in relation to the F-15 before, so maybe it's indeed just wrong information.

BTW about the Raid Assessment Mode which is often stated here. This is a high resolution mode to identify and acquire single targets within a group of aircraft flying in a close formation. That's this modes purpose.
 
Raid Assessment Mode was deleted from the F-15 some time ago.

Not sure what Dr. Kopp's source is, but I never heard of any plans for the early F-15 to be fitted with the ALQ-154 or AAR-38. I have trawled through Boeing's original document archive, and hundreds of other documents besides, and this is the first I have read about it.

Oh, and the ALQ-128 is not a launch warning system ;)
 
Scorpion82 said:
Just found this article from Kopp with some avionics details, though not much has been released:
The F-15A was unique in having several firsts in the area of avionics. The AN/APG-63 radar was one of these. Designed for use in a single seat aircraft this radar had to perform both as an air superiority sensor and air defence system, both missions having specific requirements. For the air defence role the radar employs its all aspect look-down/shoot-down capability, 130 nm+ range and track-while-scan (8 targets), range-while-search, AIM-7 Sparrow illumination modes and it couples to an IFF interrogator.

For its period the radar had excellent ECCM ability, though this was necessary as the synthetic computer generated symbology displayed on the head-down ANMI scope conceals a lot of the observable effects of jamming.

Targets are adaptively tracked in ground clutter with Kalman tracking filters and the APG-63 has been known to lock onto fast moving road vehicles, outside of legitimate bogies. Early models were sensitive to Jet-Engine-Modulation (JEM) of radar returns, which affected the tracking filters when a target was illuminated up its tailpipe.

As an air superiority radar the APG-63 represented a quantum leap with its semi-automatic dogfight modes. In Boresight mode the pilot flies the F-15's longitudinal axis onto the target and the radar locks on, in Supersearch the pilot manoeuvres the target into the 20 degree HUD field of view, which is scanned by the radar, after lock-on the radar places a square box over the target to facilitate visual acquisition. The radar supports the gun and Sparrow/Sidewinder launch in these modes, all of which are selectable by throttle and stick buttons.

The APG-63 is fully modular, uses an X-band planar array antenna and a gridded TWT transmitter, the whole system weighing in at 494 lb. The aircraft has a single IBM built 32-bit 'Central Computer', which controls the weapon system, displays and radar; for its period its 3.4x10^5 instr/sec / 16 kword RAM capability was considerable. The F-15 also had the first self contained internal electronic warfare system in an air-air fighter. The core element is the large Loral ALR-56 Radar Homing And Warning System (ESM), supported by the ALQ-128 launch warning system, ALQ-154/155 tail warning system and AAR-38 infra-red rear warning system. These were complemented by the ALQ-135 internal jammer, all EW data is displayed on the head-down Tactical EW System (TEWS) scope. Precision navigation reference is provided by the AN/ASN-109 inertial system, with 1.5 nm/hr drift and a TACAN receiver is fitted. The F-15 has a conventional head up display (HUD), which eases workload considerably.

Never heared about the ALQ-154/155 and AAR-38. So when assessing the TEWS we have:
- AN/ALR-56 RWR
- ALQ-128 missile launch warning system (probabley) detecting the CW illumination for SAHR guided missiles
- AN/ALE-45 chaff/flare dispensers
- AN/ALQ-135 ECM
- ALQ-154/155 tail warning system?
- AAR-38 IR rearward warning system

So there're the following questions:
1.) Doesn't the ALR-56 provide 360° azimuth coverage?
2.) Did the F-15 have a rearward missile approach warner (AAR-38)?
3.) Could it be that the 6-20 GHz frequency coverage was just achieved by multiple systems? (could also be an explaination for the "various" systems)?

1) Somewhat - ALR-56 provides 360 degree horizontal coverage and 180 degree vertical through five antennas with six elements. The right wing RWR antenna contains a downward facing element, and there is a low band antenna located on the nose gear down (a small, black nubby one).

2) No, not that I've ever heard.

3) I'm not sure what you're referring to here - The AN/ALQ-135 consists of 7 main LRUs on the F-15E, a Band 1.5 system and Band 3 (Band 1.5 combined F-15 Band 1 and 2 capabilities) with two amplifiers each (and a preamp interface for RWR detection and jamming).

