SPO-10 RWR behavior

[Bogey Jammer]

Hello all

 

The thread here turned to be a SPO-10 behavior discussion and was asked to us the users to tell to devs what is wrong with it.

 

But instead to tell what is wrong, I've javascripted a toy to show how I think it should behave :

 

 

It's just a html file to open in your browser.

 

Known bugs:

• doesn't work well at the red's 6 o'clock
  

 

:D

Bogey Jammer's SPO-10 simulator.zip

[il_corleone]

Thanks this is very usefull

[Bogey Jammer]

Well, I don't know if it is supposed to be useful. It's made to compare what it is expected with the current model of the SPO-10 in game.

 

So I can say:

 

• The SPO-10 IRL seems to be completely passive without any analysis algorithm or active filtering. The eyes just blink only when a radar ray pass over the antennas.
  
• Most aircrafts, or at least, players, use scanning with multiple lines for searching the area, so the blinking frequency of the SPO should be reduced 2 or 4 times in this case.
  
• The blinking frequency raises abruptly at 35km for all AI aircraft radars. That must be a pilot behavior which switches to a narrower angle for monitoring the bandits, not a SPO-10 feature. I don't know yet how it works against human players.
  
• The current lock rendering in game is in my opinion in need to be reworked. A lock is a continuous illumination of the SPO's antenna, and whatever how much the lock range is, the SPO must have continuously lit eye(s).
  
• I set a threshold of 5km for the activation of all the four eyes, blinking or continuously lit, as stated in the real pilot's manual, probably because the radar beam is so strong at this short range that all the SPO's antenna can sense it.
  

[Custard]

Wow, that's a fantastic tool there. Illustrates the necessary changes far more clearly than the walls of text that have been used previously.

Edited June 22, 2016 by Custard

[Cobra847]

[*]The SPO-10 IRL seems to be completely passive without any analysis algorithm or active filtering. The eyes just blink only when a radar ray pass over the antennas.

 

That does not seem to be the case from the MiG-21 manual and its' description of the SPO-10.

Once an enemy aircraft begins to approach the MiG-21 in a scanning cycle, the frequency of the audiovisual feedback increases.

 

[*]Most aircrafts, or at least, players, use scanning with multiple lines for searching the area, so the blinking frequency of the SPO should be reduced 2 or 4 times in this case

 

Why should we artificially decrease the frequency of our SPO if our implementation is being told by DCS that it is being pinged? Also, see above with regards to frequency increase.

 

[*]The blinking frequency raises abruptly at 35km for all AI aircraft radars. That must be a pilot behavior which switches to a narrower angle for monitoring the bandits, not a SPO-10 feature. I don't know yet how it works against human players.

 

Probably just a simplification with AI radar strength in the sim.

 

[*]The current lock rendering in game is in my opinion in need to be reworked. A lock is a continuous illumination of the SPO's antenna, and whatever how much the lock range is, the SPO must have continuously lit eye(s).

 

This is directly at odds with the documentation:

"When the illuminating radar switches from SCAN to LOCK-ON mode, the frequency of the sound signal sharply grows and the lights begin to flicker."

 

Additionally- this is highly dependent on the illuminating radar and selected radar mode.

 

[*]I set a threshold of 5km for the activation of all the four eyes, blinking or continuously lit, as stated in the real pilot's manual, probably because the radar beam is so strong at this short range that all the SPO's antenna can sense it.

 

The range this happens at is defined as between 5 and 2 km.

[grunf]

"When the illuminating radar switches from SCAN to LOCK-ON mode, the frequency of the sound signal sharply grows and the lights begin to flicker."

Is this properly implemented in latest open beta?

 

The way I see how the RWR should operate, is one light on (or two, depending on threat direction) in TWS mode, changing frequency as it gets closer. This part is fine.

 

After the lock on, the same light (or two) should start to flicker and the tone should change. I've never seen this happen, although I only tested it against AI, so i can't be sure when exactly I was locked.

 

When the distance closes, all ligts should go on. Currently this happens too early, at around 7km. I know this depends on radar strength, but it happens even with MiG-21's radar which is not the strongest. Again, AI.

[Bogey Jammer]

I repost the manual page with the SPO instructions :

 

 

That does not seem to be the case from the MiG-21 manual and its' description of the SPO-10.

Once an enemy aircraft begins to approach the MiG-21 in a scanning cycle, the frequency of the audiovisual feedback increases.

 

Why should we artificially decrease the frequency of our SPO if our implementation is being told by DCS that it is being pinged? Also, see above with regards to frequency increase.

 

OK I missed to read that in the manual. But what is strange about this feature (blinking frequency x3 to be precise) is the fact that it should happen almost all the time against an aircraft because of the forward oriented radar, except when the MiG-21 is fleeing from a slower enemy at the 6. And how technologically the SPO can tell if the signal is approaching anyway ? memory ? Also how does it deal with the multiplication of enemy's scan's frequency ? Does it wait for a new beam pass to measure the period time, assuming it can tell that the second pass comes from the exact same source ? or does it just burst 3 blinks immediately then waits for another pass to burst 3 blinks again ?

