SBP raw data pixel size

[Finn]

Hi Guys

I have understud that the accuracy of the SBP data comes directly from GPS. So that with standard L1 GPS the accuracy is about 45 feet.

Should i inteprete this so that with standard GPS pixel size is 45 by 45 by units vertical resolution?

Not any kind of software would help if the raw data is so crappy from the first place.

It's worse than NASA: s data from the surface of Pluto!!

Correct me if i am wrong please.

Salud

Finn

 

Amazon Forum Fav 👍

Cry Of The Eagle: History And Legends Of The Cherokee Indians And Their Buried Treasure - Grab it through Amazon!

[Finn]

I want to add some details. Every "ping" that sub bottom profiler sends, and receives of course, is tagged with position data. Now when the position data is not accurate, these "ping"

points are placed in anywhere in GPS accuracy radius, which can be like 25 feet. If the system receives good heading data (COG) from boats gyro compass and exact speed (SOG) from

good speed log, so those may help A LITTLE, but only when the survey data is processed with some software.

Now if there is for example a 8 foot iron canon in a soft mud (easy matrix) and SBP passes over it so what will we see on the screen? Maybe nothing because of the roll and pitch and

accelerations of the boat and the towfish. It,s like trying to draw a human hair on paper with 1 inch by 1 inch felt tip pen.

I am trying here to convince myself that SBP is not a good primary survey tool when looking for a shipwreck or other man made object. I am not reading this from any book or wikipedia

but trying to figure it out based on facts that are available. If you have better info, please comment.

Salud

Finn

 

[AUVnav]

Accuracy does not come from the GPS, it comes from the unit itself, some going down to cm range on resolution. The surface vessel GPS unit should be WAAS or DGPS, this is very accurate.

The SBP readings are geo-referenced by the shipboard unit. The tow speed should be slow enough so that the variations are accounted for, and the onboard unit kalman filter will provide the location anyways, ie where you were, not where you are, and will adjust.

Now, about the unit, what freq/pulse is it running at? As an example, the Edgetech unit is of course, a chirp signal from 4-24kHz, with pulse length of 10ms. You will get 10 measurements per second. As this is sound, the depth is a function of the speed of sound, which is why the 10ms chirp depth is 40m, and the 20ms chirp depth 80m. This is another variable to review when choosing a system, or understanding what the varibles mean.

Using a downrigger is far more accurate way to locate where the unit is below the surface, and allows to fix a very accurate depth. This also allows for much more roll/pitch of the surface vessel, without affecting the towfish.

It is really a perfect tool for searching. Why limit yourself to what is showing on the seafloor, when you can see what is buried?

 

[Finn]

I was involved in a shallow water (10-65 feet) survey at last year where over side pole mounted Knudsen Pinger(CHIRP) with adjustable pulse length, pulse delay and gains was used on 27 feet RIB equipped with track following autopilot.Aimed towing speed was 3 knots. That usually varied from 2,5 to 3,5 knots. We were looking for ballast piles, canons and parts of wooden ship structure. Bottom was loose, shifting sand and gravel sediment layer varying from 0 to 30 feet over the bedrock. We tried with 15 kHz and got no penetration. 3,5 kHz got us penetration but vertical resolution with such low frequency is only about one feet.

We past several times over 20 inch pipeline, buried down to 6 feet in bottom, which goes out from refinery. Few times we got this parabolic shape reflection on the screen when we past the pipeline but it was not every time. It would have been impossible to say if there was something or not if we didn't knew that pipeline.

WAAS is not available here because there are no land stations, DGPS beacons either. And anyway those would give base accuracy of ten feet which is not so good either.

The system needs a coordinate grid and NAV data for measurements. The only source for those is support vessels NAV instruments and the GPS receiver. It the input data is not accurate so how could the output be.

Maybe in a river delta or at lake where there are no swell and shifting currents and the sediment is relatively soft and even and an accurate position data is available.

Salud

Finn

 

[AUVnav]

okay, looking at what you have stated, and a brief look at the online information.

