"Combat Tree" AN/APX-80/81 and QRC-248

Oops... I didn't notice that there was a separate category for this, so a repeat of what I wrote in the Soviet IFF thread:

I think that the book "Clashes" by Michel might be a good starting point for Combat Tree research. The system seems to be rather secret on the F-15, but not nearly so regarding its use by the F-4 in Vietnam.
 
According to now declassified U.S. sources, Combat Tree was an IFF
interrogator that could interrogate MiG IFF transponders. It was first
installed on a small number (8?) of F-4Ds in late'71 or so, which were
operated by the 432nd TRW (The serial numbers are given in Anthony
Thornborough's "The Phantom Story"). I've checked his list against the Red
Baron reports, which identify whether and which a/c in a flight are Combat
Tree equipped and which contacts were due to Combat Tree, and cross-checked
those incidents with listings of those kills which include the a/c serial
numbers. They match).

Because of the small numbers available, typically only the flight lead (01)
and (when available) the element lead (03) would be in 'Tree'-equipped birds.
02 and 04 were often in F-4Es, to provide some gun capability without
forcing the F-4Ds to carry gun pods. Because of the loss of several of these
a/c, some more (about 20?) were modified in mid '72 (available July, I
think). Later, when the Rivet Haste slatted F-4Es showed up in theater to
re-equip the 555th, they were apparently all equipped with it.

Combat Tree had several positive effects. First, it allowed U.S. fighters to
acquire MiGs at much greater ranges than they could from a primary return
(i.e. a skin paint). Second, it allowed them to make contact looking down in
many cases. This meant that free-roving MiGCAPs were much more effective
than they had been; previously, the NVn would just vector there MiGs around
U.S. MiGCAPs by keeping them low until they'd gotten out of our a/c's radar
arcs. It also meant that we were allowed to take more BVR shots with the
AIM-7 than we'd previously been able to. An example the initial engagement
of Oyster flight on 10 May 1972. Lodge in 01 and Ritchie in 03 were both
flying Tree-equipped a/c, and Lodge's and Markle's kills were both fired BVR
because of a Combat Tree ident.

It wasn't perfect. There were limits as to how close a Tree contact could be
to other a/c, before a BVR shot was allowed. In at least one case, an a/c
ID'ed as a MiG by Tree was visually ID'ed by TISEO gear as an F-4,
fortunately before a shot was taken. In addition, the NVN was montitoring our
radio coms. When our pilots started to report contacts on MiGs at previously
unheard of ranges, they started to get suspicious, and curtailed their use of
transponders, only turning them on at turn points or the like instead of
using them continuously as they had previously.

For the best, most accessible discussion of Combat Tree, QRC-248 (a similar
interrogator installed on board the EC-121s from 1967; Michel, quoting from a
declassified document, says theat QRC-248 would tyypically allow an EC-121
Connie which might have trouble detecting a MiG at 100nm at medium altitude,
to detect them at 175nm at low altitude) and the rest of the Vietnam air
campaign, you should find a copy of "Clashes," by Marshall Michel (sp?).
This is based on declassified U.S. documents now available from the
Historical Research Center in the archives at Maxwell AFB, checked by
Michel's own experience. The HRC has a website listing some of what's
available, although it tends to be a bit out of date, and last time I
checked, none of it was on-line; yopu have to order the stuff by mail, or go
there. I have a friend who works there, and he's made copies for me.

I got my copy of "Clashes" from amazon.com for 30% off a year or two ago,
when it was published. If you're interested in how an air campaign is
fought, this is one book you've got to have. Another excellent book for
details of the F-4's weapons is the one by Thornborough mentioned above.

Guy
Click to expand...

'Combat Tree '- info sought about

groups.google.co.uk groups.google.co.uk
 
Prior to the active use of QRC-248, the MiGs had really got their act together, and first demonstrated their new tactics on August 23rd, 1967. Quoting from "Clashes":