 

It would be interesting to get the complete electronic schematics of the SPO-10, to check everything.

 

This is directly at odds with the documentation:

"When the illuminating radar switches from SCAN to LOCK-ON mode, the frequency of the sound signal sharply grows and the lights begin to flicker."

 

Additionally- this is highly dependent on the illuminating radar and selected radar mode.

OK but not completely, then the eye(s) should go from flickering to constant illumination depending of the radar beam's strengh. But I doubt that it blinks purely rhythmically with an increasing tempo. It might be a random pulse noise with increasing alpha.

I will update my toy to reflect these clarifications.

[Cobra847]

Is this properly implemented in latest open beta?

 

This may be a bug. It was implemented.

[Toybasher]

Only issue I've noticed is the way it detects ground radars like from SAM sites.

 

I tried SA-3, SA-10 sites etc.

 

I never noticed a change when locked. From what I've been reading I should be seeing the lights blinking faster, etc. but with ground targets I get no lock warning whatsoever aside from all 4 lights coming on when I'm close enough.

 

I get scan warnings (Beep and a blink) from the search radars, but when I actually get locked I can't notice any change in RWR behavior. Often I'd get startled when I eat a missile without any warning whatsoever I've been locked or fired upon.

 

From what I've read, it looks like air radars show locks, but not ground based radars.

[Frederf]

Why should we artificially decrease the frequency of our SPO if our implementation is being told by DCS that it is being pinged?

 

I think he means in 2-3-4 bar scan modes that some bars may miss above or below the MiG-21 in altitude. Sweeps in azimuth which miss the MiG vertically obviously shouldn't be detected but it sounds like the "is pinged" code already does this discrimination.

[foxbat155]

It would be interesting to get the complete electronic schematics of the SPO-10, to check everything.

 

No problem my friend:

 

Whole set for fighters ( bit different was for bombers, transport etc ). 1. Antennas, 2.Power unit, 3.Light and sound indication unit, 4.Indicator, 5. Video amplifiers.

 

SPO-10 functional scheme.

 

Indicator scheme.

 

Video amplifier scheme.

 

Inside light/signal block.

 

Light/signal unit scheme.

 

S3M-11M block scheme.

 

Board 1M scheme.

 

Board 2M scheme.

 

Board 3M scheme.

 

S3M-4A scheme.

 

S3M-4 scheme.

 

S3M-10 scheme.

 

Bit description:

Board 1M - ASO automatic dump unit,

Board 2M - cascade amplifier,

Board 3M - blocking and sound signal unit,

S3M-4/S3M-4A - steering unit,

S3M-11M - light signaling channel ( four in total ).

[Bogey Jammer]

Thanks a lot mate !!

 

(Станция предупреждения об облучении СПО-10) ;)

[foxbat155]

If you read russian I can send you whole manual.

[Harle]

If you read russian I can send you whole manual.

 

I do.

Gimme! Gimme one! )

[Bogey Jammer]

Thanks but I just found it on the web: http://bookre.org/reader?file=614809

 

Maybe Cobra wants it ? :D

[foxbat155]

Yep that's the same manual. Enjoy.:thumbup:

[Frederf]

I don't understand the SPO-10 in any particular detail so the following is theory-based speculation. Here goes:

 

The system consists of 4 moderately directional antennas sensitive to airborne intercept radar frequencies. The indicating system is four lights and one speaker. Amplification, time-constant circuit, and threshold filters are in place to limit a nominal directional source to light up only one light for the antenna which is pointed in the general direction of the source for a human-readable duration. Also a tone accompanies the light.

 

In cases where the source is directionally in between the axes of two adjacent antennas it may fall in a sensitivity overlap causing both signals to exceed signalling threshold and both illuminate. The sensitivities of the antennas are adjusted such that the overlap regions are equal in size to non-overlap regions and 8 distinct direction-zones can be identified by illumination of 1-2 lights.

 

Sources which are highly powerful in either peak energy or frequent enough that threshold capacitors are not discharging between peaks will cause light signals to read threshold despite being in the less-sensitive more-off-axis directions. In extreme situations any number of lights may be signalling at once.

 

The entire system may be modeled as 4 antennas with angle-sensitivity response, a temporary charge bank (discharge all if any signal, time constant decay if not), and a threshold for activation. E.g.

 

Signal threshold 5

 

Source power 10 (100% range)

[TABLE]Antenna|angle to source|sensitivity at angle|charge after|signalling

1|30°|60%|6|yes

2|60°|10%|1|no

3|120°|5%|0.5|no

4|150°|1%|0.01|no[/TABLE]Source power 80 (35% range)

[TABLE]Antenna|angle to source|sensitivity at angle|charge after|signalling

1|30°|60%|48|yes

2|60°|10%|8|yes

3|120°|5%|1.6|no

4|150°|1%|0.08|no[/TABLE]Source power 500 (14% range)

[TABLE]Antenna|angle to source|sensitivity at angle|charge after|signalling

1|30°|60%|300|yes

2|60°|10%|40|yes

3|120°|5%|250|yes

4|150°|1%|50|yes[/TABLE]One can imagine that if a time-decaying charge was stored between events one could arrange for a signal on non-threshold reception if it occurred rapidly enough. This would cause long-range lock on behavior to replicate very powerful short range single-sweep response. The sounding channel could simply receive the discharge from the signalling circuits. The more rapidly the signalling circuits fill, threshold, and discharge the more rapid the sounding is heard.

