problem reading BinLength from 'User Configuration' header

Posted by Dan Kelley at May 24. 2009

If I understand the System Integrator Guide (page 31), bytes 37 and 38 of this hea\

der are meant to contain the bin length.

I am having problems reading this value from the binary file. (I am doing this in\

self-written code in the R language, not using the Nortek-supplied software.)

I have a setup in which the bin length is 0.04m, with 25 bins covering 1 metre. T\

hese values I know from the deployment plan, and from the .hdr file. The problem \

is that when I read these things, I get the value 1674.

I do not think I am reading the file incorrectly, because I am familiar with readi\

ng binary files and because my code reads correctly the quantities before and afte\

r the cell size.

QUESTION: Is this 1674 in some unit that I can convert to 0.04m?

PS. the code is as follows, in case anyone wants to try it in R. (You'll need you\

r own file, of course.)

<pre>

f <- file("sl08AQ01.prf","rb")

hardware <- readBin(f,n=48,size=1,what="raw")

head <- readBin(f,n=224,size=1,what="raw")

user <- readBin(f,n=512,size=1,what="raw")

cat("cell size=", readBin(user[37:38], "integer", n=1, size=2, endian="little"), "\

\n")

</pre>

Re: problem reading BinLength from 'User Configuration' header

Posted by Sven Nylund at June 19. 2009

Hi Dan,

Our apologies for the late reply. Here is the answer:

CS = (1674/256)*0.00675*cos(25) = 0.04 m, for a 2 MHz instrument

For a 1 MHz instrument you must multiply by twice the sampling distance, i.e. 0.0135 m

Best regards,

Sven Nylund

Re: problem reading BinLength from 'User Configuration' header

Posted by Marco Schrijver at June 22. 2009

Previously Sven Nylund wrote:

Hi Dan,

Our apologies for the late reply. Here is the answer:

CS = (1674/256)*0.00675*cos(25) = 0.04 m, for a 2 MHz instrument

For a 1 MHz instrument you must multiply by twice the sampling distance, i.e. 0.0135 m

Best regards,

Sven Nylund

Since I've have the same question (with other numbers) your explanation isn't clear to me:

I've used a setup with a 2 MHz profiler with 35 bins with bin size = 0.10 m. The number I'm reading in the binary file is 1182. In the formula mentioned above I''m not sure what you mean with sampling distance and how this variable can be calculated from the parameters already present in the binary file. Is it possible to give a more detailed explanation?

Best regards,

Marco Schrijver

Re: problem reading BinLength from 'User Configuration' header

Posted by Sven Nylund at June 23. 2009

Hi Marco,

The above explanation is only valid for HR instruments. The regular profilers use a different sampling frequency (or distance).

For the regular Aquadopp Profilers you must substitute 0.00675 with
0.0956 for 2 MHz
0.0478 for 1 MHz
0.0797 for 600 kHz
0.1195 for 400 kHz

Best regards,

Sven Nylund

Re: problem reading BinLength from 'User Configuration' header

Posted by Marco Schrijver at June 23. 2009

Previously Sven Nylund wrote:

Hi Marco,

The above explanation is only valid for HR instruments. The regular profilers use a different sampling frequency (or distance).

For the regular Aquadopp Profilers you must substitute 0.00675 with
0.0956 for 2 MHz
0.0478 for 1 MHz
0.0797 for 600 kHz
0.1195 for 400 kHz

Best regards,

Sven Nylund

Hi Sven,

Since it's not a HR Profiler this is probably not the correct forum, apologies for that. We've setup a 2 MHz Aquadopp profiler, configured with a blanking distance of 0.1m and 35 bins with a bin size of 0.1 m. Current has been measured in shallow water, so with varying depth. This last effect is validated during post processing. When I use the given formula above 1182/256 * 0.00956 * cos(25), the result is: 0.44. Could you tell me what I'm doing wrong? Thanks.

