Message Search is your friend

My bad...

For being lazy!

Many thanks for the pointer!

A few more questions

Dear Prof,

Allow me to ask a few more questions after a night of research. I read through pretty much everything and am a little confused.

After looking at what decoders are available and a reasonable price, two have come to mind; Digitrax Soundbug and the MRC diesel sounds, both just shy of £40. The MRC does multiple engine types which can selected via a CV, even more attractive.

Looking at the blog of Lance (great read btw), this is purely sending the signal to a headphone without any method of altering the volume other the programming this via the CV, either on the fly or via prog track.

I am looking not to send it to a headphone but to make it line level, for example 250Mv to be fed into an amp with an input impedance of 100Ohm. I might use headphones sometimes but can plug these into the amp (though I  might build a socket on the fascia as well).

Assuming the decoders will send a signal that can be varied up till about 500Mv, into 8Ohm, how do I go about matching this?

I can buy an Audio Output transformer similar to the one used by Lance, is that the answer? Or due to the fact I don't go to a headphone, can I go straight from the speaker output wires into the inputs of the amp?

Your insight is appreciated!

Many thanks from a snowy England,

- Henk

Not all headphones are _Passive_ devices...

Dear Henk,

In order of appearance:

- in the "stationary speaker" version, IIRC Lance used PC multimedia speakers which do indeed have a built-in power-amp and active gain(volume) control.

In the final Sennheiser RS170 wireless headphone system, the Sennheisers themselves have an active gain(volume) control on the headphones themselves.

http://en-de.sennheiser.com/wireless-headphones-home-audio-rs-170

Ergo, the statement "no volume control other than Master Volume CV" is incorrect in these examples.

- Irrespective of what the target _Active_ speaker/headphone device is,
(Active = has a Line-level input, driving circuitry),

you are entirely correct. The output of the TSU (or other decoder) is generally an amplified output.

It is intended to drive _Passive_ speaker drivers (IE directly-drive electromagnets with paper cones attached ).

IE it is _NOT_ reccomended to simply "connect the (amplified speaker-level signal) purple wires" to any form of _line_level_ input!

(Cue nasty horribly distorted noises, and strong possibility of releasing the magic-smoke that makes all electronic things work... :-( ).

This is where the 2 options noted in the previously-noted posts come in:

Option 1 : Radioshack impedance-matching transformer
  http://www.radioshack.com/product/index.jsp?productId=2103254

http://www.lancemindheim.com/20AprilSound%20011L.jpg

or

Option 2 : Soundtraxx Tech Note #11 : Resistor divider "Pad"
http://www.soundtraxx.com/manuals/Tech_note%2011.pdf

Applying either of these solutions will provide a near-enough-for-model-railroading "Line Level" output,
which you can then adapt and connect to whatever amplifier/"power speaker"/active-headphone system you choose. (Just because Lance ended up using the Sennheiser wireless headphones does not mean the solution is _only_ suitable for use with wireless headphones)

NB that the RS170s are _wireless_ headphones, IE the basestation that recieves the "audio signal" is an active device, with an active Line-level input, just like the amp you are thinking of using...

(Possible mis-understanding in the use of the term "headphones".

"Normal" headphones are just passive speaker drivers, which require some form of amplified signal.

http://www.djwarehouse.com.au/media/catalog/product/cache/1/image/600x/5e06319eda06f020e43594a9c230972d/p/i/pioneer_hdj_1000_headphones_3.jpg

"Wireless" headphones a la Lance's Sennheiser RS170s

http://www.frontroomcinema.com/wp-content/uploads/2012/02/sennheiser-rs-170-appart-550x550.jpg

http://www.lancemindheim.com/20AprilSound%20013.jpg

use and present an _active_ input on the seperate basestation, requiring only line-level input signal,
just like any other "amplified speaker system".

The signal is then transmitted wirelessly from basestation to the "headphone unit",
which includes it's own battery/power supply and "power amp", 
thus operating the speaker-drivers on either side of the wearer's head... ) 

For Reference, a step-by-step walkthru

https://forum.mrhmag.com/post/could-a-blue-tooth-application-be-an-answer-12191786

Hope this helps!

Happy Modelling,
Aim to Improve,
Prof Klyzlr

Sound

A sound decoder is only as good as the sounds programmed into it.  Soundtraxx Tsunami seems to have the best sound files available at this point. 

MRC sounds aren't all that good, and I wouldn't want you to sink money into something you may end up not wanting. 

