|
|
Thread Tools | Display Modes |
#81
|
|||
|
|||
Ranging and Pioneer
"Oh No" wrote in message ... Thus spake Spud ... gr-qc/0104064 Many thanks. If I understand Anderson correctly it is merely an engineering constraint. With a (perhaps greatly) amplified signal it should be possible to reduce integration times to achieve correlation so that the signal can be returned without a substantial range delay. The problem was more fundamental. Reading between the lines, and from some personal correspondence, the rate of change of the carrier frequency was I believe limited in choice by the design of the exciter or correlator equipment. The narrow bandwidth required to get an acceptable SNR at the craft meant the minimum sweep rate was too fast and the craft lost lock on the uplink. Increased gain at the craft wouldn't help, only more transmit power, but that was in the 100's of kW already and being routed through their largest (70m) dishes. Anyway that is the basis on which I am working at the moment. But I am not an engineer, and I was hoping this might be confirmed. My arguments would take quite a different form if this was a fundamental constraint. It was purely an equipment limitation, possibly exacerbated by radiation damage during the Jupiter flyby though that is my speculation. BTW, I am a digital and systems engineer in a company working significantly in HF comms. George |
#82
|
|||
|
|||
Ranging and Pioneer
test sending via ISP
"John (Liberty) Bell" wrote in message ups.com... wrote: John (Liberty) Bell wrote: wrote: Finally you have perhaps the biggest problem of measuring the switch-off of the signal with millisecond accuracy through a receiver chain with a bandwidth of less than 1 Hz which is what was needed to hold lock on the carrier. That applies whichever method of timing you use at the transmitter. Notwithstanding my prior comment on this point, it is obvious that many positions in this receiver chain have much higher frequency responses than 1 Hz. True but the siganl is far les than the noise prior to the filter so not accessible. More below. You are confusing me here. We seem to be talking about several filters simultaneously. There are several filters involved and you seemed to be discussing tapping off a signal from earlier in the chain where the bandwidth was wider. Furthermore, in a feedback loop, it is no longer meaningful to discuss earlier and later, only physical points in the loop. Again you were discussing "positions in this receiver chain [which] have much higher frequency responses". Those are before the phase-locked loop where the narrowband filter resides. Before taking your points separately, the manuals can be found he http://deepspace.jpl.nasa.gov/dsndocs/810-005/ The index lists them by number on the various pages, the 100 series numbers are on the "Space Link Interfaces" page and the 200 series on "Station Data Processing". Clearly, at one point we must have a high Q bandpass filter which is tuned to the carrier frequency of ~ 2 gig. Its frequency response is, presumably, the carrier frequency + - 1 Hz It isn't that simple. The dish has a low-noise amplifier (LNA) in the pedestal room which amplifies the whole S-band range from 2270 MHz to 2300 MHz. A structural diagram is shown in manual 203 figure 15 on page 34 and functional block diagrams are given in manual 101, pages 26/27. The "D/C" in manual 203 figure 15 is a 'downconvertor' which is further detailed in the manual 209. Read para 2.3 on page 7 and look at Figure 2 on page 10. The amplified signal from the LNA is heterodyned to an IF frequency of 300MHz, then it is filtered with a bandwidth from 265MHz to 375MHz (centred at 320 MHz) and that is then heterodyned to a centre of 64MHz. That signal is then digitised at 8 bit resolution and a sample rate of 256 M-samples/s and fed to the digital down-convertor (DDC) which extracts subchannels of 16MHz width. Clearly at another point we must have a low pass filter providing a control signal. Its effective frequency response is, presumably, 0 to 1Hz I would be happy to stand corrected on this, but I don't think these two 1Hz figures are the same thing. This very narrow band filtering is done in the digital domain and is described in manual 207. Start with section 4 on page 11, mainly the first two paragraphs. You can probably skip para 4.1 though the first three paras on page 13 add useful background to the process for determining when the system is "in lock". Go on to para 4.2 and read the first two paragraphs, and note the graph of Figure 4 on page 19. It is this loop which can go down to 0.01 Hz but note the first para of 4.2 says "Once phase-lock is declared at this bandwidth, the loop bandwidth is gradually reduced to its final value - the desired operational value - in a manner consistent with the retention of phase-lock." The trouble is I don't know what the 'desired operational value' was at any time. In the latter days at extreme range I am sure it was 0.01Hz but in previous years they might have selected a wider value. Anyway that means the filter would ring with a decay time constant of 16s, 1/(2 pi bandwidth). As you said above, there were earlier stages with up to 30MHz analogue bandwidth or 16MHz digitized but these would contain vastly more background noise than signal. George |
