1
00:00:10,838 --> 00:00:15,779
Hello, in this five-minute video we
are going to see how Heliciel checks,

2
00:00:15,779 --> 00:00:19,028
in CFD, a boat propeller
drawn in its designer:

3
00:00:21,187 --> 00:00:27,649
its real thrust, the pressure on its
blades, and the risk of cavitation.

4
00:00:29,583 --> 00:00:37,192
Here is the marine propeller supplied
as an example with Heliciel;

5
00:00:37,192 --> 00:00:41,458
it is 1,140 millimetres in diameter,

6
00:00:41,458 --> 00:00:50,219
has 7 blades, and turns at 105 rpm to
push a boat at 1.6 metres per second.

7
00:00:50,269 --> 00:00:54,238
In sea water, the designer
calculates by the

8
00:00:54,238 --> 00:00:58,568
blade element method a
thrust of 1,156 newtons.

9
00:00:58,618 --> 00:01:02,174
This is the figure we are
going to put to the test.

10
00:01:04,324 --> 00:01:09,264
Let's now open the PRO
CFD menu of the designer.

11
00:01:14,651 --> 00:01:18,704
It sends the propeller to the test
bench exactly as it is drawn.

12
00:01:18,754 --> 00:01:21,682
We have nothing to describe: the fluid,

13
00:01:21,682 --> 00:01:25,886
the speed and the rotation
speed come from the project.

14
00:01:28,063 --> 00:01:33,497
In the bench, the only decision
to make is the mesh preset.

15
00:01:33,547 --> 00:01:38,102
From Level 1, for an express
test, up to the finest levels,

16
00:01:38,102 --> 00:01:44,202
Heliciel announces for each one the number
of cells and the expected duration.

17
00:01:44,252 --> 00:01:50,712
Here, Level 5 plans about 145,000 cells
and 7 minutes of meshing.

18
00:01:50,789 --> 00:01:54,235
Two buttons follow, Mesh
then CFD calculation,

19
00:01:54,235 --> 00:01:59,105
and the bench runs the two phases
of the calculation on its own.

20
00:02:01,610 --> 00:02:06,526
Here is the result, on
the calculated case.

21
00:02:06,526 --> 00:02:10,871
The designer announced 1,156 newtons;

22
00:02:10,871 --> 00:02:18,646
the bench finds 808 newtons on the
blades, that is 30 percent less.

23
00:02:18,696 --> 00:02:22,649
The second phase, where the
propeller really turns,

24
00:02:22,649 --> 00:02:25,284
confirms this order of magnitude.

25
00:02:25,334 --> 00:02:28,758
This gap is to be looked
for first on the mesh side:

26
00:02:28,758 --> 00:02:32,272
to describe correctly the
boundary layer that creates

27
00:02:32,272 --> 00:02:35,786
the real suction and lift
on the back of the blades,

28
00:02:35,786 --> 00:02:38,372
the mesh must be of very high quality.

29
00:02:38,422 --> 00:02:42,907
The message for the design office:
keep a margin on the thrust,

30
00:02:42,907 --> 00:02:45,684
and refine the mesh before concluding.

31
00:02:47,786 --> 00:02:53,574
With the 3D model hidden, we see the
pressure on the skin of the blades.

32
00:02:53,624 --> 00:02:59,893
On this face, turned towards the
hull, the back of the blades is blue:

33
00:02:59,893 --> 00:03:07,236
the water is sucked there, and this is the
face that produces most of the thrust.

34
00:03:07,286 --> 00:03:13,648
The red nose of the hub is the stagnation
point, where the water hits head on.

35
00:03:13,698 --> 00:03:19,335
On the other side, the face that
pushes the water is in overpressure,

36
00:03:19,335 --> 00:03:22,030
green, with red along the edges.

37
00:03:24,174 --> 00:03:27,707
It is on the sucked face
that cavitation threatens:

38
00:03:27,707 --> 00:03:32,280
if the pressure falls below the
vapour pressure, the water boils.

39
00:03:32,330 --> 00:03:35,766
The scale of the pressure
cut carries a marker:

40
00:03:35,766 --> 00:03:40,372
the vapour pressure of sea water,
1,300 pascals at 10 degrees.

41
00:03:40,422 --> 00:03:46,700
The lowest pressure of the whole
calculation is 107,681 pascals,

42
00:03:46,700 --> 00:03:53,762
far above it: at this operating point,
the propeller does not cavitate.

43
00:03:55,853 --> 00:04:02,482
The cuts colour the water
itself. In the vertical plane,

44
00:04:02,482 --> 00:04:08,282
we see the jet accelerated
behind the propeller,

45
00:04:08,282 --> 00:04:16,568
and in the centre, in the shadow of
the hub, a trail of slowed water.

46
00:04:16,618 --> 00:04:20,646
Across the shaft, the
ring of the jet draws

47
00:04:20,646 --> 00:04:24,767
the stream tube that
the propeller creates.

48
00:04:24,817 --> 00:04:31,402
A cylindrical cut follows the water at
a given radius, through the blades.

49
00:04:31,452 --> 00:04:37,335
Finally, the iso-surfaces wrap the zones
where the pressure crosses a threshold:

50
00:04:37,335 --> 00:04:42,776
this layer shows how far, in the water,
the effect of the blades extends.

51
00:04:44,827 --> 00:04:49,338
The flow lines follow the path
of the water: it arrives slowly,

52
00:04:49,338 --> 00:04:53,991
crosses the turning propeller,
and leaves accelerated behind it.

53
00:04:54,041 --> 00:04:58,163
Everything we have seen comes
from one preset and two buttons,

54
00:04:58,163 --> 00:05:01,221
without setting a single
numerical parameter.

55
00:05:04,111 --> 00:05:07,212
With Heliciel, a boat
propeller goes from the

56
00:05:07,212 --> 00:05:10,589
drawing to the CFD simulation
in three gestures,

57
00:05:10,589 --> 00:05:15,138
and the bench shows us what the
classical calculation cannot see.

58
00:05:15,188 --> 00:05:18,810
We could now recalculate
the same propeller

59
00:05:18,810 --> 00:05:22,096
with a finer mesh preset, and compare.