"The MiGs [two Mig-21s] stayed in the ground clutter to avoid interception by U.S. airborne radars, but their transponders were picked up by one of the
orbiting EC-121's QRC-248 [in passive mode]. The EC-121 called out the MiGs' location, but the escort did not react. Once the North Vietnamese GCI saw the MiGs were abeam the strike flights and outside the radar range of the MiGCAP F-4s [actually escorts], they instructed the MiGs to climb quickly to 28,000 feet. This put the MiGs above the overcast and on one of the flanks of the strike force, still unseen by the F-4 radars. The MiGs closed on the
force and, on GCI's command, dived out of the overcast and swept down at high speed on Ford, one of the rear F-4 strike flights. The MiGs attacked with Atolls; the first warning the F-4s had was when Ford 3 saw a missile hit Ford 4, which blew up in a ball of flame. At the same time, Ford 2 watched an
Atoll pass by his wing and destroy Ford 1. The MiGs escaped unscathed."
Click to expand...

f-104 in vietnam

groups.google.co.uk groups.google.co.uk
 
Last edited:
With the Combat Tree in your Phantom you don't need to activate your radar. You just point the radar antenna into direction you want to scan, and activate the APX-80. It will immediately show you (it uses the radar display) all the "MiGs" with active IFF-transponders out to 50-60kms in the air and on the ground in front of you. I.e. it will show you also those MiGs which your radar couldn't detect except they are flying at ranges closer than 15-20kms. All of this regardless of the level at which the MiGs operate (the system detects even those SRO-2s on the planes which are on the ground, but are eventually testing them, or rolling to take off etc.).

And then, by pushing "a button", you can also trigger IFF-transponders of enemy fighters which are not set on active.

With other words: you know what is going on ahead of you, where are potentional enemies, and what are they doing, but they don't.
Click to expand...

Tom on acig.org, somewhat overstating matters as usual.
 
QRC-248 could interrogate only SRO-2 IFF. But on single EC-121K was mounted Rivet Top system, and in Clashes is writen that it could also interrogate SRO-1 and SOD-57 IFFs. What were designations of Rivet Top EIFF systems? Somewhere I have found that APX-84, and 85 were used. Have anybody some details about it?
Other parts of Rivet Top were Rivet Gym stations used for COMINT between GCI and MiGs.
Click to expand...

Bunny on acig.org
 
Last edited:
Indeed, if you read "Clashes" you'll see that this capability and bad COMSEC almost certainly led the NVN to believe that the SRO-2 was no longer secure. Apparently, the Tree-equipped a/c were able to pick up squawking MiGs at 50-60nm, even in ground clutter, and on several occasions Tree contacts were called out on the radio with these ranges given (with the crews getting their butts chewed back at base). The N. Vietnamese (and the Soviets) were listening in all the time, so this sudden doubling of radar range undoubtedly made them perk up. Shortly thereafter, it was noticed that the VPAF had begun to use their transponders much less frequently, only turning them on at turn points and at other critical moments in intercepts, rather than using them continuously as had previously been the case.
Click to expand...
 
Last edited:
Since the APX-81 used the dipole array on the APQ-109 antenna (normally used by the APX-76 IFF interrogator), you would only get a reponse from the bad guys's IFF when the antenna was pointed at it. The APX-81 video was inserted on to the b-sweep and lit it up whenever the antenna was pointed in the right direction. Now, if you did this while your own radar was in stby then normal returns didn't interfere with the APX video. Of course you didn't get range info, but a quick radar look with the antenna on the target gave enough data to sneak up on the bad guy without setting off his RWR.
Click to expand...


The Soviet tactical doctrine was that fighters were to be under ground control. This entailed the ground controller knowing where the fighter was. To do this, the fighter's IFF unit needed to be kept on at all times.

In the mid-1960s, someone working for the USAF found a way to interrogate Soviet IFF units. The resulting QRC-248 was first used onboard EC-121s operating in the Vietnam conflict. This capability was top secret and initially, the EC-121 crews were only allowed to monitor when the Soviet IFF sets were interrogated by ground radar, a so-called "passive" mode. Eventually, they were allowed to actually interrogate the IFF sets themselves, in an "active" mode. This capability was first used in 1967 and resulted in a dramatic improvement in US air-air results. The EC-121s could issue MiG warnings to specific aircraft, with position information.

Eventually, a APX-80 Combat Tree system was installed in F-4s operating in Vietnam. This allowed the fighters themselves to interrogate their foe's IFF units. The F-4 crews could then see on a cockpit display where the opposing MiGs were.

The North Vietnamese caught on and started turning their IFF units off. However, now the ground controllers had a hard time controlling them, diminishing their effectiveness.
Click to expand...

