SimFeedback-AC DIY Motion Simulator 10Khz Research

Hey all,

I'm really keen to build one of the SFX-100 rigs, but I've heard (been in the presence of) my brothers and the 10khz servo/driver noise really does my head in, so much so that I would not be able to put up with it. 5-10m of it, and I'm done. This is a darn shame :)

So, I'm going to see if I can get rid of it. Either electronically (preferred) or with accoustic treatment(s).

In this thread I'm simply going to document my journey. At this point I am unsure if it'll come to anything. Maybe I won't succeed. Maybe I'll get half way. Who knows.

The story begins with Neil buying himself a single driver and servo, a 3d printer, and saying to himself "hmm. what next" :)

My full notes are here, but I can't promise you'll understand them: https://docs.google.com/document/d/1J4AUnKNYtFRujTR9BFBB_lJudZRP_02BOt6lfAIN45I/edit?usp=sharing

Look below, I'm going to try to post a more understandable version of my journey!
 
Last edited:
Just to weigh in with some ideas but first a query.

Have you guys tried to use active noise-cancellation?

Use a microphone near the source, have this then go to an amplifier and to a speaker. Now reverse the phase of the connections to the speaker. The speaker(s) via reverse-phase should cancel out the sound emitting from the source into the room. This may work better however with lower frequencies than high frequencies.

It's also possible that some room environments may be worse than others as they may have more reflective walls and objects within the room. For this its possible to consider materials like acoustic panels or acoustic tyles. Be aware though that the cheapest foam tiles on your eBay may not be that great.

Another material to consider is Mass Loaded Vinyl (MLV) if purchased in a thin layer (2mm) it could easily be wrapped around an object but perhaps worth trying.

I have purchased materials for my own build from Stp that are specific for vibrations, sound-deadening and noise block. Not sure how well these work for high frequencies but they greatly reduce road/engine noise in cars when different materials are combined for low and high-frequency control.

You can find various videos on these topics and products or making d.i.y acoustic panels.

Important info regards noise cancellation and frequencies.



Without even using a microphone for active cancellation as we know the culprit Hz. If someone could try generating a pure tone of the Hz needed but invert/reverse its phase (perhaps in audacity) then save this as a playable sound file and playback via a speaker to see if it helps at all?

I would even recommend trying a "tactile exciter" attached to the unit to use as the speaker in an attempt to cancel out the noise generated.
 
Last edited:
Upvote 0
Regarding materials, I found this helpful:


Particularly with regard to choosing a material that directly addresses the need to kill high frequency. It's the reason I didn't bother with many foams, and went straight for polyester fibre.

"Polyester fibre is spectacular for it's unique blend of heavy density (approximately 2000g/m^3) and porosity. It's sound absorbtion increases with the frequency of the sound, hence it's most effective at high frequencies."

Unfortunately, the same article doesn't provide any info regarding how well MLV might address high freq. specifically. One datasheet for a particular MLV product showed an increase in SRI performance as freq increased (https://www.pyroteknc.com/products/wavebar/wavebar/, then click link). At 5khz SRI was 34.6 for 2kg/m2 (1.2mm thickness) and 48.7 at 10kg/m2 (4.9mm thickness). Those kinds of reductions would seem to fit the ticket (ideally we want a 40db reduction @ 10khz on the motors); but it might come out different depending upon the type of MLV used. It'd have to be tested. Looks like if you went the MLV route you'd definitely want to pick the 10mm stuff, for sure.
 
Upvote 0
I think it would be worth trying various materials and options.

Here are some other examples of usage scenarios...
(Possible to use subtitles)


Due to my experimental/custom and a rather excessive transducer installation. I have 3 layers of StP materials being applied below the bottom 8mm aluminum plates my rigs tactile isolator supports are going to be bolted too.

Heres an image of the tactile/isolator frame support solution my seat and pedals will be mounted on and then bolted to the 8020 chassis for the SFX 100 motion. Six of these dual isolation towers will be used to balance/support the custom design steel tubing frame being made.

While this is not directly similar to the issue here, I believe the combination of these materials will help greatly with not just vibrations but possible reverberations and harmonics being generated. Its the next phase of my own build I have to do, so at the moment I can't personally say from testing how well they work.



Description, on each box, if interested, read what each material is used for, as seen in one of the videos above it can greatly reduce the frequencies in pink/white/brown noise, so it may be effective with this issue being discussed:

Just trying to help...







