PDA

View Full Version : NoiseCam goes 3D



Keith Oakley
4-Mar-10, 06:48 PM
We've now been using the NoiseCam system for 3 years to identify noise sources on individual hovercraft and several craft are now quieter as a result of targetted remedial action. The existing system gives a 2D picture of where loud noise sources are as a series of red spots on a cross sectional graphic of the craft. Sometimes though 2D doesn't tell the full story - yes that noise comes from the silencer but where along it? - the outlet at the back or that leaking joint at the front?

To answer that and improve picture quality I've been developing a 3D upgrade with now 24 microphones and new software. Here are some of the first test 3D noisepics. Each noisepic shows the noise in 3 pics, the first from the right hand side of the craft, then a plan view, then the familiar head on view we got from the 2D system.

This is a noisepic of a car air horn, producing an 833hz tone, placed on top of the engine on my F35. I've shown a graphic of the horn so you can see that the noise is pretty accurately located. Ideally the spot would be round. As you can see its elongated particularly vertically. Thats due to astigmatism problems on the current microphone array, I've got to do some more work on that.
4420

This is a noisepic of the noise at 312hz produced by turbulent air hitting the fan. The turbulence was deliberately created by a 150mm square plate in front of the fan. (actually gaffer tape across 9 squares of the fan guard) As you can see there are two sources, one emitting from the front of the duct and other where the turbulent plume emerges from the rear of the fan. The 2 sources are about 180degrees out of phase (one blows when the other sucks!) confirming its a dipole source. The source is very directional - its much louder to the front of the craft than at the sides.
4421

This is a noisepic of the same 312hz but I've used some new software features to more accurately locate the louder of the two sources - the front one.
4422

Keith Oakley

Ian Brooks
4-Mar-10, 07:23 PM
I never cease to be impressed by this system! If this had been commercially available it would have helped me with a serious issue at work over Christmas - knowing where the noise came from would have been very helpful.

In the end, I used a "scanning laser vibrometer" - a device that measures the oscillation of the surface in a line-of-sight and superimposes the motion over a photograph of the specimen - great for locating mechanical resonances in silencers, etc. But at 1200 per day, a little pricey!

Ian

KipMac
4-Mar-10, 08:11 PM
I never cease to be impressed by this system! If this had been commercially available it would have helped me with a serious issue at work over Christmas - knowing where the noise came from would have been very helpful.

In the end, I used a "scanning laser vibrometer" - a device that measures the oscillation of the surface in a line-of-sight and superimposes the motion over a photograph of the specimen - great for locating mechanical resonances in silencers, etc. But at 1200 per day, a little pricey!

Ian
Borrow a 3 year old kid.
They can hear and detect a packet of crisps being opened from 300 metres

Kip

Brian G. Reynolds
4-Mar-10, 09:06 PM
Very impressive Keith, well done indeed!

Out of interest, could this system be used on the cruising craft at some stage?

I would love to make mine quieter and would be very interested in a starting point.

Keep up the good work, even if I do not understand a word you say :-))))))

Brian.

broad5186
4-Mar-10, 09:34 PM
excellent work kieth, you really must put this work forward for a Masters by research, having said that it's uniqueness could easily qualify for Phd

others should see and learn from your findings

atters
5-Mar-10, 06:18 AM
Would love too see how the Sevs do.

Keith Oakley
5-Mar-10, 11:04 AM
The Sevtecs are just as prone to obstructions in front of the prop/fan as any other design. I think John R gained a couple of dba off by moving 2 jockey pulleys further away from the prop.

I've tested over 130 craft so far, cruising craft as well as racing - for example if Brian's Osprey5 is similar to others I've tested then the splitter is a major source of noise.

This is part of a paper I presented to a recent international scientific conference on this 'seeing noise' technology in Germany. Most of the other papers came from Profs & Phd folk from Europe and N America. Sadly mine was the only one from the UK. To some extent we can achieve more than the professionals. For example when I discovered the silencer is a major problem on F3 TZRs, and Ricky Goosey came up with a silencer that knocks 12dba off, without hurting performance, it was reasonably easy to get that fitted to the whole F3 fleet. Conversely if a major EU funded R&D programme finds noise from Airbus wing flaps for example, its a big exercise to get maybe 2dba off. It probably doesn't even start until a new aircraft is being designed.

I'll post some more pics over the next few days on the Builders Corner Forum with generic recommendations for members as to how to minimise noise.

Keith Oakley

hector46
5-Mar-10, 11:57 AM
Good work Keith,

I want to reduce the noise signature of the next Sev build as much as possible. I am interested if a change from 25mm square tube to 25mm round tube on engine framing makes much difference, factoring in the distance from the prop or fan?

Steve

Brian G. Reynolds
5-Mar-10, 12:21 PM
I've tested over 130 craft so far, cruising craft as well as racing - for example if Brian's Osprey5 is similar to others I've tested then the splitter is a major source of noise.

Keith Oakley

That is a bit scary.... can I pose this question on another forum? Sounds like I need some mods...