Passive detection

groups.google.com groups.google.com
 
I don't know if it was the same as the "Combat Tree" set, but the F-4Es I worked on had a Soviet IFF interrogator called the APX-81 system. There was a receiver-transmitter, the RT-961, located inside Door 19 behind the rear cockpit, and it shared a control head with the APX-76 IFF/SIF interrogator system, the APX-76 being also located in Door 19 but consisting of the RT-868, SA-1568, and SN-416B. The control head was located in the rear cockpit on the lower part of the rear main panel, just above the WSO's left knee. The control head was labeled "APX-80", which I believe may have just been a sort of average of the designations APX-76 and APX-81. I don't believe there ever actually was an APX-80 system. The APX-81 specific switches were labled something similar to "Type 1" and "Type 2" (can't recall the exact verbage just now), and both switches were three positioned: OFF, PASSIVE, and ACTIVE. Both APX-76 and APX-81 displayed their data on the radar scopes; there wasn't an independent display. Both APX-76 and APX-81 shared the "T" shaped dipole antennas mounted on the radar dish, four on the F-4E and I believe 8 on the F-4D. The APX-81 didn't have a destruct charge when I worked on it, and I was assigned to flightline maintenance so never got to see the inside of the RT-961, so I don't know if there was provisions for a destruct charge inside.

To get to the point, there might not have been any really obvious cockpit differences for a Combat Tree F-4D, I've never seen any cockpit photos or diagrams calling it out as a separate system. My guess is it also used the APX-80 control head, if it wasn't the APX-81 system I worked on.

Scott Wilson
Click to expand...

Looking for Photos of a Combat Tree F-4 Phantom

Hi there I was reading some Osprey Phanton books (Iranian and USAF Phantoms) and both of them mentioned that part of their success at air success was related to some equipment devenloped after 6 days War (the Israeli supply the Iff transpoders from URRS planes, so they can interrogate planes with...
www.arcforums.com www.arcforums.com
 
I don't think so. You might be able to find a cockpit shot or drawing of the control panel, but that's about it. It only has mode (OFF-PASSIVE-ACTIVE) and type (1 or 2) so it's not a very interesting control panel. It displayed on the radar screen.
 
Thank you. So far the search for the drawing or intelligible cockpit shot was fruitless, alas.
 
Last edited:
The only images of the APX-76 and APX-80 on the F-4E & G (item #35) and F-4J (item #5) that I managed to find.

F-4E RIO.jpg F-4J RIO.jpg

Considering that a separate IFF (APX-72 or similar) was available to the pilot in the front cockpit, I wonder what kind of "code" was input on the control unit in RIO's place? The APX-76/-80 control unit has five thumb dials. The first one is for MODE, while the remaining four are marked CODE.

How many figure-long was the IFF code on the SRO-2 Khrom? Three or four? If the "Combat Tree" systems (APX-80/-81) were able to trigger responses from adversary IFF transponders, then the code used on the very day by the Soviet/VN AD/GCI must have been known before in order to trigger the response. I understand that if a wrong code was sent, the SRO-2 would simply not reply. Right?
 
yahya said:
The only images of the APX-76 and APX-80 on the F-4E & G (item #35) and F-4J (item #5) that I managed to find.

View attachment 753916View attachment 753917

Considering that a separate IFF (APX-72 or similar) was available to the pilot in the front cockpit, I wonder what kind of "code" was input on the control unit in RIO's place? The APX-76/-80 control unit has five thumb dials. The first one is for MODE, while the remaining four are marked CODE.

How many figure-long was the IFF code on the SRO-2 Khrom? Three or four? If the "Combat Tree" systems (APX-80/-81) were able to trigger responses from adversary IFF transponders, then the code used on the very day by the Soviet/VN AD/GCI must have been known before in order to trigger the response. I understand that if a wrong code was sent, the SRO-2 would simply not reply. Right?
Click to expand...
The SRO-2 didn’t exactly transmit codes. What the combat tree is taking advantage of is that SRO-1/2 did not need a particular “code” to respond, but just an interrogation of the right frequency. So combat tree is sending out the fake interrogation, and SRO-1/2 is sending back its response no matter the characteristics of the interrogation.

