Naval missile guidance thread - SAM systems

[Tam]

PESAs may also use solid state signal generators, so no difference there.

That defeats the purpose of a PESA. Solid state amp has voltage breakdown limits. You can only pump that much power to one before it burns out. So in order to have a large amount of power using solid state amps, you would have to take a large amount and bring them to a single cluster. You don't have that problem with analog CRT based amplifiers which is also going to be cheaper too.

I won't do it for a brand new PESA design as it presents no cost advantage to an AESA and might even be more expensive. But if you plan on upgrading an existing PESA to add LPI, that would be possible in paper. In fact you can also upgrade older small and lower power radars to solid state while retaining their mechanical properties, such as rotating antenna.

 

[Tam]

Not quite. Every module has its own HPA, but the signal generator is typically centralized, either for a subarray or an entire array. In this regard it may not be any different than a PESA. I am not saying your example is wrong, rather that there are AESAs that are built differently.

Here is another way to look at it.

 

[nlalyst]

That defeats the purpose of a PESA. Solid state amp has voltage breakdown limits. You can only pump that much power to one before it burns out. So in order to have a large amount of power using solid state amps, you would have to take a large amount and bring them to a single cluster.

No, it doesn't defeat the purpose of PESA. We are talking about the unamplified signal here.

I won't copy pictures here due to copyright restrictions, but Google "TWTAs Still Dominate High-Power and mmWave Applications". You will find a block diagram of a RF signal path that includes a SSPA driver amplifier in the RF path before a high power VED. In fact, I recommend reading the entire article to understand why and where VED still dominate even GaN SSPAs, in particular for 5G and SATCOM applications.

 

[Tam]

No, it doesn't defeat the purpose of PESA. We are talking about the unamplified signal here.

I am talking about putting solid state on PESA. That defeats the purpose of PESA, which is to provide an analog transmitter alternative.

I won't copy pictures here due to copyright restrictions, but Google "TWTAs Still Dominate High-Power and mmWave Applications". You will find a block diagram of a RF signal path that includes a SSPA driver amplifier in the RF path before a high power VED. In fact, I recommend reading the entire article to understand why and where VED still dominate even GaN SSPAs, in particular for 5G and SATCOM applications.

I am not arguing that a VED or CRT based alternatives are or are not viable. They are, especially for cost and reliability reasons, and not to mention that GaNs tend to be nonlinear. I am asking where you got this, where I mark out in italics.

Not in the diagrams that I saw. The HPA was clearly after the phaseshifters in an AESA. There was a preamp before the phaseshifters.

Like Quote Reply

 

[nlalyst]

I am talking about putting solid state on PESA. That defeats the purpose of PESA, which is to provide an analog transmitter alternative.

That's not what you claimed in the post I replied to. But technically, you could make a PESA like that.

You said that SSPA allows for better frequency agility:

Far more important than multibeam is frequency agility. Each AESA module uses a solid state digital amp and signal generator, while a PESA has to use an analog unit, like klystron, amplitron, TWT and so on which are all essentially cathode ray tubes. The result of this is that AESA can change its waveforms much more rapidly and drastically.

Although it is unclear which part in the chain provides the superior frequency agility? Maybe it is the digital solid state signal generator, as you say later:

PESA has implemented its own frequency agility measures but its nowhere like an AESA can do with its digital solid state signal generators.

We just established that PESA can also use solid state drive amps.

But I would like to learn what is a digital amp?

As for the phase shifter placement. Check out:
1. An introduction to
commercial radar sensors, HITECH, slide 13
Technically, a MIMO radar but the same principle applies.

2.Introduction to Radar Systems
Radar Transmitter/Receiver, MIT Lincoln, slide 27-28
There they show both configurations, one with phaseshifters before and the other after the HPA. So you are also right about this.

 

[nlalyst]

I am talking about putting solid state on PESA. That defeats the purpose of PESA, which is to provide an analog transmitter alternative.

