Beware when you log sonar files with the HDS GEN2 Touch!

I found out the hard way that it sometimes starts to lag in the position tagging of the sonar pings, after a 30 minutes logging it was 80 meters off! The GPS receiver is not causing this error.

I discovered the error when I did some mosaic images in RM, the error is seen in the same way when you make a structure map so RM isn't doing anything wrong.

My theory is that the live structure map loads the processor too much at certain ranges so it can't keep up with the data. It stores the data alright but the position tagging get delayed BIG time! I had a range of 60 meters set during this logging.

See attached pictures for the error.



This is the correct position



This is the faulty position!


This is the same position using structure map

/hakan

I see three different logs in the pictures.

The conditions for all pictures are unknown. Such problem as in pic 2 are possible, if You not wait for WASAS fix or end of EPE changing.

Bad weather conditions are also possible.

Can You comment this?

Some more info:

The weather was calm.
The two last pictures are from the same log.
I've seen this error in another log and it is a gradually increasing positioning error so the GPS isn't having any problems.
The logging was done in 4 knots of boatspeed.

/hakan

I have somtime crash with humminbird ONIX if I record for RM as I do Autochart live to see what I record in SI (4 channels record, 2 for 2D hight and low Chirp, SI and DI), and screnshot too, some strange things happend not the position but holl in record, so better not to do to much and make a choose!

hakan wrote:
Sounds as though you haven't eliminated the possibility of multipath errors and bad DOP. Do you use an internal or external antenna ? Without having the actual time / date of the problem its not easy to determine the DOP, but a quick look here

www.anglingcharts.com/gpscartographytutorials/gpserrors.html

might give you some ideas about where to start looking without resorting to delving into processor speeds etc. I'd be interested to see the DOP chart for the day and your obstruction diagram. Your latitude isn't overly high, but could easily contribute to some pretty marginal results when satellite geometry is questionable.

T

If you look at the sonar mosaic, comparing the correct one with the bad one you can see that it isn't an offset wich would be the case for a GNSS error, it's a delay between the logging of the sonar strip and the position tagging.

I passed a rectangular object on port and in the correct mosaic it's straight and in the wrong one it's bent since the data is positioned with a delay. I passed the harbour and made a turn to avoid the land.

I've discussed the bug with Matt and also forwarded it to Navico, no reply from them yet.

I was using the internal GNSS receiver in the Lowrance unit and it had been powered on for about an hour or so.

And as I stated before, I have another log with the same bug and there I have some reference objects along the way which shows a gradual increase in the position error during the logging. The bug seems to be triggered when you are logging with a range above a certain threshold, it could be a combination with live structure map which is my theory.

/hakan

I'm afraid that there is not enough information in the images for anyone to be able to make any determination at all. You'll get gradually increasing errors, in any direction, with multipath or DOP errors, in fact a gradually increasing error is more akin to a section of the sinewave error you'd expect. How did the error resolve itself ? Did it just "go away" ?

To get multipath you need to have some big objects close by to give you reflections, that isn't the case here.

You see the same structure all three mosaic images, and on two of them the struture is in the wrong position and badly distorted. How much more do you need?
I can of course put the SL2 files on my server and put the link here if you want do have a look.

The Lowrance unit has given me very good positions of objects on the bottom and on a patch test where I passed an object on the bottom about 15 meters away from both directions the position error was around 1 meter when I checked the mosaic, that's really good!

The position error disappeared when you started a new log, and if you selected a smaller range it worked just fine. I have disabled live structure map and hasn't seen the error since then. But I would prefer to use live structure map to keep track of the coverage.

The purpose of this post is to inform the users on this forum about this problem so they don't loose sonar logs due to this.

I've been working with embedded software development for 20 years and bugs occurs all the time, the trick is to catch them with code reviews and testing. This bug is a bit tricky since it required a long duration to show a noticable position error and required a range above a certain limit in combination with live structure map and maybe even the Navionics chart. You can't test all combinations....

/hakan

You'd wonder if the HDS was using the SD card as working storage ? I imagine you've excluded that possibility. High end SD cards seem to be the order of the day to circumvent a lot of problems with latency in many devices these days.


I dont have the bandwidth to download masses of sl2 mid-ocean I'm afraid, unless you want to chop it and leave only the offending section.

I've seen equally bad swaths for many reasons, thousands of them, and would need rather more information than in your images to work out what was happening. I've also seen huge latency problems caused by devices on the NMEA2k network, as well as slow SD cards, so there are a few things to eliminate before diving into the software.

Thanks for the ideas for trouble shooting!

I have a fast 16Gbyte SD card (Transcend SDHC) which should have no problem handling the bandwidth from the sonar unit. I have used this card fo my GoPro HD cam, Sony HD Cam and Sony Nex6 cam to record HD movies at high data rates without any problems and this data rates are way higher than the data rate from the sonar unit, so that should not be an issue. And it seems like the data rates reaches it's maximum at short range, and I was using longer range here.

I used the internal GNSS receiver in the lowrance unit so no NMEA2000 problems unless they use an internal NMEA 2000 bus, but that one should be bullet proof in that case. I had no external NMEA 2000/0183 modules connected when I did this test.

I do marine geological survey around the Swedish coast for a living now (got tired of sitting at a desk all day long ;) ), my job is to keep all equipment running. We use multibeam, swath sonar, side scan sonar, sub bottom profilers, seismic equipment. For positioning we use GNSS Rtk receivers for centimeter precision (Seapath 330 for the ship and Hemsiphere A325 for our small work boat). I've been trouble shooting a lot of equimpent after three years on the ship so I'm used to things like this. You find a fault, report it, get it fixed by the manufacturer...no big deal :)


I use the Lowrance HDS Gen2 for my personal use and the combo of ReefMaster and HDS Gen2 Touch is really impressing!

/hakan

Hakan,

yes - I thought you may be hopping around the Baltic on a little boat with wires everywhere and tech support just up the road. Here with the proper big gear we have to fix it ourselves ;). I'd be interested to hear the outcome of this, good luck !

T

I guess it's a bit more tricky when you have several days to travel to the closest harbour ;)
Just curious, do you work with sesmic surveys?

I was at the Oceanology International show in London this spring and the 3-D seismic system is really impressive....but way above our budget!!!

We have a small sleeve gun aboard our 40x12m survey ship and a 40 meters analog streamer which is enough for shallow water work. Last year we worked together with the Danish guys and used their digital streamer that was 650 meters long and a bigger air gun and then we got really good penetration (more than 1000 m).

/hakan