Sound advice(?)

[Jim Starbuck]

Has anyone tried making the shell be part of the box?  I was picturing a design where the entire box/speaker assembly is turned upside down and the ceiling of the shell formed the base of the box, and the speaker is firing upward into the box/shell.  It would require the base of the box frame be profiled to the shape of the ceiling to maintain a seal (which could be challenging), but the acoustic coupling to the shell would be much more direct and controlled.  This is basically how the piezo wafer speakers are intended to work.

I’ve experimented with this technique on a few switchers where I’ve made part of the cab the speaker enclosure. I know we’re talking apples and oranges here as far as the engines you’re doing the installs in.
I’ve had mixed results with two basic techniques. The first is to build the enclosure out of .015 or .020 styrene with the outside wall dimensions the same as the speaker (usually 8x12) with no floor and as much height as chassis and speaker thickness will allow. I contour the top of the box to fit the inside of the cab roof then reach through the open box and seal it to the cab roof with canopy glue. When that’s dry I seal the speaker to the top edge firing into the enclosure.  This just adds whatever the thickness of would have been to the volume of the enclosure by making the cab part of it. This has worked pretty well except on one of my SW1s but I suspect I didn’t get an adequate seal. I simply haven’t opened that one back up yet to see why there isn’t much volume.
The other technique was to make a wafer of .020 styrene that fit the inside of the cab with the speaker inset into the wafer and sealing the perimeter essentially making the cab into the enclosure. It’s more difficult to get a tight enclosure this way though.

Jim

[peteski]

I think I should mention something about totally sealed speaker enclosures.  The atmospheric pressure changes quite a bit depending on the weather. Those pesky highs and lows.  The internal pressure in the enclosure will be whatever the barometric pressure was on the day it was sealed.  Then when the ambient pressure is higher, the speaker cone will not longer be in a neutral position - it will be pressed in. Reverse will happen when the barometric pressure is lower than the pressure inside the enclosure.  That might cause distortion since the speaker cone will have limited travel in one direction. I recommend a small pinhole (like #80 drill size or smaller) opening to equalize the pressures.

[John]

Gary -- sounds like a good undergrad physics project for your students .. design an optimum speaker enclosure for this sound file 

[Cajonpassfan]

What a great thread, very informative and thought provoking.
Interesting notion about the barometric pressure variation suggested by Peteski above. I've heard about the pinhole treatment before, never tried it. What's the collective wisdom here?
Thanks,
Otto

[kiwi_bnsf]

I think I should mention something about totally sealed speaker enclosures.  The atmospheric pressure changes quite a bit depending on the weather. Those pesky highs and lows.  The internal pressure in the enclosure will be whatever the barometric pressure was on the day it was sealed.  Then when the ambient pressure is higher, the speaker cone will not longer be in a neutral position - it will be pressed in. Reverse will happen when the barometric pressure is lower than the pressure inside the enclosure.  That might cause distortion since the speaker cone will have limited travel in one direction. I recommend a small pinhole (like #80 drill size or smaller) opening to equalize the pressures.

This is an interesting point, but I don't think these cell phone speakers are ever 100% sealed (their IP ratings for water ingress are not perfect). Gradual changes in atmospheric pressure should equalise through the layers and joints in the speakers.

I did quite a few experiments with tiny holes and even long folded ports, and they all sounded pretty terrible compared to a sealed box.

I could imagine that maybe a particularly well sealed box might have issues with a dramatic altitude change, but if you are drillings tiny holes to fix a bad sounding speaker, I would try and seal the hole up again after equalising. (Ideally you would use a slightly permeable material for the pressure equalisation but I think that this is overkill, and there are much more fundamental issues going on with these tiny speakers).

[peteski]

Tim,
I'm not talking about tiny "holes" just single small hole. 0.013" or smaller.  But if you think the actual speakers are not all that well sealed, then that will allow for the pressures to equalize.

[jdcolombo]

@jdcolombo is there an optimal thickness for the enclosure wall?

TIA,
Ed

Stiffer is better.  I usually use at least .030 styrene, .040 if possible.

[kiwi_bnsf]

Tim,
I'm not talking about tiny "holes" just single small hole. 0.013" or smaller.  But if you think the actual speakers are not all that well sealed, then that will allow for the pressures to equalize.

I tested 0.6mm and 0.4mm printed tubes in single, dual, and quad configurations. I tested tube lengths varying from 1 to 4mm (i.e. non-folded and folded). I tried tapered and non-tapered.

I also tested a single 0.2mm and 0.1mm hole drilled through a 1.6mm solid section of the enclosures.

All versions with holes sounded a lot worse than the standard sealed version of the exact same design

Since those tests, I've stuck with tightly sealed speaker surrounds for all my designs and I've not had any issues with variability.

I also strongly recommend against drilling any holes in locomotive shells or fuel tanks as it also has a detrimental effect vs a "closed" negative space around a speaker enclosure.

Cheers

[GaryHinshaw]

I had a chance to try a few more things this evening: first I removed the kapton tape and secured the 9x25mm box with double-stick tape instead (for now).  I can't say I noticed a dramatic difference, but I'm certainly prepared to believe that keeping the back of the speaker clear is better.

