Don Gibson wrote:How can a EQ make the bass any better if the speaker can't reproduce it?It can't. EQ can reduce highs to fool your brain/ear that you are hearing more bass (by removing the highs/mids).
How can a EQ make the bass any better if the speaker can't reproduce it?
It can't.
EQ can reduce highs to fool your brain/ear that you are hearing more bass (by removing the highs/mids).
Good way to put it Don. In actual fact a really good tweeter (1" or even smaller) can vibrate from 10Hz (Vibrations per second which is roughly 1 and 1/2 octave below the range of human hearing) to 20,000 Hz or more. The problem as has been said is a small speaker can't move enough air to get a hearable Bass (or low midrange) wave to your ear from 2 or 3 feet away. EQing out the midrange (2,000Hz t0 4,000Hz typically) means that the percentage of Bass and Treble in the mix has been upped. Hence it sounds like there is more Bass there. Since most on board sound tends to be set extremly loud to start with, turning the volume down and if possible lowering midrange EQ will indeed help the quality of sound. I keep telling people mixing sound for me that lowering the midrange in effect is the same as raising the Bass and Treble and doesn't make the output of the channel clip..... Even 3 Db of change in the midrange can make a big difference in the sound.
"With EQ, you can also screw things up royally" - Mackie Mixer Owner's Manual
BTW, Electrical speakers have been around commercially since the late 1920's, I believe. Very little was done to improve them until the Mid 1980's when several improvements took place. Again in the mid to late 1990's another round of major new ideas improved and allowed speakers to be much smaller and still produce better sound. Bose was instrumental in leading the research into speaker placement being a major factor. According to some people in the Pro-Audio profession, we are on the verge of some more major changes in speaker technology. I would not be at all suprised if within 20 years the sound we are all looking for in a model loco becomes a reality. Unfortunately by that time I will be sitting around in my wheelchair in the nursing home saying, "WHAT, Speak up, Huh?"
Edit: I must have been distracted when I posted this..... It should be Midrange 400Hz to 4000HZ (depending on who you read). Sorry! That puts it in musical terms from about the A above middle C to about 4 octaves ubove that. The improvement made reducing the midrange assumes that you have some bass there to begin with which onboard sound really doesn't have. So ignore most of this post!
Ray Seneca Lake, Ontario, and Western R.R. (S.L.O.&W.) in HO
We'll get there sooner or later!
Seems that QSI "heared" me....lets see what happens with this new speakers...but all you think ahead, 3, 5 years ahead in this DIGITAL WORLD!!!
Try this:
http://www.qsisolutions.com/news/qhbspk_072907.html
Today's sound decoders use digital technology similar to that found in CD players to store and reproduce the sounds associated with our favorite engines. The original recordings are usually reasonably well done, and the digital technology allows the recordings to be stored and played with a quality that is very close to the original sound. Indeed, listening to the output of one of today's top sound decoders through a good set of headphones can be just like standing inside the engine compartment of a classic or modern diesel (standing inside the steam chest of a steam engine is probably not a good idea).While the quality of the sound at the output of the decoder is quite good, it must somehow be coupled into the surrounding air to create the sounds waves that we hear. There is the weak link in the system: the speaker. Unless you operate one of the larger gauge models, there is probably only room for maybe a 1" speaker (or smaller) in your model. In speaker theory, there is a frequency, designated Fs, that is the free air natural resonance frequency of the speaker. The critical information related to Fs is that the speaker is physically unable to reproduce any frequency significantly lower that Fs. Thus, the lower Fs, the lower the minimum frequency that the speaker can reproduce and the more bass. Fs is determined primarily by the moving mass of the cone and the stiffness of the suspension. Small speakers have a small cone, and therefore a small cone mass, and normally have a relatively stiff suspension so that the speaker cone does not move too far and damage the cone. The net result of this design is that most of the speakers used in model train sound reproduction have a relatively high Fs and are therefore are unable to reproduce the low sounds associated with "diesel rumble" or the throaty sounds of a steam chuff. Even worse for the model sound enthusiast is that when you place a speaker in an enclosed baffle (such as an engine car body or other small enclosed space), the value of Fs increases due to the stiffness of the air trapped behind the speaker (note that you do need the baffle, because if it is not present, the sound behind the speaker will simply flow around the edge of the speaker and cancel the sound in front).QSI Solutions has found a speaker that significantly improves on the existing situation. It is shown in the picture below:
Today's sound decoders use digital technology similar to that found in CD players to store and reproduce the sounds associated with our favorite engines. The original recordings are usually reasonably well done, and the digital technology allows the recordings to be stored and played with a quality that is very close to the original sound. Indeed, listening to the output of one of today's top sound decoders through a good set of headphones can be just like standing inside the engine compartment of a classic or modern diesel (standing inside the steam chest of a steam engine is probably not a good idea).
