Inclining test shifting weights

Dear all,

I am arguing with a consultant about the type of shifting weights to be used during an inclining test. I am talking about a small to middle size vessel (60 m).

As per IMO Intact stability code, annex 1, paragraph 4.3, it is quite clear that 8 different heeling moments have to be measured and that, after a linear regression, a straight line is to be fitted on the heeling moment vs. tangent of heel graph.

I suggested, in order to measure 8 different heeling moments, to use 4 different weights and a constant shifting distance. However, the consultant argued that in order to get a straight line on the graph, he doesn't accept to use different shifting weights.

After consultation of several others inclining test procedures around me, I found that everybody is using the same weights and the same distance and therefore is recording only two heeling moments (shift to starboard side and shift to portside). So now, I am quite puzzled.

Is there anybody who can advise me on this issue?

Thanks a lot in advance.

VM

 

The displacement during the inclining test shall not change so no weights can be embarked or disembarked during the test (and the centre of the weights should also not change either vertically or longitudinally during the test).

You can however move either partially or all of the weights in a constant or variable distance, noting that you should be able to replicate each reading in order to verify it.

I've done it both ways and if well executed either method is acceptable the choice depending on the particular arrangements on the particular vessel (e.g. the existance of a deckhouse may prevent from using different shifting distances).

Rodrigo

 

The requirement is for eight different "moments"; not points on each side of the line.
A common sequence with 2 weights is:

#0 1 CL : 2 CL
#1 1 P : 2 CL
#2 1 P : 2 P
#3 1 CL : 2 P
#4 1 CL : 2 CL
#5 1 S : 2 CL
#6 1 S : 2 S
#7 1 CL : 2 S
#8 1 CL : 2 CL

If everything is fine, #'s 0, 4, & 8 should all fall on the same angle, otherwise the vessel lolls. Additionaly, if the weights are equal (which they will never truely be), #'s 1 & 3 and #'s 5 & 7 should match. You just need 3 points on each side of center, and 3 points on center (i.e. 8 moment points+the starting point) to check that the line is straight and the vessel is not lolling.

 

Hi Rodrigo,

Well understood your message. However, I think there have been some misunderstanding regarding my question. My bad.

I know that no weight change is allowed during the experiment neither change of actual LCG and VCG. Maybe I should take an example.

Let's give an available transverse distance of 10m so 5m away from centerline on both side. Let's imagine we need a heeling moment of 150 ton.m to achieve the minimum requirement of 1 degree of heel or 150mm measure on the measuring devices. This leads to a minimum weight to be shifted of 150 / 10 = 15 tons.

If I put 4 weights of 15 tons on the main deck and shift them, according to standard shift procedures, alternately of 10m, I will only get two different heeling moments: +150 ton.m and -150 ton.m. Therefore, we don't get the 8 moments required to plot the chart.

Let's say, we change the weights to be: W1 = 15 tons, W2 = 16 tons, W4 = 17 tons and W4 = 18 tons, then we shall record 8 distinct moments: +/- 150, 160, 170 & 180 tons.m. Then, we shall be able to obtain the proper chart.

I hope it is clearer now, sorry about that.

I see you are located in Singapore which means close to me (I am in Batam, just in front of you). I have checked a few inclining test procedures made by others shipyard in Batam and my consultant comes from Singapore as well: all of them use 4 similar shifting weights and shift them of the same distance, which means that they apply the same heeling moment at each shift (only change the sign of the moment, depending if they shift from portside or from starboard side). I am wandering if it comes from an usual way to perform inclining test in this region or not.

Hope you can help.

Regards,

VM

 

@ jehardiman:

Thanks for the answer. Alright, then I shall show you what has been proposed and ask for your opinion.


Here is the consultant proposal:

http://imageshack.us/photo/my-images/513/consultant.png/

Here is my proposal (amendment in red):

http://imageshack.us/photo/my-images/191/neweyz.png/

I have the feeling the new proposal is more willing to be approved by flag administration, isn't it?