I also played quite a bit more with the individual sound slot levels and realized that the traction motor channel was cranked up to 140, which was not playing well with the dynamic brake channel at all.  (The ESU documentation on which slot the traction motor resides in is vague.  I had to determine by trial and error that it was slot 25, not 19.)  I played around with a bunch of other levels as well and I'm now quite happy with the dynamic brake effect.  The prime mover effect also benefits from the traction motor channel being reduced, but run 8 is still a borderline hot mess in my setup.  Run 5 sounds pretty good though, so I think I'm converging on a decent short term configuration. 

I'm still going to play around with box configurations - especially an integral shell design - to see if the prime mover can be made a bit more robust.

It's pretty daunting how many combinations of channel levels there are to assess though.  Even if you reduce it to the 10 most important channels and say that each can be only low, medium, or high, you're still talking about 3^10 = 59,049 possible combinations!   

[peteski]

It's pretty daunting how many combinations of channel levels there are to assess though.  Even if you reduce it to the 10 most important channels and say that each can be only low, medium, or high, you're still talking about 3^10 = 59,049 possible combinations!   

Ah, this  is the scientist in you talking. 
Sounds like a possible banner material.  Too bad DKS isn't doing that anymore.

[BCR751]

I think Lance Mindheim was on the right track with his idea of having outboard decoders specific to a particular loco and running the sound through an audio mixer and Blue Tooth to a set of headphones.  I haven't seen anything about this recently so maybe it has lost favor but, to me at least, it would be preferable to the tinny sound of N-scale installations.  I've heard a lot of locos with sound and no matter how the installation was done, it still doesn't work for me.  It always sounds as though the speaker output is being played in a five gallon pail.  I wish I was an electrical engineer and had the where with all to design a system that would do the trick but I'm not.  So, for now, I avoid decoders with sound entirely.

Doug

[woodone]

Well although SoundTraxx Tsunami’s do not have a drop in decoder for the Kato’s I find that the (sound equalizer )does wonders managing some sounds.
I have been happy with the results using a 9X16 speaker with the largest enclosure I can fit in.

[GaryHinshaw]

Doug, I share your skepticism about on-board sound, but I am coming around to it in certain circumstances.  At low volumes, your brain can fill in a lot of detail, and the resulting power of suggestion can be quite dramatic.  But there is a fine line between dramatic and irritating!  For filling in the bass, I do think a set of well-placed sub-woofers around the layout (a la SurroundTraxx) could be used quite effectively.  It's something I hope to explore some day.

I tried an implementation of Lance's headphone sound at Mark Dance's layout a few years back and I didn't like it at all.  (Mark didn't really, either.)  It certainly produces a full rich sound, but the effect of having the sound in your head while the loco is 'out there' is disorienting.  It felt like you were inside the loco itself, while at the same time it was off in the distance there.  I also didn't care for the isolating effect of the headphones while you were around other people in an operating session -- which is a social event, after all.  (I have the same feeling about wearing headphones outside.  I don't like to be cut off from the environment.)

[BCR751]

Doug, I share your skepticism about on-board sound, but I am coming around to it in certain circumstances.  At low volumes, your brain can fill in a lot of detail, and the resulting power of suggestion can be quite dramatic.  But there is a fine line between dramatic and irritating!  For filling in the bass, I do think a set of well-placed sub-woofers around the layout (a la SurroundTraxx) could be used quite effectively.  It's something I hope to explore some day.

I tried an implementation of Lance's headphone sound at Mark Dance's layout a few years back and I didn't like it at all.  (Mark didn't really, either.)  It certainly produces a full rich sound, but the effect of having the sound in your head while the loco is 'out there' is disorienting.  It felt like you were inside the loco itself, while at the same time it was off in the distance there.  I also didn't care for the isolating effect of the headphones while you were around other people in an operating session -- which is a social event, after all.  (I have the same feeling about wearing headphones outside.  I don't like to be cut off from the environment.)

Maybe I'm a bit premature in my assessment.  I didn't really consider the fact that the sound stays with you no matter where the train is but in my case, I have a small shelf layout and I'm never more than a foot or so away from the train. Also, I am the sole operator on the layout so there's no consideration for other people.  I would very much like to try out this system but, as I mentioned, I do not have the capability to build it.  So, I'm just going to stick with the non-sound decoders.

[ednadolski]

I tried an implementation of Lance's headphone sound at Mark Dance's layout a few years back and I didn't like it at all.  (Mark didn't really, either.)  It certainly produces a full rich sound, but the effect of having the sound in your head while the loco is 'out there' is disorienting.  It felt like you were inside the loco itself, while at the same time it was off in the distance there.  I also didn't care for the isolating effect of the headphones while you were around other people in an operating session -- which is a social event, after all.  (I have the same feeling about wearing headphones outside.  I don't like to be cut off from the environment.)

IIRC Lance said something to that effect as well.  This is what I like about the subwoofer setup - I only set it loud enough to be there, but not enough to become bothersome or intrusive.  The sound generally does seem to be coming from the engine.

Ed