4) RE: The ALQ-128 - it's a highly classified system, to the point where F-15 TOs even have next to no information in them - everything about it is in classified TOs. As such, its operations simply can't be discussed here :)
 
I've been looking on the net for some APG-63(V)2 pictures, surprisingly theres only one repeated everyhere , and even more surprisinly , maby its just a light trick , but does the AESA antenna in thst pics seems to be covered whit something ? i meen like a plastic cover of some sort ? the T/R modules appears to be aranged diagonally , but i'm not sure ...are there any clearer pics of the thing out there ?

Thank you very much.
 

Attachments

  • apg63v2.jpg
    apg63v2.jpg
    39.1 KB · Views: 182
While not a part of the F-15's on-board avionics, I have been looking for information about the AN/ASQ-236 which is some sort of reconnaissance pod. I've never heard of the USAF using F-15s as reconnaissance aircraft, but the F-15E has been spotted carrying it on a few occasions.
 
Now, my question is: What would Silent Eagle cockpit looks like? Will it use an MFD? Just like the JSF or Super Hornet Block 3? I don't think todays pilot will like even an SE if the cockpit looks orthodox...
 
http://www.flightglobal.com/articles/2010/07/13/343998/farnborough-silent-eagle-makes-quiet-progress.html

Since the F-15SE was unveiled, Boeing has also revealed a potential new upgrade for the Silent Eagle's cockpit displays. Customers have the option of selecting a large area display measuring 11 x 20in (280 x 508mm), Jones says. The panorama display is being offered in tandem with the export version of the F/A-18E/F Super Hornet.
 

And after two long days, the seven hackers found a mother lode of vulnerabilities that — if exploited in real life — could have completely shut down the Trusted Aircraft Information Download Station, which collects reams of data from video cameras and sensors while the jet is in flight.

They even found bugs that the Air Force had tried but failed to fix after the same group of hackers performed similar tests in November [2018] without actually touching the device. …

The hackers lobbed a variety of attacks — including injecting the system with malware and even going at it with pliers and screwdrivers. When I saw it, the metal box that's usually secure on the aircraft had wires hanging out the front.


The hackers revealed their success to Will Roper, the Air Force’s top weapons-buyer, Marks wrote. “They were able to get back in through the back doors they already knew were open,” Roper said.

Roper is “trying to turn that around,” Marks explained. The acquisitions chief is “hopeful about the results of the U.S. government's newfound openness to ethical hackers.”

This is a drastic change from previous years, when the military would not allow hackers to try to search for vulnerabilities in extremely sensitive equipment, let alone take a literal whack at it.

But the Air Force is convinced that unless it allows America’s best hackers to search out all the digital vulnerabilities in its planes and weapons systems, then the best hackers from adversaries such as Russia, Iran and North Korea will find and exploit those vulnerabilities first.

“There are millions of lines of code that are in all of our aircraft and if there’s one of them that’s flawed, then a country that can’t build a fighter to shoot down that aircraft might take it out with just a few keystrokes,” Roper told Marks.


The F-15, which after 40 years of service is still the Air Force’s main air-to-air fighter, was the target of an earlier hack that wasn’t at all “ethical.”

Starting in 2014, North Korean hackers infiltrated a computer network belonging to a South Korean aerospace firm’s computer network.

South Korea operates one of the same models of F-15 that the United States does. Korea Aerospace Industries builds the F-15’s wings under contract with Boeing, the number-two U.S. defense firm. Boeing has described KAI as a “valued supplier.”
.

North Korea has neither the know-how nor the resources to copy the F-15 or even adapt the Eagle’s blueprints to its own designs. “North Korea will never build a serious air force,” Robert Edwin Kelly, an associate professor at Pusan National University in South Korea, told The Daily Beast.

The hack nevertheless exposed a vulnerability, one of many that the U.S. military hopes to address. In 2020 Roper wants to invite, to Nellis or Creech Air Force Bases near Las Vegas, what Marks called “vetted hackers.”

There the hackers “can probe for bugs on every digital system in a military plane, including for ways that bugs in one system can allow hackers to exploit other systems until they’ve gained effective control of the entire plane,” Marks explained.

“We want to bring this community to bear on real weapons systems and real airplanes,” Roper told Marks. “And if they have vulnerabilities, it would be best to find them before we go into conflict.”
 