You know, the inverse is also possible: to build an AESA radar with TWTAs. VEDs have come a long way in miniaturization and reliability. Although their power density tends to be lower nowadays compared to SSPA.

Back to phaseshifters. If they are placed after the HPA, then the AESA would suffer the same losses as a PESA would. What's the advantage then?

 

[Tam]

That's not what you claimed in the post I replied to. But technically, you could make a PESA like that.

You said that SSPA allows for better frequency agility:

Although it is unclear which part in the chain provides the superior frequency agility? Maybe it is the digital solid state signal generator, as you say later:

Yes, we are talking about SSPA. You can go read this one.

Please, Log in or Register to view URLs content!

Actually this one also shows phase shifter behind HPA in one of the illustrations.

We just established that PESA can also use solid state drive amps.

Technically speaking but no one has done this in real life. unless you can provide an example. But it is interesting.

But I would like to learn what is a digital amp?

Sorry i mixed up. You find those on audio.

As for the phase shifter placement. Check out:
1. An introduction to
commercial radar sensors, HITECH, slide 13
Technically, a MIMO radar but the same principle applies.

You should just provide a link instead of me googling it up. I find it looking at the MIMO. It looks like its putting the time delay behind the amplifier.

Please, Log in or Register to view URLs content!

There are three alternative methods where to put the time delay.

Please, Log in or Register to view URLs content!

Here are for example, phase shifters used on the waveguide.

 

[Tam]

You know, the inverse is also possible: to build an AESA radar with TWTAs. VEDs have come a long way in miniaturization and reliability. Although their power density tends to be lower nowadays compared to SSPA.

They did this before long time ago, with ballistic missile radars. I think one example would be China's LPAR. The problem with VEDs is that they are large, and no one makes them anymore. Maybe some factories in China. Those high end audio amps that use vacuum tubes still source those vacuum tubes from China and they're not even cheap anymore. I don't know if they are still making in Russia or in places like Hungary but I guess these factories have been bought out long ago and have changed their business.

If I were to build a modern LPAR, and I think China did make one, the size of each element which can be about a meter or so, might be big enough to have an entire vacuum tube amp powering one element solely. I would think that ballistic missile radar is the one application where you need the sheer brute power of such an arrangement.

Please, Log in or Register to view URLs content!

Back to phaseshifters. If they are placed after the HPA, then the AESA would suffer the same losses as a PESA would. What's the advantage then?

Much of the losses of the PESA comes from a very long path from the transmitter to the array. If you put TDU before or after the HPA on an AESA module, the path is only difference in centimeters. Power loss and signal distortion takes place after the signal is converted from digital to analog. With AESA, the ADC is on the module itself, it is only analog by a few inches to the antenna, but on the PESA, the long line path from the transmitter to the Tx element is analog.

If you want to reduce these losses, you have to consider changing the medium used for these lines. These lines can also be affected by other forces, such as magnetic forces within the system itself, along with conductivity of the medium. For this reason, these feed lines have to be manufactured with high tolerances, they have to be well protected from interference and such. They add to the cost of the phase array. Probably why PESA ain't really cheaper than AESA.

Maybe you should read about photonic phase array. Here the feed lines are optical. This also requires an optical phase shifter to do the time delays.

Please, Log in or Register to view URLs content!

Of course, there are phase arrays that don't use any feed lines at all. The signal from the transmitter is directly projected to the back of the phase array through a horn feed. This is common with phase arrays used with SAM systems. Signal loss caused between the transmitter and the phase array, can be blamed on the air.

 

[nlalyst]

There is a master clock on every phased array radar, phase shifters , and radiating elements.

Now, the difference between AESA/PESA is in the case of PESA the signal amplification happens before the phase shifter, in the chase of AESA it happens after the phase shifter.

It means the AESA has a good 3db advantage on the transmitter and 3db on the receiver side.

I am curious, where do you get the 3db figure from?

The SPY-1A from the late 70s had a max permissible insertion loss of <1.15db at high and low test frequencies, 0.85dB everywhere else. This was using a 4-bit ferrite phase shifter that cost an arm and a leg, but was impervious to EMP blast (unlike those on MMICs) and not a single one failed in the history of the AEGIS program. Source:

Please, Log in or Register to view URLs content!