 
Last edited:
Upvote 0
I have a fair amount mlv left over from car sound deadening projects and could do some tests on that too. 1lb per ft^2. Might take me a while. In my experience Mlv will definately block the freq but it probably has to be completely sealed.
 
Upvote 0
@cfischer9009 if you do manage it, any chance you're able to measure similar to what I've done? i.e: some kind of mic about 150mm away from the noise source, then basline recording + with material in place? It'd give us a relative measure of difference. Completely understand if you can't (or it's too much of a pain!).

Also, I don't know if I wrote about it but I think I did show it in the pic: I was able to get a 40db reduction with a "tube" of poly fibre. I didn't close either end (although the fibre did collapse to the center when put over the motor; so there's some amount of it over the top of the motor).

I found it necessary to extend the fibre cloth below the motor itself. It made a differnce to the measurements. I did measure the noise emission above, below, on the sides of the motor, and also the motor axis before covering in cloth. From what I could measure; MOST of the sound is coming from the sides.

What I should do next is measure the sound reduction along the axis of the motor with the cloth in place; to see if there's still loads comnig out the top. Prob I have at the moment is that even if I turn it off; my ears are still ringing from testing/putting it all in place. I guess that means I have to trust my instruments :)
 
Upvote 0
10kHz testing.jpg
 
Upvote 0
Have you noticed the harmonics on single phase driver on 20000Hz, 30000Hz???

Also the 10Khz difference between the two is about 10dB... That's what I measured when I was comparing single phase 220v vs 110v power...
It's worth punting out on the chart but I wouldn't say I noticed it in person. Maybe if I did the two conditions back to back. As mentioned in the image the test was done on different days.
 
Upvote 0
Interesting how there's a similar dB peak at around 10k from the motors, in both single and three phase tests.

Looks like the noise emitted by the driver goes away in 3 phase. That might mean that an alternate 3 phase supply to the drivers might elimitate the noise coming from those (no idea of cost effectiveness nor what noise might be generated from that unit itself, if any).

Some kind of mitigation would still be required for the motors then... (sits and ponders).

Thank you for doing this!
 
Upvote 0
Interesting how there's a similar dB peak at around 10k from the motors, in both single and three phase tests.

Looks like the noise emitted by the driver goes away in 3 phase. That might mean that an alternate 3 phase supply to the drivers might elimitate the noise coming from those (no idea of cost effectiveness nor what noise might be generated from that unit itself, if any).

Some kind of mitigation would still be required for the motors then... (sits and ponders).

Thank you for doing this!
The noise in the driver definitely does not go away for 3 phase. Its better, but its still there. The motor does seem much less affected by the phases.
Glad I could help.
 
Upvote 0
I think the reason for the sound difference is that the input will get filtered by more diodes (8 instead of 4) because using 3 phase will utilize two rectifiers instead of one.
I didn't expect the sound to go away, it will still be switched at the same freq.
 
Upvote 0
I also experience the noise - so much so that I've stopped using the rig altogether, it's simply too painful.

This is a good news story by the way - keep reading.

Based on my reading of various articles about absorbing / blocking sound I have built heavy wooden boxes using 18mm thick beech, lined with a Basotect foam material. These wrap both the motor and the driver units. The basic principle appears to be to provide a way for the sound wave energy to be absorbed (open pore foam) in combination with a high density (reflective?) exterior shell. The idea isn't mine, I borrowed it from ideas I've seen in articles that describe how to wrap heavy machinery and heat pumps - the latter often uses a multiple shell technique with baffles.

I will be the first to admit that I am not following a strictly scientific approach here. I am not taking each material or combination in isolation; nor do I have a "clean room" (anechoic chamber) within which to measure the differences in sound absorption / reflection - heck; the only tools I've got to measure the sound are my ears and the Spectroid app on my Android phone.

The latest version of the driver unit box is a double shell design that uses 70mm thick Basotect acoustic panels along with the 18mm beech panels. Air flow is achieved by a large CPU fan pulling air through the entire unit via a series of 3 baffles (also padded with Basotect) at each end. The box isn't yet finished; I'll report more in the coming days on it's effectiveness.

Scornflake's recent discovery that Polyester fibre dampens the noise was a motivator.