Getting info out of the aircraft if they're allowed to physically dissect the electronics isn't too startling - bit like asking someone to check how good your front door lock is by handing them a chainsaw! The whole point of modern avionics architectures is that systems talks to each other internally, so if you get in, you'll have access to that data flow. And they're designed for external access because that's needed for maintenance and fault-finding. If black hat hackers are getting physical access to the aircraft you have a much bigger security problem than just the aircraft systems.

Remotely accessing the aircraft would be a much higher level of threat.

Worth noting that the F-15's original architecture dates back to when hacking really wasn't a thing outside of a few phreakers hacking the phone network. Everything from fridges to pacemakers has now had to be hardened against malicious interference. Hopefully the F-15's hardware has also been hardened since then, but there may be historical architecural vulnerabilities such as poor separation of data and control flows that it's difficult to change.
 
Last edited:
I’ve been reading Steve Fino’s book Tiger Check

In the chapter on the F-15, it goes into detail on AIMVAL/ACEVAL and the early F-15s, A model.

I’m surprised that TWS was NOT a function of the APG-63 right out of the box since they expected the F-15 to be outnumbered at least in some situations. The problem of the Sparrow requiring lock all the way to impact was known.
Yet it took AIMVAL/ACEVAL in 1977 to show the need to:
1. Stay out of the HOBS WVR fight
2. Shoot from BVR by proper “sorting”

Single target track was an impediment to deal with numerous BVR targets. The pilots said they needed and eventually got TWS.(and AMRAAM)
Did Hughes and/or the USAF drop the ball here in the initial development of the APG-63?
 
I’ve been reading Steve Fino’s book Tiger Check

In the chapter on the F-15, it goes into detail on AIMVAL/ACEVAL and the early F-15s, A model.

I’m surprised that TWS was NOT a function of the APG-63 right out of the box since they expected the F-15 to be outnumbered at least in some situations. The problem of the Sparrow requiring lock all the way to impact was known.
Yet it took AIMVAL/ACEVAL in 1977 to show the need to:
1. Stay out of the HOBS WVR fight
2. Shoot from BVR by proper “sorting”

Single target track was an impediment to deal with numerous BVR targets. The pilots said they needed and eventually got TWS.(and AMRAAM)
Did Hughes and/or the USAF drop the ball here in the initial development of the APG-63?
It looks like a great read, sadly very expensive or I'd have a copy already.

The F-15 avionics suite were heavily cost-cut from the original plans, including losing the ATAR TV sight and other changes. AN/APG-63 was always a much lower cost design than the AWG-9, and it had no weapon to employ in a TWS mode - it would have to switch to STT mode for Sparrow launch. What interests me is that AN/APG-65, normally described as derived from the AN/APG-63, had TWS modes.
 
So, the key element is likely the computing power. The PSP backfitted to the improved AN/APG-63 was based on the AN/APG-65's one. The time between F-15 IOC and F-18 IOC was a time of immense improvements in computing technology.
 
So i did a little digging on F-15's Avionics. I found this power point slides for "F-15FX" offers to Japan. It probably dated back to 2008-2009 time period.

The point of interest to me would be this part.

Cooling capacity.png

So apparently the baseline F-15E have cooling capacity of 52 KW and present load of some 35 KW. Meaning there would be a headroom of 17 KW. If we then take the old APG-63 away (this one having peak power of about 5 KW and perhaps 1-2 KW of average power) We would have further margin of 18-19 KW.

In order to approximate the actual required cooling for "Airborne Early Warning System Concepts" Which put the electronic load is about 20% from Antenna load. If we start from "original" APG-63/70 with 1.2 KW (25% duty cycle). The electronics would be about 240 Watts. And with another "rules of thumb" that dissipated heat must be about 30-90% of generated power in kilowatts (Take the 90%), bring the "required" cooling of original APG-63 to about 1.2 or 1.5 KW. We can perhaps take 2 KW for "margin of safety"

If one question on why not using the 5 KW ? Well Radar cooling requirement is determined from Average power as it's the one that actually determines the duty cycle which is how long your transmitter is actually "firing".

If one assumes that the entire 19 KW capacity is used only for the AESA radar. The result would be considerable.

This assumes 3 cm wavelength and 33% PAE For the AESA modules. It however discount any possible cooling required for additional electronics. Reason being i would describe later.

Cooling capacity F15Max.png

As seen the radar would be capable of 6.3 KW (maximum) average power and 25 KW peak at 25% duty cycle. Assuming the entire 19KW margin is used only for Radar. The required power is then about 6.3/0.33 = 19 KW of electrical power required by the radar.

