Re: Sound in LL P2K engines

That's fine, but allow me to add an adjunct to Commandment 12. Thou shalt ADVERTISE outside thy home domain.

Always remember, Who tooteth not his own horn, getteth not his horn tooted.

And no, I do not have a recording of an Alco 539T engine nor do I have a recording of the horns. I CAN recreate the horn sound synthetically, but I do not have the capability to recreate the sound of the diesel engine. It most certainly can be done, but requires equipment much more sophisticated and expensive than I can muster.

If there was good sound available of the 539 engine and the 244 engine, that would be all I needed to have sound for the GM&O. I suspect that you have numerous 244s and

251s running around in downside-up land that you could record. I would not be surprised if you had some 539s as well. I could record a 539 at a museum if I could get them to run one for me, so could a sound card manufacturer.

I would like to play around with a programmable processor to see if, for example, you could have two sound bytes of the same recording played back with a 250 microsecond delay on one of them and get the illusion of twin engines through a single speaker. Maybe a human cannot detect 250 microseconds, so you might need longer, or that might be too long and need to be shortened.

Anyway, it sounds like the Oz blokes are on the right track, they just need to keep developing the user options. It certainly sounds better than the crap that the NA guys are flogging off on us.

..........F>

Yes, plenty of 251s still running. 244s are a little harder to come by, there are only two surviving in NSW that I know of, one is a runner of sorts, and the other soon will be. The other 40 is rumoured to be a stripped and empty shell.

I don't think so - weren't they used in the early S-series switchers?

I've been thinking about some sort of tension feedback from the drawbar.

I've been getting better with the on-off timing, but still far away from the PFM whistle lever(s).

Yeah, I still have PFM/GSB sound in my G-scale 1:20.5, that in conjunction with the 4" speakers in the tenders kinda warps my opinion of things.

I'm working with several different baffle designs to get the frequency response lower. A set of Bose IVs are one of my inspirations. I've also got a crossover design (5000kHz) that allows the use of a tweeter to pick up clearer highs, freeing the 1.25 inch "woofer" to work on the lows it can do. Unfortunately all of these consume an entire unpowered "B" unit. As I perfect them I hope to combine the two ideas.

Another thing I have just discovered is how much the performance of the capacitors helps the sound quality. I always knew this, I just didn't apply it to the trains until I started building the cross overs - Duh. Replacing the cheapo capacitors that come with some of the units with audiophile grade "Multicap Metallized Polypropylene" or similar capactiors really helps. Of course that assumes you don't mind spending another $28-$40 per-capacitor on the project. And even then I don't know the sound produced by the unit is good enough to justify this.

I am going to do some more experimenting with more mid-range capacitors such as:

see where the cost/performance curve balances. Has anyone else had any experience with this brand of capacitors?

Hmmm. Can't find anything to hand, but I have an idea either West Aussie or South Aus had an early ALCO switcher, but it's just a feeling of something from an old book I read many moons ago. What did the NSW 73 class diesel-hydrauilics have as a prime mover, Mark?

Steve

G'day Steve, the 73s have a medium-sized Caterpillar - after spending countless hours working on the things I cannot recall the model number!

Was it one of the Pilbara region iron ore railways that had a early Alco? Hamersley Iron had the demonstrator Alco C-415, perhaps they had something else before that? I'm fairly certain the croweaters didn't have anything other than EE powered shunting units...

All the best,

Mark.

Afirmative. The last year that I know of for new 539s in railway service is

1959. I suspect that some marine applications may have gone on a bit longer. I have no idea how many units continued outside the North American venue.

My point was, among other things, that the tension in the drawbar on a model is *NOT* the same, in proportion, as for the prototype.

The tension in the model's drawbar would be closely related to the current draw on the motor (except for drive line binds, etc.).

But the real point is that the dynamics of the model TRAIN are very different than the prototype. The model train will not 'load' the locomotive in anything like the same manner that the prototype train does. And I'm NOT referring to magnitude here, but mostly timing and oscillation factors.

For example, when a typical real train leaves a yard it takes the locomotive several miles to work the train up to speed. That's when the loco is working the hardest and has the greatest tension in the drawbar. On a model, however, the train can be brought up to speed in a few feet. Inertia is proportional to mass, and is near zero (a 'cube' function,

1/658,503 for HO to be exact) for the model train in comparison to the real train.

On the other hand, once the train is running at a steady speed, things change the other way. Friction dominates the model. For HO the friction (a 'square' function) is about 87 times greater in proportion than for the real train.

Thus real trains start hard, but pull easy. And model trains start easy but pull hard.

And that's just ONE aspect of what I'm talking about. Add in the dynamics of real railroad braking, the dynamics of a long train on several different grades at the same time, friction in curves, coupler slack, cushion underframes, and other factors.

Also, the gearing in the model has considerable to do with it. Spur and helical gears can COAST, worm gears (single lead at least) cannot. This greatly affects the load on the model's motor when coasting or going downhill. Add all these factors together, and you have a *VERY* different model to prototype 'applied forces' environment.

Dan Mitchell ==========

SleuthRaptorman wrote:

Yes, but don't forget the wonder of software. Since we know the physical characteristics and differences as you explained, they can be programmed into the software of the decoder/sound unit to adjust accordingly. Combine that with other feedback devices such as a grade detector, speed, and then store a little history from each sensor and the adjustments should be able to get pretty close. But getting back to the thread topic almost anything load related is closer than using "voltage to the motor".

Further thought, extrapolate this into the future of DCC. If the dispatching software knows the composition of the train and what the "pretend" load of each car is and hence its "pretend" weight, it could then in turn adjust the speed curve of the locomotive to simulate the mass of the train. This same adjustment signal could be used to adjust the sound also.

Read my post. I **DID** say the effects would have to be SYNTHESIZED!

My point being that simply sensing the load in the drawbar, or the current level, as suggested, will NOT give the desired effect.

Dan Mitchell ==========

SleuthRaptorman wrote:

Ah yes, I missed that. The second clause of "placed totally under the control of the operator" overwhelmed my attention.