#83
|
|||
|
|||
Ranging and Pioneer
"John (Liberty) Bell" wrote in message
oups.com... Craig Markwardt wrote: ..... I refer you to my communications with Jonathan Silverlight, and to gr-qc/0104064 for confirmation that spacecraft transmissions were indeed "switched off." repeatedly (and switched on again successfully). I quote you, personally (from the reference provided by Jonathan Silverlight): "Turyshev admitted two things: the round trip signal time is recorded in some form, but it is not precise enough to constrain the anomaly. Second, he said they did not use that form of measurement technique anyway. As I've already gone into in a different post, I wanted to investigate this more deeply, so I went to a primary document, the DSN procedures manual. I already referred to the table in that manual, where they describe that it takes a certain amount of time to acquire a signal lock, somewhere in the range 0-4 seconds, but perhaps more time, depending on the pre-acquisition bandwidth. Thus, the signal receive time cannot be known to a precision better than a few seconds. Hence, satellite distance discrepancies of 0.5 light seconds or smaller would not be measurable with such a system." Your first paragraph appears to confirm that means to control spacecraft transmissions from Earth do, in fact, exist. Your second paragraph appears to confirm that the lack of accuracy in the already recorded light time data is due to the time taken to achieve a signal lock. The transmitter can be switched off and on but after switching on it can take a long time to lock on. At extreme range a narrow receiver bandwidth has to be used on the craft and the uplink signal is swept through the range of frequency where the craft might be listening hoping it will lock. The time taken depends on the particular frequency hence on the absolute accuracy of the on-board reference. This would seem to suggest that, once a signal lock has been achieved, the primary obstruction to obtaining more accurate ranging data has already been overcome. Yes, a better approach is to consider the switch off transition. Once the uplink and downlink have been locked, there should be a clear indication if the downlink is switched off. One problem is that the downlink keeps lock by using a very narrow bandwidth, probably less than 1 Hz so the response time may be more than a second. Another problem is how it is achieved. For normal operational purposes, they probably had software that would switch the transmitter off and on at specified times to match the ground station schedules. Using that requires an accurate clock on the craft and synchronisation becomes significant. There may not be a command to do an immediate switch off, but again if there were then software response times would need to be deterministic and known. Since the links use forward error correction and possibly repeated commands, it may not be certain which copy of the command caused the transmitter to go off. Whilst it is true that signal levels are, by now, already below the noise threshold, that problem too can potentially be overcome, via the development of still lower noise detectors. Not necessarily, the problem can be that the signal falls below the galactic background. The only way to extract it then is to narrow the bandwidth so that all the signal power is seen but less wideband noise is allowed through. Of course reducing the bandwidth has the problem of increasing the response time and hence measurement uncertainty. George |
#84
|
|||
|
|||
Ranging and Pioneer
George Dishman wrote: "Oh No" wrote in message ... The position of Pioneer was calculated from Doppler information. Ranging was not available. Can anyone explain why ranging could not be used? Is this just a limit on available technology, or is there a more fundamental reason? To get a range measure, the DSN applies a "ramp" to the uplink. I believe that means very slow sawtooth FM. The time delay to the corresponding downlink modulation via the transponder gives the range. Because of the extreme distance and low signal levels, the bandwidth had to be very narrow both at the craft and the DSN. From personal emails from one of the team, I believe that they did attempt to apply the modulation soon after the Jupiter flyby but it always caused the spacecraft to loose lock with the uplink. I would just like to add to this that one of the earlier refs, or a link therefrom, does indicate that the cause was a failure within the spacecraft, quite possibly radiation damage. However, it does seem that both Pioneer 10 and 11 suffered the same fate in this respect, which is suggestive of a design flaw. Ranging on other missions worked fine. John Bell. |
#85
|
|||
|
|||
Ranging and Pioneer