B.

atters
5-Mar-10, 01:20 PM
Many think that the noise is what is holding hovercraft back from becoming a mainstream recreational toy and tool. With your research I am sure that very soon craft will be as quiet as other watercraft and ATV's

Well done so far.

Keith Oakley
5-Mar-10, 04:15 PM
We've now found a 'recipe' of noise reduction measures that so far has got 2 F35 craft signifantly below the EU Directive requirements for a Personal Watercraft (ie Jetski). That calls for a craft to be below 75dba on a fullpower flyby at 25metres- contrast that with the racing limit of 96dba at 25metres. So we're getting there but its still work in progress.

I'll include strut noise and recomendations in the builders corner posts. Any other noise areas folk would like covered?

Keith Oakley

trev
5-Mar-10, 04:53 PM
Nice to see your still working on that Keith keep up the good work.

Brian my craft was 87db at 25m at Loch Long 3 years ago

Trev

Keith Oakley
7-Mar-10, 07:39 AM
I've done a post on strut noise and I've got a post on splitter/stator noise underway. Any others folk want to see covered?
Keith

Hovertrekker
7-Mar-10, 12:07 PM
I've done a post on strut noise and I've got a post on splitter/stator noise underway. Any others folk want to see covered?
Keith
How about a proper airbox vs. straight filters. I'm thinking in terms of a snowmobile 2 stroke engine, but aren't 4 stroke engines equally susceptible to intake noise?

Ian Brooks
7-Mar-10, 01:24 PM
Hi

Here's an interesting paper on the subject of noise. Palmer has done considerable work over the years on hovercraft efficiency and noise reduction - here we can see the origins of the Sevtec designs, and one might get a glimpse at the analytical nature of the design process he employed.

4427

I have other papers which I would like to share but I probably need to gain his blessing for those that are not already in the public domain.

Ian

Keith Oakley
7-Mar-10, 06:31 PM
Interesting paper from '75 concluding thats it's mostly tonal noise at the engine firing frequencies plus harmonics and the fan blade passing frequency plus harmonics; and that moving the prop further away from obstructions and improving the exhaust silencer reduces the noise. I'd agree wholeheartedly with all of that.

Interesting to see he got 83dba flyby at 50 feet from 7bhp engine running at 3080 rpm - which equates to about 80dba at 25metres. We've currently got 70-72dba from 28bhp at 4100rpm flyby at 25metres.

On the airbox I recently published the formula (from a book) for designing them - I'll repeat it in the builders corner series. My tests suggest only a properly designed airbox achieves noise reduction - any old box you may have on there for spray protection has no effect on noise. I mostly see intake noise as a problem on 2 strokes but I think I have seen it on higher revving 4 strokes. I've got one experimentally on my low revving 28bhp B&S 4 stroke mainly to check it didn't drop the rpm. I need to look back through the data to see if it had any noise effect.

Keith Oakley

Ian Brooks
7-Mar-10, 08:03 PM
Interesting to see he got 83dba flyby at 50 feet from 7bhp engine running at 3080 rpm - which equates to about 80dba at 25metres.
This was probably pretty good at the time (1975) - a modern built to plan Vanguard (25hp) was measured at 76 dB (Steve Holland) 'out of the box' in Ardgarten. My Surveyor was last measured at 77dB, although this wasn't in the most 'standard' of conditions, it's been getting quieter each year as I address the obvious issues. The Scout measures about 74dB with a 13hp engine fitted.


We've currently got 70-72dba from 28bhp at 4100rpm flyby at 25metres.
This is very encouraging, I'm keen to see how that has been achieved, and keen to encourage all cruisers to do what they can to bring their noise signature down.

Ian

atters
8-Mar-10, 05:53 AM
Right, so I have it....

1. Make a intake air box with a foam lining
2. Move the fans as far from the frame as I can
3. Make frame aerodynamic (later)
4. Exhaust: add on silencer
5. Engine cover with foam lining NOT eliminating air flow.

If I understand all that has been said, these few things should reduce the noise substantially, right?

Keith Oakley
8-Mar-10, 07:34 AM
Good start but theres a few more to come yet! On 1 I haven't tried an engine air intake silencer based on the use of sound absorbing foam - sounds like a fire risk if they get soaked with fuel and you get a backfire! The successful ones I've tested are based on airboxes which by their design form a low pass filter cutting down the higher frequency harmonics.

Keith Oakley

Hovertrekker
8-Mar-10, 12:26 PM
On the airbox I recently published the formula (from a book) for designing them - I'll repeat it in the builders corner series. My tests suggest only a properly designed airbox achieves noise reduction - any old box you may have on there for spray protection has no effect on noise. I mostly see intake noise as a problem on 2 strokes but I think I have seen it on higher revving 4 strokes. I've got one experimentally on my low revving 28bhp B&S 4 stroke mainly to check it didn't drop the rpm. I need to look back through the data to see if it had any noise effect.

Keith Oakley
I have the snowmobile airbox for my Rotax793 that I will be testing in the near future with a handheld Db meter. It's all plastic, with multiple chambers, sort of like the inside of a muffler with several layers of foam filtering. Slightly bulkier than a simple aircleaner cover but not heavy. I hope to get some readings to compare.