However in reality it might not be so as it basically pushing the limit and leave no room for any possible growth in other electronics. The question is of course "How much would be the reasonable amount of cooling ?"

The electrical power available would easily support the 19 KW Radar, the additional electronics however which would consume about 3.8 KW and may require about 3.7 KW of cooling would easily exceed the margin.

Assuming 25% of duty cycle but with some relaxed parameters, maybe 5 KW average power (Thus 20 KW peak) The required cooling would be about 15 KW, same as the AN/APG-79 radar cooling demand, while the required power for the radar would be 15 KW with about 3 KW for the supporting electronics, this in turn would requires 2.9 KW of cooling. This in turn bring the entire Radar-electronics cooling demand into 17.9 KW So there is still about 1.1 KW of margin. Close but pretty good i think.

But anyway, what you guys think ?
 

Attachments

  • F15FX.pdf
    7.1 MB · Views: 74
Information book on F-15 by Soviet Ministry of Defense. 1983 Edition.


Not many pages but i see it's dense and "to the point" about what was known about F-15 at that time along with perhaps prediction of flight regimes and Avionics. The radar description is also there. Perhaps pertaining the Original AN/APG-63.

Some relevant page on the Radar. XD particularly range.
APG-63Soviet.png APG-63soviet2.png
 
Last edited:
The cooling capacity for the APG-63 is 12kw or more (I don't recall the exact details any longer, just base numbers that added up to over 12Kw, maybe as much as 13Kw) while the head-on detection range as tested on the original 70's APG-63 was 85nm vs. a 6m^2 target (T-33 in case you're wondering, it has that approximate head on RCS).
 
I have looked into above documents. Each of them shows radar system from some point of view.
1. Reliability report:
(Created in 81 , the last edit of the available report is December 81)
https://apps.dtic.mil/dtic/tr/fulltext/u2/a142071.pdf

This report written by manufacture - to show to client improvement of reliability. And rather tries to put system in good light.
There are some interesting parts in the introduction:
radar range chart:
radarRange.JPG

says something, but not clear what is "medium size target" 16m2 ?
And is also above chart is "based on radar equation", with note:

note.JPG

what is also interesting is radar modes (probably prePSP/ + PSP based system):

rafarModes.JPG

and also development of system in 70tees and early 80tees with prediction to late 80tees
There is seen some logic continues improvement improv.JPG

my comments:
- early models have some radar processor based on ferrite rings 16K of 24bit wordlength so not easy to improve.
Many functions existing in PSP were existing, (technical documentation, for PSP presents some "initial version" of software - and PSP box "just replace" existing "hard wired" processor. Anyway - in my opinion, baseline system from 74 "have a lot ability" including Medium PRF - that was novum then (not exiting in AWG-9 and earlier) - very solid, in my opinion (details, later)
- then in 77 solid memory was introduced. 24K words, this allows to introduce some changes in radar data processor based on AIMVAL/ACEVAL test (improved radar modes) This includes azimuth correction (if I am correct - if plane turn , "blips are corrected" according to turn)
- next big change was introduction of PSP in 80. Base system, I think, replace just replace (1:1?) many existing modes, adding some new A2G modes only,
pspReplace.JPG

but was base for many subsequent improvements: just after introduce Raid mode, and many more in following "tapes" (in the beginning tapes no : 1-4?):
earlyTapes.JPG

improvement2.JPG

The PSP was identical with PSP of APG-65 so probably both shares similar functionalities (they differs in blocks of displaying data):

psp.JPG

To be continued...
 
Last edited:
Not too much time to write more post. So to be concise:
F-15 Radar Set PSP Computer Program Development Specification, 1980
Very nice technical description of PSP.
It is easy to understand, and contains rather classical approach of signal processing*
*(contrary to some digital signal processing from Russian manuals - some kind of spectrum analysis I've never seen, and actually I do not understand)
PSP does a lot more I expected - from Doppler filtering - to generating image for radar "scope" , blips ageing , IFF correlation, selection of targets to track (by pilot) in "gates" etc.