Yesterday I decided to try the fibre idea. I took each of the wooden boxes I had made for the motors and ripped out the Basotect material, doubled up the cheap polyester blankets I'd purchased earlier that day and jammed them into the enclosure.
View attachment 336569 View attachment 336570 View attachment 336571
Each time I dropped one of the enclosures onto the motor while the rig was running (but stationary); there was an audible drop in the piercing level of sound within the room.

It's AINT pretty, in fact it's downright ghetto at this stage. But guess what: I almost have silence. Almost - the piercing sound is still there and of course I'm listening out for it, but the difference is no less than stunning.

The noise is without a doubt coming from both the motors AND the drivers in my case.

You will see in my pictures that the double walled box isn't yet finished; in fact right now the upper cover is just placed on there by hand and there isn't a front cover at all.

Power to my sim, CPU, drivers, the whole lot - is driven from a single wall socket. An electrician may well use a palm-to-forehead movement to properly express their dismay at the number of extension leads I'm using.

I'll attach some photos of the boxes in the spirit of sharing, I'm more than happy to receive any suggestions for improvements.
View attachment 336572

My current goal:
- from a whine/noise perspective - to go from "some" to "none" using absorption/blocking techniques

My next questions / steps (none of these have a basis in measured facts, e.g. it's more things to try):

Q: how can I quantify the improvements properly from this point forward? (aside from the obvious, stop being lazy and start measuring this you fool!)
A: try to roughly gauge how much noise is coming from the drivers vs motors (it's definately from both)
A: try using the Spectroid app at a constant distance from the source being measured and start writing some of this down properly to document the effect of each stage.

Q: does having more fibre layers increase the dampening effect?
A: increase the number of polyester fibre layers on each of the motors.

Q: what role does wood thickness / density play here?
A: increase/decrease the thickness of the wooden boxes and/or try the same number of layers with either a PLA wrap or different types of wood.

Q: is any noise leaking out the bottom of the motor mounts?
A: fully encapsulate the motors using wooden caps (try to reduce noise gaps at the base of the motor)

--
John C

PS: I'm Scornflakes brother, living in Switzerland.
Mmmm... I'm pretty sure I've got some thick cardboard tubes up in the rafters of my garage, the heavy kind that carpets and industrial fabrics come on. From memory the tube diameter seems about right for the pistons. Those could be a nice quick way to try your enclosure-and-blankets method. They're a doddle to cut to length with a hand saw, and at around 10mm wall thickness (or more) and being made from high density cardboard, I imagine their reflective/absorptive properties are pretty good.
If anyone else wants to try this, try hitting up your local carpet supplier for old tubes. They'll have them in abundance, probably glad to get rid of them and they shouldn't cost you a penny.
Edit: take a saw with you or you might not get them in the car... ;)
 
Last edited:
Upvote 0

If you recorded this, then you could try and play it back perhaps via a tactile exciter (it will act as a mini speaker) and have it attached to the unit. Play it back with reversed polarity (swapping speaker cables of + to -). You could also use a mic to constantly monitor the source and replicate it in realtime if it altered.

Such exciters can be powered with a cheap amp and it would help determine if sound cancellation (with identical source) just being out of phase helps eliminate the problem or not.

May be of some help
 
Last edited:
Upvote 0
Been doing some work on an enclosure, as I slowly build up my actuators. Thought I'd post some progress. The vid below is for the motor/actuator part. I'm also working on a fully enclosed case (with extractors) for the drivers.

 
Upvote 0
I’ve recently got some open back Beyerdynamic dt1990 pro’s and they seem to completely cut-out the sound when on but because they are open I can still hear everything else.

Possibly an option for you @Mascot as you’d still be able to use your speakers and have crew chief through the headphones? But easier and probably cheaper than trying to cover all the motors/drivers? Also more like wearing a helmet with the sound dampening ;).
 
Upvote 0
I’ve recently got some open back Beyerdynamic dt1990 pro’s and they seem to completely cut-out the sound when on but because they are open I can still hear everything else.

Possibly an option for you @Mascot as you’d still be able to use your speakers and have crew chief through the headphones? But easier and probably cheaper than trying to cover all the motors/drivers? Also more like wearing a helmet with the sound dampening ;).
Oof! £365 at Amazon. I'll add them to my notes though. Thanks mate.
 
Upvote 0

Latest News

Are you buying setups?

  • Yes

  • No


Results are only viewable after voting.
Back
Top