Thus spake George Dishman
"Oh No" wrote in message k... Thus spake Spud ... gr-qc/0104064 Many thanks. If I understand Anderson correctly it is merely an engineering constraint. With a (perhaps greatly) amplified signal it should be possible to reduce integration times to achieve correlation so that the signal can be returned without a substantial range delay. The problem was more fundamental. Reading between the lines, and from some personal correspondence, the rate of change of the carrier frequency was I believe limited in choice by the design of the exciter or correlator equipment. The narrow bandwidth required to get an acceptable SNR at the craft meant the minimum sweep rate was too fast and the craft lost lock on the uplink. Increased gain at the craft wouldn't help, only more transmit power, but that was in the 100's of kW already and being routed through their largest (70m) dishes. Considering that on Earth we pick up a signal of 8kW from Pioneer, it doesn't sound to me as though the amplitude of the signal sent to Pioneer was really too low. Loss of lock on frequency modulation sounds interesting. I might be helpful to understand this more precisely. Can you give more detail on how the signal is encoded. Anyway that is the basis on which I am working at the moment. But I am not an engineer, and I was hoping this might be confirmed. My arguments would take quite a different form if this was a fundamental constraint. It was purely an equipment limitation, possibly exacerbated by radiation damage during the Jupiter flyby though that is my speculation. I am slightly sceptical about the equipment failure theory, because it would mean the equipment from both pioneers failed in the same way at much the same place. I would also be a little surprised if radiation from Jupiter is energetic enough to cause such damage. After all other space craft have flown close to the sun and survived. Equipment limitation seems probable. It would be more satisfying if one could identify a cause for the limitation, since clearly it was not expected from the design parameters. Your response has caused me to wonder if there is an unmodeled effect in the uplink which might contribute to loss of lock. Clearly there is a Doppler variation in the signal received by Pioneer due to orbital motion of the Earth of 30,000m/s which must have been taken into account. But how narrow is the bandwidth, and how accurately does this need to be predicted? Could an unmodelled effect as small as _+-2m/s (varying annually) contribute to loss of lock? BTW, I am a digital and systems engineer in a company working significantly in HF comms. It's clear you know a lot about this kind of thing, and your input is much appreciated. Regards -- Charles Francis substitute charles for NotI to email |
#86
|
|||
|
|||
Ranging and Pioneer
George Dishman wrote:
Not necessarily, the problem can be that the signal falls below the galactic background. The only way to extract it then is to narrow the bandwidth so that all the signal power is seen but less wideband noise is allowed through. Of course reducing the bandwidth has the problem of increasing the response time and hence measurement uncertainty. This response is remarkably similar to your original response of Fri, Aug 4 2006 8:26 pm . Are we starting to go round in ever diminishing circles here? [Mod. note: a number of postings that George made earlier this year and that were lost in his ISP's system have now reappeared and a few were approved by your inattentive moderator before he noticed. This, I suspect, is one of them. Apologies for any confusion caused -- mjh] |
#87
|
|||
|
|||
Ranging and Pioneer
John (Liberty) Bell wrote:
George Dishman wrote: Not necessarily, the problem can be that the signal falls below the galactic background. The only way to extract it then is to narrow the bandwidth so that all the signal power is seen but less wideband noise is allowed through. Of course reducing the bandwidth has the problem of increasing the response time and hence measurement uncertainty. This response is remarkably similar to your original response of Fri, Aug 4 2006 8:26 pm . ... I probably sent it on Aug 4th, see the mod's note. There are a few others like that too. Just ignore them or reply if you choose. [Mod. note: a number of postings that George made earlier this year and that were lost in his ISP's system have now reappeared and a few were approved by your inattentive moderator before he noticed. This, I suspect, is one of them. Apologies for any confusion caused -- mjh] |
#88
|
|||
|
|||
Ranging and Pioneer
Oh No wrote:
Thus spake George Dishman "Oh No" wrote in message k... Thus spake Spud ... gr-qc/0104064 Many thanks. If I understand Anderson correctly it is merely an engineering constraint. With a (perhaps greatly) amplified signal it should be possible to reduce integration times to achieve correlation so that the signal can be returned without a substantial range delay. The problem was more fundamental. Reading between the lines, and from some personal correspondence, the rate of change of the carrier frequency was I believe limited in choice by the design of the exciter or correlator equipment. The narrow bandwidth required to get an acceptable SNR at the craft meant the minimum sweep rate was too fast and the craft lost lock on the uplink. Increased gain at the craft wouldn't help, only more transmit power, but that was in the 100's of kW already and being routed through their largest (70m) dishes. Considering that on Earth we pick up a signal of 8kW from Pioneer, That is 8 W, not 8 kW. The anomaly is equivalent to the radiation pressure of a signal of just 63 W, a light bulb would do it. it doesn't sound to me as though the amplitude of the signal sent to Pioneer was really too low. Loss of lock on frequency modulation sounds interesting. I might be helpful to understand this more precisely. Can you give more detail on how the signal is encoded. AIUI they had a dedicated piece of equipment that they used. It provided an output which was used to modulate the uplink carrier and it also then looked at the downlink frequency and tried to run a correlation. Once it found a match, it would measure the time delay. I don't know the exact modulation scheme but I believe it involved "ramping" the transmitter frequency, possibly by generating commands to the local signal generator rather than producing a frequency itself. The ATDF format includes "ramp" records and that facility was used as part of the signal acquisition strategy. My guess it that linear frequency ramps were used possibly with variable ramp durations to resolve any ambiguity. Anyway that is the basis on which I am working at the moment. But I am not an engineer, and I was hoping this might be confirmed. My arguments would take quite a different form if this was a fundamental constraint. It was purely an equipment limitation, possibly exacerbated by radiation damage during the Jupiter flyby though that is my speculation. I am slightly sceptical about the equipment failure theory, because it would mean the equipment from both pioneers failed in the same way at much the same place. I would also be a little surprised if radiation from Jupiter is energetic enough to cause such damage. After all other space craft have flown close to the sun and survived. Jupiter is worse because the craft get much closer and at the time they were shocked at the levels. Remember the craft were some of the first to encounter these fields and subsequent craft have had much better protection as a result. Equipment limitation seems probable. It would be more satisfying if one could identify a cause for the limitation, since clearly it was not expected from the design parameters. Your response has caused me to wonder if there is an unmodeled effect in the uplink which might contribute to loss of lock. Clearly there is a Doppler variation in the signal received by Pioneer due to orbital motion of the Earth of 30,000m/s which must have been taken into account. The effect from the rotation of the Earth is more significant due to the shorter period. But how narrow is the bandwidth, and how accurately does this need to be predicted? Could an unmodelled effect as small as _+-2m/s (varying annually) contribute to loss of lock? For the initial search I know the DSN uses FFTs over a fairly wide rang - 16kHz from memory with a check in adjacent bands. The diurnal Doppler is around 8 kHz (and is known of course) but produces a rate of change in the kHz per hour range. The FFT then told a PLL where to look with an initial bandwidth in the low Hz which then dynamically reduced to as low as 0.01Hz as lock was obtained. However, the craft hardware was probably much simpler. I'm sorry I can't help more on that but I don't have details on the craft receiver design. Against that, the uplink power was 400 kW compared to just 8 W on the downlink so I would have expected a much stronger lock. Anyway the bottom line is that I expect it was the rate of change of frequency that would break the uplink lock. It's a shame they couldn't hack the correlator box to run a smaller frequency deviation but since it was probably dedicated hardware in those days, it might not have been practical. BTW, I am a digital and systems engineer in a company working significantly in HF comms. It's clear you know a lot about this kind of thing, and your input is much appreciated. Thanks, I don't pretend to know everything but I hope it is helpful. George |
Thread Tools | |
Display Modes | |
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
30 Years of Pioneer Spacecraft Data Rescued: The Planetary Society Enables Study of the Mysterious Pioneer Anomaly | [email protected] | News | 0 | June 6th 06 05:35 PM |
Pioneer 10 test of light speed | [email protected] | Astronomy Misc | 48 | February 18th 05 05:40 AM |
pioneer 10 acceleration | Nodem Info. Sys. | Research | 19 | June 4th 04 10:15 AM |
NASA Test of Light Speed Extrapolation | ralph sansbury | Astronomy Misc | 26 | February 12th 04 03:29 PM |