While the quality of the sound at the output of the decoder is quite good, it must somehow be coupled into the surrounding air to create the sounds waves that we hear. There is the weak link in the system: the speaker. Unless you operate one of the larger gauge models, there is probably only room for maybe a 1" speaker (or smaller) in your model. In speaker theory, there is a frequency, designated Fs, that is the free air natural resonance frequency of the speaker. The critical information related to Fs is that the speaker is physically unable to reproduce any frequency significantly lower that Fs. Thus, the lower Fs, the lower the minimum frequency that the speaker can reproduce and the more bass. Fs is determined primarily by the moving mass of the cone and the stiffness of the suspension. Small speakers have a small cone, and therefore a small cone mass, and normally have a relatively stiff suspension so that the speaker cone does not move too far and damage the cone. The net result of this design is that most of the speakers used in model train sound reproduction have a relatively high Fs and are therefore are unable to reproduce the low sounds associated with "diesel rumble" or the throaty sounds of a steam chuff. Even worse for the model sound enthusiast is that when you place a speaker in an enclosed baffle (such as an engine car body or other small enclosed space), the value of Fs increases due to the stiffness of the air trapped behind the speaker (note that you do need the baffle, because if it is not present, the sound behind the speaker will simply flow around the edge of the speaker and cancel the sound in front).
QSI Solutions has found a speaker that significantly improves on the existing situation. It is shown in the picture below:
My (Don's) comment:
This is what is known as a high excursion "woofer." Because it moves further, it extracts more bass. However because the cone is traveling further, it also means that it's going outside the linear gap of the magnet. So this creates greater distortion. Luckily we don't mind a little distortion when a lot of locomotives main noise is pink/white band type which is friendly to mild distortion.
Don - Specializing in layout DC->DCC conversions
Modeling C&O transition era and steel industries There's Nothing Like Big Steam!
selector wrote: EL, I think he means the bass is a gonner anyway, but you can reduce the nasal reproductions in what the speaker will produce in abundance, and coincidentally what your ear is likely most selectively attuned to, by equalizing their volumes downard a bit...thus enhancing the somewhat higher freqs that will tell us we have improved sound in our tenders? Did I get that right?
EL, I think he means the bass is a gonner anyway, but you can reduce the nasal reproductions in what the speaker will produce in abundance, and coincidentally what your ear is likely most selectively attuned to, by equalizing their volumes downard a bit...thus enhancing the somewhat higher freqs that will tell us we have improved sound in our tenders?
Did I get that right?
Yes.
If you cut "nasal" on a speaker that only produces nasal, all you're doing is turning it down, not changing the timbre.
With this speaker, you can cut nasal and still have something left.
A speaker like this will also work better with a stationary subwoofer, since it won't leave as much of a hole in the low midranges as other speakers will.
Gentlemen: (Oversimplification)
OUR EARS hear vibrations in air. Loudspeakers move air.
Speakers were invented by Bell Telephone Labs around 1915 for the Motion Picture Industry ('Talkies').
Small speakers are grossly inefficient at LOW frequencies. (Laws of physics). Since BASS requires moving large quantities of air,The lower the frequency / the more air is to be moved - due to the inefficiencies of the human ear.
Large speakers can move more air, therefor are more efficient for producing audible lower frequencies.
Improvements have been made in Loudspeakers over the years, but they've been relatively minor - Permanent Magnent have replaced Electro magnets - Cone materials and suspensions have changed resonent frequencies somewhat in 92 years - but Laws of physics are still in place. Most 'improvements' have been in controlling propagation (dispersion) and resonances).
Future: small 'satelite' speakers with a sub-woofer have become ''Theater or Surround" for leading edge types. Digital transmission is already here.
SOUNDTRAXX already has a 'black box' for use with existing "Surround" Speaker systems. They cancelled entry into the market for various reasons, including Co$ts and the impending threat of litigation from MTH re: 'Back EMF".
Multi-channel Sattellite sound with Sub-Woofers is the wave of the future - unfortunately it requires the big bucks. 'Theater' Sound is only as good as its sub-woofer (the most costly part) of the system. Theaters use multiple 15" drivers (moves air).
Like with Audio systems, what we buy is what we think we can afford - not necessarily desire. My $200 Radio Shack sub-woofer won't go below 50 hz with any amplifier. (To lower by one 8va requires 5X the cost).
Maybe SOMEDAY?
Don Gibson wrote:Like with Audio systems, what we buy is what we think we can afford - not necessarily desire. My $200 Radio Shack sub-woofer won't go below 50 hz with any amplifier. (To lower by one 8va requires 5X the cost).