For SRO-2M, because of combat tree, it was then added that a specific type of “interrogation” was needed before a reply was sent, and if the “wrong” interrogation was sent then no response was sent back, versus SRO-1/2 which would send a response no matter the form of interrogation.

There is a good released CIA paper on SRO-2, you might be able to find it. Here are some images of the response.

My understanding of combat tree is that this IFF response was just shown as a normal radar “brick” on azimuth range on the radar display; so no “code” shown. This other secret projects thread has a lot of good info and links https://www.secretprojects.co.uk/threads/soviet-russian-iff.792/
 

Attachments

  • IMG_8459.jpeg
    IMG_8459.jpeg
    62.1 KB · Views: 11
  • IMG_8460.jpeg
    IMG_8460.jpeg
    171.7 KB · Views: 9
  • IMG_8461.jpeg
    IMG_8461.jpeg
    225.5 KB · Views: 10
  • IMG_8462.jpeg
    IMG_8462.jpeg
    290.7 KB · Views: 15
Amazing. That would mean that the original SRO-1/SRO-2 were badly designed. Thus, spoofing them would be feasible even for tech kids not to mention a major power.

By the way, did the APX-80/APX-80A control unit's challenge function simply trigger the responses from SRO-2, or a different equipment had to be used for that purpose?
 
yahya said:
Amazing. That would mean that the original SRO-1/SRO-2 were badly designed. Thus, spoofing them would be feasible even for tech kids not to mention a major power.

By the way, did the APX-80/APX-80A control unit's challenge function simply trigger the responses from SRO-2, or a different equipment had to be used for that purpose?
Click to expand...
I do not know, but yes it was a design that assumed the enemy would not try to take advantage of it. Thus, combat tree would work on MiG-15 with IFF, MiG-17, MiG-19, early MiG-21, but not later MiG-21 and beyond.

Whatever transmitted the fake interrogation, it was something that could do this 670 mhz and 0.8 microsecond pulse width transmission in a series of 3 and 4 micro seconds apart.

Since APQ-120 is X band and SRO-2 responds to that, I would not be surprised if the F-4 native IFF transponder was able to do it, simply being directed by the combat tree unit.
 
Aeria Gloria is not really correct here.

Most people re thinking of Western Mode XII IFF here, which uses 4 digit codes (10,000 possible codes) and encryption. SRO-2 is much older and simpler.

The primary technical deficiency of the SRO-2 IFF is that there were only 12 possible codes selected by a rotary dial. These codes were simply 12 specific pulse repetition rates, so the IFF interrogation signal sent was just three sets of pulses (pulse trains) with specific PRR and correct spacing between the pulse trains sent at 670MHz (I band). When the third correct pulse train was received, SRO-2 would send a response, which delayed response resulted in the interrogating radar drawing an arc or line at a slightly bigger range on the display around / above the target blip (depending on scope type).

The additional S and X band antennas on SRO-2 were used so that when a compatible surface GCI radar (s band) or airborne radar (x band) pulse was received at the same time as the I band (670MHz) IFF pulse, the response was sent immediately after the first pulse train was received not the third, so the reply would be only very slightly delayed compared to the main radar return.

Why?

Typically this would result in a "bird" shape rather than an arc or line IFF mark as the wider beam of the IFF interrogator would draw "wings" offset in range either side of the target, where only I band was triggering it, then in the centre the IFF would receive both L band (IFF) and S or X band (main radar) from the narrower main radar beam and the delay was almost removed, thereby drawing the IFF line closer to the target in the centre of the line. This presumably helped anchor the IFF "wings" to the correct target blip visually.

If the SRO-2 device were to fall into enemy hands however, they would know all the needed parameters for the system, and they could build a device can produce that can produce the correct 12 codes.

Combat Tree can simply try all 12 possible codes in rapid succession until a response was triggered - sending a wrong code had no negative impact with SRO-2. Once you know the right code you could even store it and use it.

You can just send L band IFF, and know the response is delayed a little and compensate for the slight position inaccuracy when drawing the target.

Or you send the L band IFF pulse train at the same time as an S or X band pulse, then the SRO-2 responds almost instantly with its L band reply giving a very accurate location on the screen.

Either way, you can easily tell that the target has an SRO-2 in it and is hence hostile.

Combat Tree is inherently lookdown capable, as you are receiving L band signals direct from the target.
.
The issue was compounded by the linked SOD-57/SOD-57M unit which would send out the aircraft height, fuel state etc (this being valuable info for the GCI radar operator) which was super handy for the F-4 WSO to know.