Much of the losses of the PESA comes from a very long path from the transmitter to the array. If you put TDU before or after the HPA on an AESA module, the path is only difference in centimeters. Power loss and signal distortion takes place after the signal is converted from digital to analog. With AESA, the ADC is on the module itself, it is only analog by a few inches to the antenna, but on the PESA, the long line path from the transmitter to the Tx element is analog.

I am curious. Can you quantify these losses? Waveguides are known to have lower losses than coaxial cables, and I've seen coaxial cables that have losses as low as 0.03db/m in the lower part of the S-band region and rated for peak powers of 300-425kW.

 

[Anlsvrthng]

No. In both cases, it happens before the phase shifter, but with the AESA, it happens just below the phase shifter, while on the PESA it happens way way back behind the phase shifter, like way way way way back behind the array itself. There is a central amplifiier and signal generator and then you have parallel lines from this unit to each element. Signal transfer from the central amp/waveform generator to each emitter element on the array has insertion losses which AESA only experiences minmally.

So, the difference is the line losses. B

Many PESA, in fact I tnink most PESA have seperate receiver elements. IRBIS is nothing different from them.

Example ?

Nope. In the case of fighter AESA, these can also be rotated.

How ?

This thickness ,mass and cooling requirement is way higher than for PESA.

I already said filtering. For example the computing on the receiving end removes unwanted signals or clutter, so only the relevant signals are sent down to the radar's main computing unit. On the output end, the transmit element would adjust itself based around the output of the neighboring elements.

Again, without having the combines signal from all spatial elements how can you make any filtering ?

The elements has no clue about what is the signal, even the noise per element can be higher than the signal , but the combine spatial signal will be higher than the noise.

Another example, just read the introduction.

Please, Log in or Register to view URLs content!

It is on the receiver path, so irbis does the same, and it require the phase combined signal from multiple element.
NOT calculation on one element.



Phase shifter control is through the SoC implemented through the FPGA, and these are all connected by a bus throughout the entire array.

No. Not with a single central amplifier and signal generator. In a PESA, all the waveforms are generated by a single source, and nothing else modifies it along the way to the emitter. So all the emitters emit the same waveform. With AESA, every single module has their own amplifier and signal generator, and you can vary the frequencies and the timing across the face.

No. The Superheterodyne transmitter needs a central oscillator on the carrier frequency. You can't distribute the oscillators to thousand of places - it will loose the coherence .
They can delivery the IF signal as well separately, theoretically there is no real reason why the IF signal can not generated locally at the amplifier, but it means that the frequency of the DSP and all other circuit has to matched extremely well with the carrier frequency oscillator (hundredth of clock cycle).

Far more important than multibeam is frequency agility. Each AESA module uses a solid state digital amp and signal generator, while a PESA has to use an analog unit, like klystron, amplitron, TWT and so on which are all essentially cathode ray tubes. The result of this is that AESA can change its waveforms much more rapidly and drastically. This allows the AESA to rapidly change its waveforms in response to an ECM response, constantly varying frequency and phase. PESA has implemented its own frequency agility measures but its nowhere like an AESA can do with its digital solid state signal generators. For this reason, AESA or solid state radars are better for implementing LPI, or Low Probability of Intercept. This means the signals being emitted by such radars would have a much more difficult time of being identified by an enemy receiver or ESM. These signals would look like static or background noise, and enemy ESM would be fooled into disregardi

ng them as such, and thus would not cause any alarm or notification.

LPI integrates well with stealth craft, planes or ships, because you essentially have 'stealthy' radar.

Can you explain why a TWT can't be as flexible in waveform and frequency like a solid state transistor ?
I mean, with something that is not an advertising material : )
Actually ,can you explain what can be magically more flexible in a solid state transistor than a vacuum tube or in any other switching elements ?
Of course , using the same complex control electronics : )