That's why I used to love my 18" JBL Pro subwoofers They could reproduce full usable waveform down to 10Hz. Think earthquake. And, they could handle something like 3500 watts RMS. They also had a 9" throw in the cone. They could move some major air. Of course today they run about $1500 each.......... I also get to wear hearing aids these days.
There are many ways that sound for model railroads can (and will be) improved in the next decade or two. The question is: Will You and I be around to see, or hear it?
From the far, far reaches of the wild, wild west I am: rtpoteet
jeffers_mz wrote: You can do this, and it will work. You can probably devise a capacitance filter that will roll of the volume grasdually on the speaker for the block being exited, and bring it up gradually in the block being entered.
You can do this, and it will work. You can probably devise a capacitance filter that will roll of the volume grasdually on the speaker for the block being exited, and bring it up gradually in the block being entered.
Perhaps. OTOH, I have no idea how something like that might work let alone put one together - I fully understand how relays work and how to use them but not how to use a capacitance filter in such a capacity.
You're still going to need a sound source, and probably a way to at least approximate prime mover speed with train speed. Manual would work, and I have nothing against playing the guitar, and nothing against playing the trains.There is software for the PC, free software, that will let you live mix a 5.1 spread using a simple joystick. Put your manual ptentiometer in the joystick handle and it wouldn't be too tough to make it look like a throttle.Once the PC is involved, you can try some of the 16 bit samples suggested earlier, and I suspect that once you hear them, 8 bit sound might get old pretty fast.
You're still going to need a sound source, and probably a way to at least approximate prime mover speed with train speed. Manual would work, and I have nothing against playing the guitar, and nothing against playing the trains.
There is software for the PC, free software, that will let you live mix a 5.1 spread using a simple joystick. Put your manual ptentiometer in the joystick handle and it wouldn't be too tough to make it look like a throttle.
Once the PC is involved, you can try some of the 16 bit samples suggested earlier, and I suspect that once you hear them, 8 bit sound might get old pretty fast.
Once a PC is involved why not use it to control the trains as well?
What I don't understand is why some bright boy at one of the controller companies hasn't realized that they are designing digital equipment to mimic analog controllers, when all they realy need to do, is make a interface ( USB or wireless ) to hook in a PC between the control pack and the track - it's not like the PC can't send out and recieve thousands of signals a minute - even an old 386 could do the job and you can get those for free. Yes, I know some people have done it on their own, but, not everyone has that type of skill.
Put a simple sensor system along the track and combine it with motor feedback, the computer could caculate the scale speed to less than a .5 mph, and sync the sound to the proper set of undertable speakers accordingly, and let the user adjust the sound with the computer's volume & equalizer controls to get just the right tones. If sensors 23 through 36 are on a incline, and the computer knows the train is going uphill, it can increase the sound of the engine like would happen prototypicaly.
Then the company could design a control pack to look just like a real throttle, break, horn and bell controls, instead of push buttons and dials. Control a few thousand trains and other things on a layout all at the same time.
In a way, this is kind of a neat time to be modeling sound. Back in the days of HF train control, (and in the days when chase syncing DA to MIDI cost $100k ballpark), there was all kinds of experimentation going on, and new ideas appearing frequently. Sort of like the wild west before and after it got all settled and civilized.Sound is just another palatte to paint with, except right now, it's all new and exciting, instead of programmed options you choose from a list. The next few years should be pretty interesting, if nothing else.
In a way, this is kind of a neat time to be modeling sound. Back in the days of HF train control, (and in the days when chase syncing DA to MIDI cost $100k ballpark), there was all kinds of experimentation going on, and new ideas appearing frequently. Sort of like the wild west before and after it got all settled and civilized.
Sound is just another palatte to paint with, except right now, it's all new and exciting, instead of programmed options you choose from a list. The next few years should be pretty interesting, if nothing else.
Either they will be interesting or the companies will fall flat on their face because they failed to listen to the customer - again.
Don Gibson wrote: Like with Audio systems, what we buy is what we think we can afford - not necessarily desire. My $200 Radio Shack sub-woofer won't go below 50 hz with any amplifier. (To lower by one 8va requires 5X the cost).
I humbly disagree. The point of this sound is not to have your whole house shaking from a lil HO locomotive. Too much noise will drive you from your layout. (Especially mine with my 18 locomotives.) And it's not like we need the most accurate bass possible. Most of the "chuff" and "rumble" we here isn't produceed by a Phil. Harmonic Symphony. So accuracy of bass isn't key here. A simple re-eq'd 5" port tuned tunnel sub will do the job quite nicely from 80Hz->40Hz (Audible Threshold). This is actually somewhat overkill. These speakers can be had <$50.