One interesting thing - US IFF was also I band, so it was possible to reuse the existing IFF dipoles on the F-4 radar - and the F-4 had an X band radar too.

The only real countermeasures once the SRO-2 was compromised were:

Turn off your IFF set
Change the PRRs (these were implemented as 12 physical filters so presumably you could in theory retrofit new filters with different PRRs, but you need to change them in everything)
Implement a larger number of codes (new hardware)
Penalise wrong interrogation attempts (e.g. if a "wrong code" is received, stop listening for x seconds)

I haven't read up on SRO-2M changes. SRO-2M was also compromised later on after an example was obtained.
 
Last edited:
overscan (PaulMM) said:
Aeria Gloria is not really correct here.

Most people re thinking of Western Mode XII IFF here, which uses 4 digit codes (10,000 possible codes) and encryption. SRO-2 is much older and simpler.

The primary technical deficiency of the SRO-2 IFF is that there were only 12 possible codes selected by a rotary dial. These codes were simply 12 specific pulse repetition rates, so the IFF interrogation signal sent was just three sets of pulses (pulse trains) with specific PRR and correct spacing between the pulse trains sent at 670MHz (I band). When the third correct pulse train was received, SRO-2 would send a response, which delayed response resulted in the interrogating radar drawing an arc or line at a slightly bigger range on the display around / above the target blip (depending on scope type).

The additional S and X band antennas on SRO-2 were used so that when a compatible surface GCI radar (s band) or airborne radar (x band) pulse was received at the same time as the I band (670MHz) IFF pulse, the response was sent immediately after the first pulse train was received not the third, so the reply would be only very slightly delayed compared to the main radar return.

Why?

Typically this would result in a "bird" shape rather than an arc or line IFF mark as the wider beam of the IFF interrogator would draw "wings" offset in range either side of the target, where only I band was triggering it, then in the centre the IFF would receive both L band (IFF) and S or X band (main radar) from the narrower main radar beam and the delay was almost removed, thereby drawing the IFF line closer to the target in the centre of the line. This presumably helped anchor the IFF "wings" to the correct target blip visually.

If the SRO-2 device were to fall into enemy hands however, they would know all the needed parameters for the system, and they could build a device can produce that can produce the correct 12 codes.

Combat Tree can simply try all 12 possible codes in rapid succession until a response was triggered - sending a wrong code had no negative impact with SRO-2. Once you know the right code you could even store it and use it.

You can just send L band IFF, and know the response is delayed a little and compensate for the slight position inaccuracy when drawing the target.

Or you send the L band IFF pulse train at the same time as an S or X band pulse, then the SRO-2 responds almost instantly with its L band reply giving a very accurate location on the screen.

Either way, you can easily tell that the target has an SRO-2 in it and is hence hostile.

Combat Tree is inherently lookdown capable, as you are receiving L band signals direct from the target.
.
The issue was compounded by the linked SOD-57/SOD-57M unit which would send out the aircraft height, fuel state etc (this being valuable info for the GCI radar operator) which was super handy for the F-4 WSO to know.

One interesting thing - US IFF was also I band, so it was possible to reuse the existing IFF dipoles on the F-4 radar - and the F-4 had an X band radar too.

The only real countermeasures once the SRO-2 was compromised were:

Turn off your IFF set
Change the PRRs (these were implemented as 12 physical filters so presumably you could in theory retrofit new filters with different PRRs, but you need to change them in everything)
Implement a larger number of codes (new hardware)
Penalise wrong interrogation attempts (e.g. if a "wrong code" is received, stop listening for x seconds)

I haven't read up on SRO-2M changes. SRO-2M was also compromised later on after an example was obtained.
Click to expand...
Thank you for the correction, very interesting. Wonder where I heard what I thought earlier.
 
What I wrote is from reading the documents, so its possible I am wrong, but it fits the anecdotal evidence I have about Combat Tree.

The USSR started development of Parol in the 1960s. The developers of SRO-2 suggested just moving to a system with hundreds of thousnds of codes, but still easily compromisable if the hardware fell into enemy hands. Parol developers suggested an encryption based system instead. It took a long time and serious effort to get done.

I've seen a source suggest that even first gen Parol was compromised.
 