Best regards,

Marco

Re: problem reading BinLength from 'User Configuration' header

Posted by Sven Nylund at June 24. 2009

Hi Marco,

My mistake, here is the corrected table with the value for 2 MHz changed:
0.0239 for 2 MHz
0.0478 for 1 MHz
0.0797 for 600 kHz
0.1195 for 400 kHz

Now you will get the correct cell size, 1182/256 * 0.0239 * cos(25) = 0.10 m

Best regards,

Sven

Re: problem reading BinLength from 'User Configuration' header

Posted by Paul McCarthy at August 11. 2009

Hi Sven,

Can the same formula be used to derive blanking distance?

I suspect not, but is there a generic formula that will work for all Paradopp instruments? My specific need is for AWAC and Continental instruments.

Is it possible for a description of this calculation to be included in the next version of the System Integrator Manual?

Thanks,

Paul McCarthy.

Re: problem reading BinLength from 'User Configuration' header

Posted by Sven Nylund at August 11. 2009

Hi Paul,

For the AWAC the table for cell size looks like this:
0.0478 for 1 MHz
0.0797 for 600 kHz

And for the Continental:
0.0945 for 470 kHz
0.2221 for 190 kHz

To find the blanking distance you must first calculate the cell size as described above using the appropriate value from one of the tables. The blanking distance in meters can then be calculated from the T2 value (at offset 6) like this:
BD = T2*0.0229*cos(25) - CS

The formula for blanking distance applies to all profilers except the HR Profiler where the scaling parameter must be substituted by the same value as for the cell size, i.e. 0.00675 for 2 MHz and 0.0135 for 1 MHz.

Best regards,

Sven

Re: problem reading BinLength from 'User Configuration' header

Posted by Paul McCarthy at August 12. 2009

Hi Sven,

Thanks for your response. I'm happy to use a lookup table as you suggest, but am curious as to the relationship between the head frequency, and the multiplication factor used to calculate the cell size. Is there a generic formula which can be used to automatically derive the factor from the frequency for any instrument? The formula I've come up with (too simplistic, I'm sure) is:

factor = 47.8 / frequency

which holds for all frequencies exceot the continental (190kHz and 470kHz). Are you able to shed any light on this?

Regards,

Paul McCarthy

Re: problem reading BinLength from 'User Configuration' header

Posted by Marco Schrijver at August 23. 2009

Hi Sven,

In a previous post you gave the following numbers:

0.0239 for 2 MHz
0.0478 for 1 MHz
0.0797 for 600 kHz
0.1195 for 400 kHz

In your post to Paul the frequency is 0.0229 for the 2 MHz. Which is the correct one?

Thanks,

Marco

Previously Sven Nylund wrote:

Hi Paul,

For the AWAC the table for cell size looks like this:
0.0478 for 1 MHz
0.0797 for 600 kHz

And for the Continental:
0.0945 for 470 kHz
0.2221 for 190 kHz

To find the blanking distance you must first calculate the cell size as described above using the appropriate value from one of the tables. The blanking distance in meters can then be calculated from the T2 value (at offset 6) like this:
BD = T2*0.0229*cos(25) - CS

The formula for blanking distance applies to all profilers except the HR Profiler where the scaling parameter must be substituted by the same value as for the cell size, i.e. 0.00675 for 2 MHz and 0.0135 for 1 MHz.

Best regards,

Sven

Re: problem reading BinLength from 'User Configuration' header

Posted by Sven Nylund at August 24. 2009

Hi Marco,

The only place I find the number 0.0229 is for calculating the blanking distance (BD), so this is for a different parameter than the cell size (CS).

Best regards,

Sven

Re: problem reading BinLength from 'User Configuration' header

Posted by Reyna Jenkyns at May 21. 2010

For an HR-Profiler, 2Mhz and transducer faces 30 degrees from the horizontal (instead of the usual 25 deg), am I correct to use the following:

The configuration was to have 15mm bins, and I've inserted decimal equivalents from the user configuration (a500....). Can you give any guidance on rounding? I want to then determine the range to each bin.