Digitrax sounds I believe are user-created, then uploaded to the Digitrax site, then programmed into the Soundbug with a PR3 (extra cash outlay, if you don't already have one).  Something to keep in mind. 

QSI and ESU Loksound also have downloadable sounds, with proprietary devices to load the sound into the decoders. 

I agree with the futility of onboard sound in N scale.  Just not enough room for a speaker and baffling in a powered engine to make it sound decent. 

Stationary Sound

Why go to all that bother. Why not consider Surroundtraxx. It works extemely well.

Doug

"signal path limitations", Close-coupling speaker drivers...

Dear Mike,

The Digitraxx Soundbug and equivalents come with a default pair of "steam" and "diesel" (GP38?) projects in each decoder. It's a CV change away to swap between the projects. Agreed, Digi do also allow User-loading, although with only 8kHz 8bit resolution, the audio performance is a long-way from "CD spec" (not that Digi ever promoted it as such).

NB that most all "recorded loco sounds" used in onboard sound decoders were and are originally recorded/"captured" at a minimum of 44.1kHz 16bit (Many are recorded at higher specs! 96kHz 24bit anyone?). As such, the original source files are _far_ in excess of the resolution of the actual playback circuitry and storage limitations of the decoder.

Ergo "... only as good as the sounds programmed into it..." may well be true in terms of the _tone_ of the sound, (Is the recording of #1136 with the healthy turbo, or #1137 with the blown turbo? What mic was used? Where was the mic positioned? Can someone get up on the deck and stop that engine-door rattling?),

but doesn't hold in terms of actual digital audio playback "weakest link in the chain" signal path.
(like the old Audio Guys dictum goes
"...You can make a Good Signal go Bad,
but it's almost impossible to make a Bad signal come Good...")

In terms of sonic performance, I'd agree, starting with a Tsunami is starting with "as good as it currently gets",
with the full knowledge that when held up against "normal domestic digital audio" (CD, iPod, iTunes MP3), even that "as good as it gets" is not actually all that good...

In Lance M's case, replacing the onboard loco speaker with the Senni's was a earthshattering improvement.

Then his son plugged in a bog-standard iPod into the same headphones,
and the decoder's limited performance was shown up for what it actually is.
(Lance reported to the LayoutSound group that his "earthshattering" experience was somehow not so earthshattering any more...)

In terms of onboard in N scale, a close-coupled array of smaller speaker drivers can indeed get down into the 100-200Hz range, which is what passes for "bass" in most top-20 music tunes...
(the music producers only _wish_ they had the ability to actually audition and _design_ their sound on the _actual_ speaker system the song would be played back on. We as modellers have this capability right in front of us, but how many modellers use it?)

Just gotta get the space (an EMD FB or hicube boxcar are obvious options) to get the drivers in, and enough venting for the soundwaves to "push on the air" outside the shell...

Happy Modelling,
Aim to Improve,
Prof Klyzlr

 I am not a sound engineer

I am not a sound engineer and don't even play one on the internet.  I am also an n-scaler who was interested in sound, but not in the decoders.  The selection and speakers in locos never satisfied me.  The sound were not "powerfull" enough.  I was not looking for room shaking sound.  I imagine myself in the cab of the engine as I run my trains, so I wanted something loud enough that it gave me that feeling and engine decoders never had that effect for me.

I have looked down the road you are exploring and actually made some head way.  I built a box to house 6 tsunami decoders for the various engines I was running.  I then pumped the sound through a cheap amp to some passive speakers and was very happy with the sound.  The fall down part for me was the cumbersomeness of consisting the stationary tsunamis with the engines I was running.  Also, if I had two of any one type of engines running around, I only had one of that type of decoder, so one was silent.  The Tsunamis cost me about $95 each.  I sold them all and purchased a SurroundTraxx (DSP-80) for about the same price as the ones I sold..  I am much happier!

The advantage of the DSP-80 for me is that it holds configurations for 99 engines and can make sound come from 6 of them at any one time.  The consisting is automatic as long as the soft-decoders in the device are the same address as the loco you are running... it pulls them from the virtual roundhouse and starts making noise as soon as it sees the loco address on loconet.  Digitrax loconet is required.  I do not use the built in engine following aspects of the device and am still happy.  You can do with if you want to deploy Digitrax transponding on your layout... It is a bit of work, but others are able to do it. 

For me the DSP-80 is like a box of 99 sound decoders that automatically work for my engines.  Since I don;t care about the train following abilities, I just set out soem speakers around my layout and an $15 amp handles the sound fine.    I think it is something you  might want to consider.

-SeanM