Last edited:
Reading the CIA document, I or someone I read must’ve mixed up the non coded SRO-1 with SRO-2. It doesn’t specify too much about the coding of the code in the interrogation signal, but I see what I missed now.

Despite being compromised by combat tree, it is a pretty genius and accurate system. Having the response shown on the interrogating radar’s display in various ways, including as a second target or an arc seems to be a unique and clever solution that was also used by Parol for good reason.

For Parol switching to encryption was definitely the correct choice, however difficult and time consuming it sounds like it was to engineer it.
 

Attachments

  • IMG_8892.jpeg
    IMG_8892.jpeg
    180.5 KB · Views: 6
Last edited:
Well, from my reading SRO-1 had 28 codes, but the system was very slow and the display not integrated to the radar display. SRO-2 was a more elegant solution.
 
overscan (PaulMM) said:
Aeria Gloria is not really correct here.

Most people re thinking of Western Mode XII IFF here, which uses 4 digit codes (10,000 possible codes) and encryption. SRO-2 is much older and simpler.

The primary technical deficiency of the SRO-2 IFF is that there were only 12 possible codes selected by a rotary dial. These codes were simply 12 specific pulse repetition rates, so the IFF interrogation signal sent was just three sets of pulses (pulse trains) with specific PRR and correct spacing between the pulse trains sent at 670MHz (I band). When the third correct pulse train was received, SRO-2 would send a response, which delayed response resulted in the interrogating radar drawing an arc or line at a slightly bigger range on the display around / above the target blip (depending on scope type).

The additional S and X band antennas on SRO-2 were used so that when a compatible surface GCI radar (s band) or airborne radar (x band) pulse was received at the same time as the I band (670MHz) IFF pulse, the response was sent immediately after the first pulse train was received not the third, so the reply would be only very slightly delayed compared to the main radar return.

Why?

Typically this would result in a "bird" shape rather than an arc or line IFF mark as the wider beam of the IFF interrogator would draw "wings" offset in range either side of the target, where only I band was triggering it, then in the centre the IFF would receive both L band (IFF) and S or X band (main radar) from the narrower main radar beam and the delay was almost removed, thereby drawing the IFF line closer to the target in the centre of the line. This presumably helped anchor the IFF "wings" to the correct target blip visually.

If the SRO-2 device were to fall into enemy hands however, they would know all the needed parameters for the system, and they could build a device can produce that can produce the correct 12 codes.

Combat Tree can simply try all 12 possible codes in rapid succession until a response was triggered - sending a wrong code had no negative impact with SRO-2. Once you know the right code you could even store it and use it.

You can just send L band IFF, and know the response is delayed a little and compensate for the slight position inaccuracy when drawing the target.

Or you send the L band IFF pulse train at the same time as an S or X band pulse, then the SRO-2 responds almost instantly with its L band reply giving a very accurate location on the screen.

Either way, you can easily tell that the target has an SRO-2 in it and is hence hostile.

Combat Tree is inherently lookdown capable, as you are receiving L band signals direct from the target.
.
The issue was compounded by the linked SOD-57/SOD-57M unit which would send out the aircraft height, fuel state etc (this being valuable info for the GCI radar operator) which was super handy for the F-4 WSO to know.

One interesting thing - US IFF was also I band, so it was possible to reuse the existing IFF dipoles on the F-4 radar - and the F-4 had an X band radar too.

The only real countermeasures once the SRO-2 was compromised were:

Turn off your IFF set
Change the PRRs (these were implemented as 12 physical filters so presumably you could in theory retrofit new filters with different PRRs, but you need to change them in everything)
Implement a larger number of codes (new hardware)
Penalise wrong interrogation attempts (e.g. if a "wrong code" is received, stop listening for x seconds)

I haven't read up on SRO-2M changes. SRO-2M was also compromised later on after an example was obtained.
Click to expand...

Great info! Thank you. Did your sources elaborate how exactly the CT interrogator that tested the twelve SRO-2 "codes" looked like? Was it integrated into the APX-80 or APX-81 interrogator control seen in the F-4 WSO cockpit as depicted in the entry from the link below? Did the "Challenge" button on the control unit trigger the transmission of the twelve "codes" consecutively?

 

Similar threads

Please donate to support the forum.

Back
Top Bottom