 

Inclining Test

Vinassman,

Have a look at the attached pdf. It is part of a document issued by the IACS (International Association of Classification Societies) and may help you decide.

Cheers,
Graham
Naval Architect

www.westbrookmarine.co.uk

 

Thanks Graham for the reference.

I found in IMO intact stability code that they advise to get 8 points on the curves and therefore we shall measure 8 distinct moments. So we do have to either change the distance or the weights between the shifting weights.

Now is my question to all of you: how do you do yourself? Do you use 4 different value of weights or 4 similar?

Thanks,

VM

 

If you change the weights, the overall displacement, CG, etc. are changing. The rule makes it really simple: shift the weights. Why do you want to change the method? Usually you can do it differently and then justify your system. However, if there is any failure, you will have to justify that in court which can be difficult and expensive.

 

It depends more on the weight availability and crane capacity than anything else. If you use the IACS method Graham posted, it states the weights should be approximately equal. This is to give sufficient seperation of the points and a check point, likewise in my example weight 1 should be twice as heavy as weight 2 if you want 4 points.

Note that the total moment shift must produce at least 1 degree of heel but no more than 4 degrees. This should not be a problem for you, but it really is for larger ships. We needed over 400 tons to incline the USS Iowa in 1984.

Edit, In the consultants proposal the column labled "Moment" is the moment shift, not the total moment. The total moment would be 0, 99, 198, 99, 0, -99, -198, -99, 0; i.e. 3 points (0,99,198) are all that is needed to make a straight line. In yours the total moment starts at 44 t-m to stbd, and never gets to 0 so it will not check for loll. If you swap weights 2 and 4 so the original moment starts at 0 it would be better and would give an extra point between 0 and maximum.

 

Use 4 equal weights and hope to get 5 different readings.
Use 6 weights and get seven different.
Chances of getting exact weights equal at that size is slim, even so it's really hard to get the same moments again if you replace the weights a second time round.

 

@ jehardiman:

Actually, I figured out the same indeed. The difference is that I worked with shifted moment and not total moment.

However, after I checked the GL Leaflet for Inclining experiment, they do have curves with 3 points on each side. I believe a good way to achieve this is to use let's say 2 weights of 9 tons and 2 weights of 18 tons in order to get 3 different total moments on each side and still to be able to come back to 0.

By the way, what will be the "profile" of the graph in case of loll?

Anyway, thanks for the advice and the answers.

@ gonzo: sorry but I guess I cannot make myself clear... I was just saying that I had in mind to use 4 shifting weights with each of them a different weight. However, during the operation, the displacement will not change during the test. Hope it is more clear.

VM

 

Usually the need for use more weights is related to the facility of moving them around in case the weight needed is large (2x10t weights are easier to shift than 1x20t weight).

I've been present in inclining tests, for a small fishing vessel, were only one weight was used (being shifted by different distances).

As long as you can calculate the heeling moment and corresponding heeling angle you should have no problems.

You can consult "ASTM F 1321 – 92 - Standard Guide for Conducting a Stability Test (Lightweight Survey and Inclining Experiment) to Determine the Light Ship Displacement and Centers of Gravity of a Vessel" which is quite an extensive guide on which problems related to the practical execution of the test are also dealth with.

Rodrigo

 

Vinassman,

I see you have accepted that you don't need 8 distinct moments.
Why not use a six equal weights, then you will get nice proportional movements each side of zero. I feel that this better demonstrates the change in Tan on the linear scale of incline. Still this is not necessary, you should follow the advice of your consultant rather than stepping up for such responsibility.
There is unlikely to be a graph of loll as the angle of loll will most likely exceed 4 degrees. Luckily there certainly wont be any stability booklet either for such a craft.

 

In the case of loll (i.e GM<0 for some small angle, noteably due to free surface in vessels with flat floors, so make sure you completely drain or press up all tanks) then the 3 zero moments ( #0, #4 and #8) do not fall on top of each other. Additionaly the half moments ( #1&#3 and #5& #7) may not fall on the same line depending on the extent of GM<0.