So $$$$ exspensive speakers are not necessary. (This from the man with a $50,000 B&W/ Velodyne speaker setup in his home theater)
The problem here is bass extension. You can do this several ways:
1. Create a longer excursion of the cone. (As QSI has done) This introduces more distortion...but again this isn't a problem.
2. Add multiple redundant cones that concentrate at that frequency and roll off above a certain frequencey. (Inefficient)
3. Roll off the middle/high range and have multiple speakers. (This was partially attempted by earlier QSI in BLI and P2K locos so equiped) But this distorts the middle/high ground
4. Increase cone size. (Not possible with our small engines)
But with we have right now, I am very happy. If I listen to old tapes of trains passing through a valley or mountain at somewhat a distance, it doesn't sound that much different from what we are actually hearing.
Side note:
Unfortunately our sensitivity to low and high frequencies roll off as the volume lowers. This is known as Fletcher-Munson Loudness Cruves (and hence why we have a loudness button on stereo systems to help reequalize that curve. (And why THX has a re-eq button from theater tracks))
Since "Chuffs" dominate the entire spectrum (mostly white noise), then maybe all low frequencies can be directed to a woofer. Maybe someday when railcomm is approved by NMRA we can sync woofer sounds to a small subwoofer on channel 1 of railcomm. (as the low freqs are omni directional).
I am not at all concerned with getting correct scale sound. What I want to do is convey the impact of a diesel locomotive rumbling by. I think a little exaggeration without shaking the foam off my trees would be very nice.
There is another side to this.. if you had a truly integrated sound system.. then not only would you get better bass but you could get better highs since your engine based speaker would not need to handle the lower frequency. That should be as easy as uploading samples to the sound decoder that only contain 600+ hz (or whatever your cutoff is)
As for hardware.. while a nice "Black box" solution by a DCC vendor would be sweet.. I would be plently happy with software that reads the DCC signals with PC through an interface and simply plays the appropriate sound files through the computers software mixer which should be plenty fine for the job. Then you can shop around or use whatever you want. I have a relatively inexpensive system on my PC that still put out plenty of bass for the needs of this application.
The nice thing about a PC (or mac or whatever) soultion is that you would not need any fancy hardware other then the PC itself. Further.. you could tune effects like reverb.. and perhaps in the future block and occupancy detection could further enhance the system.
Actually.. as a poor man's sound solution.. I wouldnt mind just having software to simulate sound decoders right now. My layout is small enough that if all the sound came from one spot.. or maybe high/mid spread over speakers in the corners.. it would still be plenty cool.
The kids would love it!
Chris
ChrisNH wrote: Actually.. as a poor man's sound solution.. I wouldnt mind just having software to simulate sound decoders right now. My layout is small enough that if all the sound came from one spot.. or maybe high/mid spread over speakers in the corners.. it would still be plenty cool.
A simple re-eq'd 5" port tuned tunnel sub will do the job quite nicely from 80Hz->40Hz (Audible Threshold). This is actually somewhat overkill. These speakers can be had <$50. - Digital Griffen
'RUMBLE' frequencies are all below 40 Hz. a "5" tunnel ported sub (woofer) won't go this low - sorry, and I've got the Sound Generator equipment (and calibrated microphones) to prove it.
Most Hi Fi Loudspeakers roll-off 18db per 8va below 100Hz, $250 JBLs below 70HZ, AND $350 acoustic suspensions below 50Hz. (Problem with A.R.'s was low efficiency). A 14" speaker system 'flat' to 70 hz is down -18db @35Hz!
D.G. I do not know the Resonant frequecy of your VELODYNE sub woofer but I suspect it will prove that providing Bass is not cheap. The lowest note on a double Bass Viol is 43 Hz and 'rumble' is below that.
The problem here is bass extension.
Extending the excursion of the cone - As per Edg. Villcher (AR), Henry Kloss (KLH), Rudy Bozak (Bozak Inc.), 1. 'Extending the 'throw' is a viable way of lowering bass fundimentals on a 12" speaker*, but on a 1" speaker is too small to be of consequence. Most HO speakers resonant frequencies are HIGHER than 500Hz. 'Tony's are 750Hz.
2. Add multiple redundant cones ... Passive Radiators??? JBL'S 10" & 12" soldl for $100 ea.
3. Roll off the middle/high range - 'fool the ear'? - but this eliminates the middle/highs.
4. Increase cone size. (Not possible) with HO or 'N'.
Someone on this forum (not you) mentioned a JBL 18" woofer. To my knowledge JBL has never had an 18" speaker. Electro Voice made one in the 50's for their '$1800 'Patrician'.
As I stated before, getting bass is not cheap.
* didn't design or make 15" because of difficulty of 'damping' the motion.