 

That's the simplest design I can imagine which is roughly in line with the behavior of the system people are describing. It could be fairly more complex than the above.

[Toxic]

Super cool html! thank you :)

[Bogey Jammer]

With the unique position of our lead FM programmer being an active duty MiG-21bis pilot; etc…

 

This is extremely interesting. I didn't know there's someone that deeply involved with the MiG-21 at LN. Does this pilot took part in the RWR simulation development yet ? I hope that there is no security reasons to not use his knowledge about that matter…

[Bogey Jammer]

I've done some homework.

 

According to the book Станция предупреждения об облучении СПО-10, the radar signal is captured by the СЗМ-1СМ sensor (СЗМ-7С antenna + СЗМ-8СМ detector) then it is amplified by C3M-9M unit which sends the pulse signal to the C3M-11M unit which is the key element of the SPO-10 logic. It filters the pulse signals and sends almost ready-to-use signal across other parts inside the C3M-3M which delivers sound signal to the aircraft's audio system and light signal to the СЗМ-5А cockpit panel. The C3M-10 is just the power supply.

 

Here comes an interesting part:

4.4. If the irradiating radar operates in the "Scan", the lamp will flash each time the aircraft moving beam irradiation of the station, and a beep will be heard in the form of short beeps in time with the lighting up of the lamp.

4.5. When the irradiating radar in the "Lock" station SPO-10 receives signals in the form of a continuous series of pulses. At the same time light and sound alarm will work without long pauses.

 

This plus further examination of the diagrams, make me think the following points:

1. The 4 channels are processed 100% separately. There is no reason that the 4 eyes should be lit in the same time when the aircraft is locked at 5+km.
   
2. An eye blinks one time when a radar beam passes over the corresponding antenna in scan mode. There is no blinking frequency multiplication in the circuitry when the radar approaches the aircraft. That part in the english MiG-21bis manual is bullshit. So what is shown in my toy is confirmed.
   
3. However, I've not understood yet how the signal is exactly processed when it is in lock mode, as well as audiovisual durations. I still need to translate the chapter 5.
   
4. The audio signal is 800Hz. Spot on in DCS :P
   

 

More to come later…

[MiG21bisFishbedL]

This is extremely interesting. I didn't know there's someone that deeply involved with the MiG-21 at LN. Does this pilot took part in the RWR simulation development yet ? I hope that there is no security reasons to not use his knowledge about that matter…

 

That also assumes that Serbian MiGs (Dolphin887 serves in the Serbian AF, btw) still utilize the SPO-10.

 

With the sort of modifications the type has seen, they may have been given a newer RWR at some point.

[Frederf]

It's not a blinking multiplication it's a sounding multiplication. It says that there will be 1, 2, or 3 sounding tones per illumination.

 

The 4 channels may be separate but if you point a strong enough signal at any "directional" sensor it will receive some signal. The gain vs azimuth graph for the sensors won't be zero even 180° from the intended peak gain azimuth.

[Bogey Jammer]

It says that there will be 1, 2, or 3 sounding tones per illumination.

Where ?

 

The 4 channels may be separate but if you point a strong enough signal at any "directional" sensor it will receive some signal. The gain vs azimuth graph for the sensors won't be zero even 180° from the intended peak gain azimuth.

OK but is a lock on at ANY range powerful enough to make systematically all the eyes to blink ? I don't think so.

[Frederf]

Page 72/228 Handling SPO-10 Radar Illumination Warning System

 

"While the enemy interceptor is closing upon the aircraft, the frequency of the CIlO-10 equipment operation increases from one operating cycle per illumination cycle to three operating cycles per illumination cycle.

 

Once the i l l u m i n a t i n g radar has changed over from the SCAN t o the LOCK-ON mode, •hhft freauenc-r of sound sifcnal s h a r p l y grows and the corresponding l i g h t s s t a r t f l i c k e r i n g ."

 

I don't know the system well enough to say but in principle I can't see why not. They would be sychnonized by the fact that the radar energy is sweeping over all sensors are near-enough simultaneously.

 

Also "When the range between the enemy i n t e r c e p t o r and your a i r c r a f t becomes as small as 5 to 2 km (depending on the i r r a d i a t i o n power of the i l l u m i n a t i n g r a d a r ) , the CnO-10 equipment w i l l operate i n a l l the f o ur channels ( a l l the l i g h t s b u r n i n g ). "

[Bogey Jammer]

That's a funky OCR you've got :D. But it is still not clear that it was about audio signal only: "the frequency of the SPO-10 equipment operation increases"

SPO-10 is the whole system…

 

I don't know the system well enough to say but in principle I can't see why not. They would be sychnonized by the fact that the radar energy is sweeping over all sensors are near-enough simultaneously.

 

Yeah but I forgot to say that all the 4 lights are lit during lock on at every distance in DCS (currently).