Weird noise from relatively new BLI NW-2

[joelm]

Hi Guys: Peteski is correct. When I got it, I immediately reduced the master volume. And it's always run fine at a relatively low volume until last week, when the buzzing and random sounds started.  The buzzing sound is pretty loud. I'll reset it tonight and let you know the results. Thanks. Joel

[tehachapifan]

Well Russ, the way I read the problem description was that Joel repeatedly ran that model, and every time it behaved the same way.  Sounds ok first, then 40 second later it starts the strange noise.

If something was to blow (amp or speaker), after the first time the sound would always be distorted.  Or am misinterpreting things?

That's how the one I had that failed behaved (good for about a minute), until it finally gave up. Difference is, he's describing a buzzing noise along with randomly triggered sounds, where mine degrading into garbled and crackling noises. My guess is the amp on mine was on the brink of failing for a while but hung on for a bit (perhaps because the objectionable sound would cause me to stop running on a number of occasions before it blew).

...as I was typing this response, I remembered that a few months ago I had another custom N scale install start to make the same distortions I described above. I felt that I had it turned down plenty far, so I decided to try replacing the speaker. I replaced the 8x12 speaker with a 9x16 and subsequently made a new enclosure as well. The sound has performed flawlessly since. Not exactly sure which part of the equation fixed the problem in this particular case, but I suspect the larger speaker handled the sound file I was using better. I think some files are more load on speakers and components than others, even with lowered volume. The reason I think this is possible is because I had a prior sound file loaded in this loco that had no problems. This was an ESU file that (like many ESU files) the prime mover sound seems to fade to a lower volume once the transition to a new speed step has concluded. I'm not crazy about this and went with a file that doesn't fade, but I think the fading files are easier on components perhaps.

[peteski]

. . .
Not exactly sure which part of the equation fixed the problem in this particular case, but I suspect the larger speaker handled the sound file I was using better. I think some files are more load on speakers and components than others, even with lowered volume. The reason I think this is possible is because I had a prior sound file loaded in this loco that had no problems.

From electrical standpoint, that makes no sense to me (especially "even with lower volume"), but if that's what works  for you who am I do dispute it.
And still, the description of Joel's problem does not fit that  scenario.  Repeatedly, the model starts of with clear sound, and the buzzing sound starts after a delay.  I don't believe that could be caused by a damaged speaker. The problem is likely on the decoder.  In either case, this is a factory installed DCC sound system, so if it is returned for service, they will replace whatever components are required.

[tehachapifan]

From electrical standpoint, that makes no sense to me (especially "even with lower volume"), but if that's what works  for you who am I do dispute it.
And still, the description of Joel's problem does not fit that  scenario.  Repeatedly, the model starts of with clear sound, and the buzzing sound starts after a delay.  I don't believe that could be caused by a damaged speaker. The problem is likely on the decoder.  In either case, this is a factory installed DCC sound system, so if it is returned for service, they will replace whatever components are required.

How about a problem arising with the amp on the decoder? Couldn't that cause some of the issues I describe? My understanding on this stuff is admittedly far less than expert, but what kind of issues and symptoms could arise from an amp that is not up to the task at hand because, say, the volume is set too high? The speakers we tend to use in N scale are typically rated anywhere from, what, 250mW up to 1W but an ESU decoder, for example, can put out 1W to maybe 3W (always hard to tell for sure from reading the docs) and, AFAIK, the only way to adjust that output is with volume. If the output is too high for the speaker or any other component (amp?), would a failure be instant without warning, or would it come on gradually (heating?) and first manifest itself as sound quality issues?

[peteski]

I mentioned that I believe the decoder is the problem. That absolutely includes the amplifier chip on the decoder, or any other component on the decoder.  Even if the amplifier chip is the problem, it  is not like Joel can replace the chip.

As far as "blowing"anything in the audio section, if the speaker impedance is within specs for the amplifier, then the amplifier chip is designed to handle even the loudest and most distorted signal it outputs to the speaker.  On the other hand, if the speaker was underrated with wattage (handle lower wattage than the amp can supply) then the speaker can "blow". Usually this causes the speaker coil to melt (creating open circuit), or the diaphragm suspension gets damaged. In either case, this would result in no sound, or badly distorted sound (at all times, not after 30 seconds of sounding clear).

[tehachapifan]

... if the speaker impedance is within specs for the amplifier, then the amplifier chip is designed to handle even the loudest and most distorted signal it outputs to the speaker.  On the other hand, if the speaker was underrated with wattage (handle lower wattage than the amp can supply) then the speaker can "blow"....

This is the crux of what I'm trying to get at.

I think that, with N scale (particularly narrow hood) diesel installs, speakers that are not up to the rated output of the decoder are typically used with regards to wattage (I am yet to find a 9x16mm or smaller speaker rated more than 1W and most are more like .5W to .7W). Now, again, no expert here, but I've only been able to find lots of info about what ohm speaker to use (typically 8ohm), but hardly a thing about what wattage it needs to be rated for. This seems to be an area that is not discussed much in any ESU docs I've found, but I recall seeing at one point that the decoder's sound output is as much as 2-3W. This leads me to believe that default volume settings on at least ESU decoders could be too high for a speaker rated at only 1W. I'm guessing similar situations could occur with other brands as well.

So, I guess what I was trying to articulate was, if the decoder is (over?)driving a speaker that is only rated at, say, 500mW, could this first manifest as some sort of distortion before something outright fails? And, if so, could this distortion begin after a minute or so of running?

[nickelplate759]

in general, distortion in an overdriven speaker wouldn't be a function of how long the speaker was being used.   Either the speaker is overdriven (which will sound awful) or it is not.  Normally a speaker won't get so hot as to start distorting badly, then cool off and be OK, and then repeat..

I agree with @peteski - this is almost certainly in the decoder, not the speaker.
Easily tested by hooking up a different speaker, by the way.  I'd recommend doing that.

[peteski]

So, I guess what I was trying to articulate was, if the decoder is (over?)driving a speaker that is only rated at, say, 500mW, could this first manifest as some sort of distortion before something outright fails?

That could be possible.

And, if so, could this distortion begin after a minute or so of running?

That part I have hard time with.  If this scenario happened once, then after that the buzzing was happening as soon as the loco was powered back and sound turned, then yes. But if the same sequence of working for 30 sends, then buzzing starts keeps repeating, then to me the chances of a damaged speaker are rather low.

I was also  going to mention that since this is BLIs model with their decoder, and their selected speaker, the speaker should be matched in ratings with the decoder's amp. But I'm not sure If I  want to use that reasoning, since we have seen factory models not designed very well.

I'm curious if a decoder reset cures the problem, or if the model is sent for service, what will BLI report.  It is also possible that BLI will just exchange it for a new model and keep this one for spare parts (they seem to get some of their spare parts that way).

[tehachapifan]

Well, FWIW, in both my experiences of sound distortion that I described earlier, the sound would start out fine and then begin to degrade/distort after a minute or so of running. If I shut off the loco and then tried again later, same thing would happen. I always also assumed that heating was a possible contributing factor.

[joelm]

Hi All: Finally got back to the loco this afternoon. Reset to Factory and reprogrammed. Seems to be working ok now. I will reset first in the future, before asking for help. But thanks for jumping in! Joel

[peteski]

Thanks for the follow-up Joel - those easy fixes are the best!