Something is amiss, ..this is a chirp, which means a multi-freq pulse, not a ping. Looking at the Knudsen data, it appears they have 2 channels, a low and a high.

There is CW and Chirp. Cw is continuous freq. I am not seeing a chirp in the knudsen data. (even though they call it one) 2 freqs is not chirp. Chirp is multifreq on one channel.

You mention one foot penetration, what was pulse length? 3.5kts seems too fast for that spec, did they give a max tow speed?

With mixed sand and gravel, the penetration will be minimal, so tow speed should be very slow. Pulse length and pulse delay should be minimal, again these determine depth, so you simply need to be stationary for a time to adjust to your optimum resolution.

You say it was shallow water, from 10-65 feet. The water depth affects the pulse rate, and depending on the density, this needs to be validated real time.

The specs on the unit beamwidth varies wildly on frequency, and in reality, does not match with the rest of the specs. As shown in the specs, a 30 degree beamwidth at 6 kHz, while the specs show 3.5 kHz, and a 12 degree beamwidth at 15 kHz, which matches the other spec. This seems to put the beamwidth at 3.5kHz at over 45 degrees. There is a reason they do not advertise this in the specs. the sidelobes at 45 degrees illumination make the data very corrupted. They show 30 degrees at 6, 12 degrees at 15, and 9 degrees at 200. As you can see, they advertise 30 degrees at 6 kHz, this is the lowest I would go...even that is pushing the data.

Sorry the above is a bit disconnected, but in looking at the specs, and what your are saying, the specs are really disconnected. I am very suspect when the transmits are 3 separate systems, and receivers are listed as two different systems, a hydrophone and a separate 200kHz...

I really suspect this is a multi freq pinger, and not a chirp. Chirp will give you the doppler shift in the signal, with horizontal and vertical location to transmit/rec. A pinger will give you range to the transmit/rec. There is a significant difference.

I think they are using 2 frequencies, and combining the results. ie, measuring the response at 3.5kHz and 200 kHz, and subtracting the difference, the resultant being a quasi 2d replica of the target. Think of it as a combination of 2 dots per pulse vs a swath of 2000 dots per pulse....

hope that helps...

( I like the response numbers...especially 176dB, )

 

[Finn]

Thanks for your comments AUVnav

Three knots is the minimum steering speed for autopilot in local conditions. SBP manufacturer has not mentioned any maximum speed but as far as i understand it should be slow and

steady. Another issue, i think, is the pole mounting of the transducer, it repeats every movement of the boat; 30 degree inclinations to all directions, up and down movements caused by

swell and wind waves.

One foot that i mentioned is not penetration depth but vertical resolution with 3,5 kHz, like in technical data of Edgetech 24 kHz is mentioned: "vertical resolution 4-6 cm", if i remember

correctly. So if we think raw data as 3D pixels, the depth of the pixel is one foot with 3,5 kHz.

I didn't think about the beam width the way you described, thanks. I was just calculating that when the boat inclines 30 degrees, where does the beam shoot at certain depth.

What you stated of they using two frequencies is probably correct. The 200 kHz projector is fixed in the transducer and cant be chanced, the user can swap in 3,5 kHz or 15 kHz projector.

Ringing and secondary echoes limited useful gain ranges a lot.

Maybe with other kind of setup we would have got better results, but this experience didn't make me a SBP believer.

Salud

Finn

 

[AUVnav]

Finn,

Seems like you are experiencing issues associated with a pinger vs a chirp. Since the chirp signal is multi-freq, the different freq are just a little off from each other in response time. This helps to eliminate the ringing (I assume feedback) and secondary echoes. Much of this is caused by the signal length, as it is still broadcasting when an echo is received, so the matched filter just blows up. More power just makes it worse.
I would consider using the same techniques we had to use with the older metal detectors, tuning it to the local conditions with freq and pulse length, but alas, with a 2 freq pinger, it is pretty limited.
There are gimball mounts that isolate the transducer from boat movement that may help.

As a primary search tool, I would agree that the use is limited. As a secondary, when looking at the surrounds and the site, it cant be beat..

Cheers!