Some facts, I've noticed:
- HPRF - uses two analog filter banks and then A/D - (for 30kHz bandwidth each) - they are combined in PSP.
Probably, there were no A/D fast enough at the beginning of 70 to accommodate full A?D conversion .
Much later it was possible, making possible to make modes like Range Gated HPRF (probably in the same fashion as MPRF - see below)
- Velocity search - uses 3 frame post detection integration, whereas HPRF Range While Search - 3 cycle (1 flat - for Doppler speed estimation, and with 2 linearly changed freq - for distance measurement)

The most impressing part - for me - was medium PRF.
Before , PSP step, - there is pulse compression step using 13- size Baker Codes:
1632525678547.png
This increases resolution 13 times - so range bins are more narrow. This is important - as that increases resolution and reduces level of "background" noise from earth (reflections from "discrete" objects can be removed by guard channel processing, but if resolution is tp low - discrete objects "smash" into one "background noise" that degrades radar performance in this mode)

As I understand there is 64 range bins for MPRF,
14 range bins are "blanked" during normal MRPF operation, by transmitter This corresponds exactly to 13 range bins (transmitting period) - compressed later to 1 bin resolution + 1 bin for transition from emitting to receiving (T/R switching period).
This means that duty cycle may be estimated as 13/64 - 0,2 - not so bad for med PRF, quite a lot energy, whereas retaining resolution.

(PRF - I think oscillates about 16kHz)
FFT is 32 size (16 filters). This gives 64 x 16 space - frequency matrix
There are 9 different cycles of MPRF - with different PRF - to fill gaps in "blind" ranges and speeds and to estimate real range and closing speed.
But additionally there is one non coherent period with 1 range period duration (2 range bins are "blanked") - and with 700 microseconds time range (that corresponds to max range 105km) - This was introduced in PSP version of radar.
I think just to detect close targets (energy of that single impulse is low).
Guard channel is always on in MRPF mode

Comparing to other known systems from that time (APG-66, Viggen radar, N-019 from Mig-29) - no one uses pulse compression.
For example N-019 has 12 range bins, but when use guard channel - this resolution is limited to just 6 ranges.
On the other hand - resolution in freq domain is (I write from memory) is 4x 20 channels for 10kHz

APG-66, Viggen uses Hi power TWT - so they do not need pulse compression as much as lower power TWT.
But at cost of no HPRF

What is also striking that compression is not part of PSP.
PSP was introduced later - initially just to replacing existing digital processor.
So, I can expect that this compression was not introduced with PSP, but was from very first version of APG-63, from 74 year.
This really impress me. Actually I expected that similar compression were possible only later - in late 80tees or even later...
And that compression is only possible - having full digitally sampled signal in Doppler bandwidth.
(Comparing N-019, does analog conversion to about 2.5kHz bandwidth - and then only is done "digital" spectrum computation)

All in all, combining low side-lobe array, guard channel, and pulse compression, and not so low mean power - I think all this result in quite sensible mode for all aspect detection over land, and not only "rear hemisphere" mode, even with lower peak power of TWT (comparing to hi power TWTs of MPRF "dedicated" design like APG-66, and Viggen's radar)

This is my speculation, but I think the same 13:1 compression was introduced later in from of range gated HPRF, (when speed of A/C increased, as well as PSP processing speed).
 

A couple of thoughts on these as a programmer of an only slightly later vintage. I'm not certain if the author of the Memory Loading page didn't understand what he was talking about, or was trying to spin something, but you cannot compare memory capacity with different word sizes by multiplying the number of words by the word size to get bits. Each instruction takes one word, so if you had 24kwords of 24 bit memory and swapped it for 48kwords of 16 bit, you've doubled the instruction capacity, even if you've only increased the number of bits by a third.

But it's more complicated than that, because if I'm remembering right, then some of the military 24 bit instruction sets managed to squeeze more than one instruction into a twenty four bit word, but only for some of their instructions, or squeezed some data into the back-end of an instruction word. So, for example (and I'm making these examples up because I definitely don't remember specifics and I'm not certain I ever knew them), you might have an instruction that did an addition, but part of the instruction word told it which registers to use, and another part then told it to use the result as an address to go and fetch a value from memory, something that might take you two or three instrictions in a 16 bit instruction set, or you might have the first eight bits of the word being an instruction, and the trailing 16 bits a data value.

And potentially a lot of this would be hand-tweeked machine-code, so very dense, whereas as available memory capacity went up you could increasingly rely on compilers to write machine code for you, even if they tended to be less efficient wrt memory use.