CS = (628/256)*0.00675*cos(30) ~ 0.01548 m

BD = 19*0.00675*cos(30)-0.015 ~ 0.0960 m

By the way, in the .hdr file it says 20mm for the cell size. Is it rounding to the nearest cm? Similarly, blanking distance in .hdr is 0.10 m...is it rounding to the nearest cm?

Re: problem reading BinLength from 'User Configuration' header

Posted by Reyna Jenkyns at May 21. 2010

I tried copying and pasting....sorry for that mess...ignore all that stuff between angle brackets.

Re: problem reading BinLength from 'User Configuration' header

Posted by Sven Nylund at May 21. 2010

Hi Reyna,

CS = (628/256)*0.00675*cos(30) ~ 0.01548 m

The expression is correct but there seems to be a miscalculation there, the result is 0.01434 m. Since the transducer angle is larger than normal the vertical extent of the cells will be smaller so you end up with a slightly smaller cell size than in the nominal case with 25 degree angles.

BD = 19*0.00675*cos(30)-0.015 ~ 0.0960 m

Since the blanking distance only gives you a fixed offset for all cells this is less important, but to be perfectly accurate the blanking distance should also take into account the actual cell size along the vertical. So with the above calculation of the cell size we get:
BD = 19*0.00675*cos(30)-0.01434 ~ 0.0967 m

I am afraid the current software version rounds off these values to the nearest cm, also for the cell size where it is output in mm. We will try to get that fixed in the next software release.

Best regards,

Sven Nylund

Re: problem reading BinLength from 'User Configuration' header

Posted by Reyna Jenkyns at November 24. 2010

Does the Blanking Distance refer to the distance to the start of the first bin, or the middle of the first bin?

Re: problem reading BinLength from 'User Configuration' header

Posted by Jonas Røstad at November 24. 2010

Dear Reyna,

The blanking distance is the distance from the sensor head to the start of the first measurement cell.

Best regards

Jonas Røstad

Re: problem reading BinLength from 'User Configuration' header

Posted by Reyna Jenkyns at February 08. 2011

Hi, I am having trouble getting the correct values for a few of our Nortek Aquadopp Profilers. The HR-Profilers work fine.

The HEX of the binary characters for cell size is 14 17, which is 5908. For 2 MHz profiler, we use

CS = (5908/256)*0.0956*cos(25) ~ 1.9996 m

But it should be 0.5m.

Even more surprising, I have a completely different Profiler with the same value. I'm using BinLength from the user configuration, but it happens to be the same for these bytes even though they are configured with different bin lengths. They're hdr files are correct. Here's how one of the user configurations start

a5 00 00 01 5c 00 31 00 18 00 d3 01 00 02 17 00 01 00 03 00 00 00 60 00 00 00 00 00 00 00 01 00 02 00 0a 00 14 17 0a 00 00 00 00 00 00 00 ....

What am I missing!?

Re: problem reading BinLength from 'User Configuration' header

Posted by Reyna Jenkyns at February 08. 2011

I found my mistake. I was using the wrong factor (hadn't seen the corrected values further in this thread).

Re: problem reading BinLength from 'User Configuration' header

Posted by Jonas Røstad at February 09. 2011

Perfect!

Please tell us if there are other problems.

Best regards

Jonas Røstad

Re: problem reading BinLength from 'User Configuration' header

Posted by Nick Mortimer at March 28. 2012

I'm trying to check my work I have a AWAC 1 Mhz blanking T2 =44

Now I calculate BD to be 0.4148 now storm says 0.4m same with the cell size 0.4984 vs 0.5 in storm, is this is just storm rounding down?

Cheers

Nick

Re: problem reading BinLength from 'User Configuration' header

Posted by Jonas Røstad at April 11. 2012

Dear Nick,

Yes, this should be rounding issues.

Best regards

Jonas Røstad

Sections

problem reading BinLength from 'User Configuration' header