And for the second page, you can see how it splits the available memory into A/A, A/G, BIT, Loader and Common, but the reality is that you aren't going to be using the BIT (Built In Test) code while you're in the air, and possibly not the Loader as well depending on its precise function, so the available memory is potentially quite a bit more than this would seem to indicate.
 

And for the second page, you can see how it splits the available memory into A/A, A/G, BIT, Loader and Common, but the reality is that you aren't going to be using the BIT (Built In Test) code while you're in the air, and possibly not the Loader as well depending on its precise function, so the available memory is potentially quite a bit more than this would seem to indicate.
Well, this memory refers to ROM side - so all those programs are already written in hardware. Some of them are used in the air, some of them on the ground. But not using BIT or loader in the air does not make more available memory.
 
Well, this memory refers to ROM side - so all those programs are already written in hardware. Some of them are used in the air, some of them on the ground. But not using BIT or loader in the air does not make more available memory.
You're right, now I notice that the EAROM* segment of the table just sums the PSP and RDP segments of the table, though I'm not at all clear what Words (Memory) now represents as it's not the total of Constants and Instructions and I don't see what else you'd stick in ROM.

* Electrically Alterable ROM
 
Figures that HAC would use a non-standard term. The rest of us mere mortals know these devices as EEPROMs (Electronically Erasable Programmable Read Only Memory).
 
Figures that HAC would use a non-standard term. The rest of us mere mortals know these devices as EEPROMs (Electronically Erasable Programmable Read Only Memory).
There are a bunch of variations on the EEPROM theme, so this may have been a distinct technology to conventional EEPROMs as we understand them today.
 
  • Electrically alterable read-only memory (EAROM) is a type of EEPROM that can be modified one bit at a time. Writing is a very slow process and again needs higher voltage (usually around 12 V) than is used for read access. EAROMs are intended for applications that require infrequent and only partial rewriting. EAROM may be used as non-volatile storage for critical system setup information; in many applications, EAROM has been supplanted by CMOS RAM supplied by mains power and backed-up with a lithium battery.
Wikipedia
 
Well, this memory refers to ROM side - so all those programs are already written in hardware. Some of them are used in the air, some of them on the ground. But not using BIT or loader in the air does not make more available memory.
You're right, now I notice that the EAROM* segment of the table just sums the PSP and RDP segments of the table, though I'm not at all clear what Words (Memory) now represents as it's not the total of Constants and Instructions and I don't see what else you'd stick in ROM.

* Electrically Alterable ROM
Yes, this table seems to be strange as no of words is not simply sum of two previous columns (constant and program) o_O
But, for PSP:
the last column is 2 x number of instruction + number of constants.
What does it mean? Is instruction 2 word size - I mean 2 x 24 bits ?
For RDP - still no sense, as number of words is lesser than number of instructions...
 
Yes, this table seems to be strange as no of words is not simply sum of two previous columns (constant and program) o_O
But, for PSP:
the last column is 2 x number of instruction + number of constants.

Good spot on the 2xInst+const, but that doesn't hold for Common, which is 4xInst. *headdesk*
 
So apparently apg-63(v)1 entered service in April-may of 2001 but the AESA apg-63(v)2 actually entered service a few months before in December of 2000. Only one squadron at elmendorf afb, presumably due to its anti cruse missile mission. What’s the deal with this mysterious AESA? It entered service just two months after the j/apg-1 five months before the radar it’s supposedly based on, five years before the next US AESA radar, it’s apparently no closely related to apg-63(v)3 (which is apparently a closer relative to apg-79) and only ever ordered in small quantities. So like what’s the deal? Is their a story here?
 
Not much beyond that. The US had several AESA's in the pipeline around that time. These were in production systems but their official in-service date was tied to aircraft milestones and not their own maturity. I believe the first F-22 fitted with the APG-77 flew around the same time as the F-15's did with their AESA's.
 
Not much beyond that. The US had several AESA's in the pipeline around that time. These were in production systems but their official in-service date was tied to aircraft milestones and not their own maturity. I believe the first F-22 fitted with the APG-77 flew around the same time as the F-15's did with their AESA's.
I actually did some digging and antidotally I found APG-63(v)2 is much heavier then any other member of its family, like 400kg. It requires significant weight be added to the back for balance. The software was tailored for finding cruise missiles and normal modes found on other versions of apg-63 weren’t added till around 2005. They would be paired with a normal F-15 for cruise missile hunting. Looking for something official looking that says that now.
 

Similar threads

Back
Top Bottom