Speaker Problem?

[EL3632]

Hi all, I just want to see if my train of thought is on the right track (no pun intended).

I have a Kato Kobo PA-1, with ESU Loksound installed. It runs well, and sounds good, too; or at least it does for the first ten minutes or so of running. The sound becomes all staticky after around 15 minutes, and I am unsure of why, I have three more of these, and all are fine for any amount of time I have run them so far. Doing some research, I have found that the issue is most likely the speaker itself or the wires connecting to the speaker. I am leaning toward the speaker with the same type of wires running to three more engines with no issues.

Givens and druthers: I reprogrammed the sound file from a 244 to a 251, as the loco I am modeling is a PA-4, rebuilt with a V12 251B. I also reprogrammed one of the others with this file and has no issues. Some advice I got was to lower the master volume and see if that solved the problem; it did not, and continued to start being staticky around 15 minutes in. The other 251 decoder was also lowered and shows no issues.
 
Here is a video of the unit in question, it had been running around 15 minutes at this point:

Here is a video of the other 251 equipped unit, with no issues, also run around 15 minutes:

Not sure what to do at this point, but wanted to ask the experts before I tried tinkering with it and broke something expensive. I believe I have the skills required to replace the speaker if need be, but have never done a sound installation myself, so any tips and tricks would be appreciated.

[Mike C]

  I'd say its a bad speaker , unless there is something touching the cone .....Mike

[peteski]

If it takes that long for the problem to show up, to me it sounds like possible overheating of some component on the decoder (like the audio amp. chip).  Maybe you could remove the shell and try running it "naked"?

[tehachapifan]

I'm no sound expert by any stretch, but how much did you lower the master volume and do you know the wattage rating of the speaker? When I first started doing my own installs, I had similar behavior with one install prior to the decoder failing entirely. I did not realize at the time that the default master volume setting with many/most/all? ESU sound files (192 I believe) is too high for many speakers that are in a size suitable for N scale. While I haven't found a solid guide or formula to go off of (wish there was one), I usually cut the master volume roughly in half and then slowly creep it up until it sound about right and there's no distortion that starts occurring like you described. I'm currently using speakers that are 0.5w (0.7w max) up to 0.8w (1w max). Your mileage may vary.

....I've been wishing for a while that there was more clear-cut documentation on all this. It isn't even clear, at least the last time I checked, on how many watts ESU decoders put out for the speakers. I've seen anywhere from 1 to 3 watts in at least some of the past documentation. Hard to determine a max volume number without that output figure when, for example, you're using a 0.7w speaker....unless I'm missing something. There's always been a lot of emphasis on using 8 ohm speakers (or a range somewhere around there), but that's about it.

(Edit) One more thing....I also typically turn the prime mover volume and some other individual sounds way down while leaving the horn up. I find the default setting for the prime mover is typically far too high as well.

[Steveruger45]

Hmmm.  I’ve had this problem on an earlier install.  In my case it was a bad solder joint/speaker wire. I just desoldered amd renewed the suspect speaker wire. Actually I changed both speaker wires and resoldered while I was at it.  Never been a problem since.

[jdcolombo]

Instead of guessing, I'd go about a systematic troubleshooting.  You've actually done a lot of this already, by changing the file in the decoder (essentially reprogramming it) and finding no improvement.

My next step would be to re-solder the speaker wires, as Steveruger45 indicated.  It could be a cold solder joint.

If that doesn't work, then step 3 would be to replace the speaker.

If THAT doesn't work, then by process of elimination, it's a bad component on the decoder, and you will need to replace that, which is going to be a bit more complicated, particularly since this loco probably uses a LokSound Select Micro wired decoder that is no longer made.  But no need to worry about that until you have tried steps 2 and 3.

Take a photo of the unit with the shell off and post it here - it may help with additional troubleshooting down the road.

John C.

[peteski]

Also, since this is a custom factory-installed decoder (Kato Kobo), another possibility would be to return it to Kato for service.

[jdcolombo]

Also, since this is a custom factory-installed decoder (Kato Kobo), another possibility would be to return it to Kato for service.

But what fun would THAT be??????



John C.

[nightmare0331]

I'm not sure I really want to post this here but here goes....

I'm actually working on four of these that were returned for similar reasons (I'm guessing the OP bought one of the recent ATSF ones and used that since Kato didn't have any D&H units left when we got around to modifying PA1's).  After about 15 minutes they get staticky and the decoder gets quite hot.  I've seen this a few other times recently.

Previously before this run, we had done a bunch of DRGW PA1's with LokSound (select micros) and there was next to no issues with them.  So, logically, everything is exactly as before since the V5 micros won't fit (this is why the recent run of ATSF PA1s have select micros on them.)

When these decoders were manufactured, it was past the end of the Select Micro production span and these were reworked from another OEM's updated order that switched to V5's along with leftover spare parts.  I'm not sure if this has something to do with the issue as perhaps as parts ran out, bits got substituted or something. 

Now, troubleshooting down the line, I have seen some issues with speakers that are sealed having sound issues at different elevations.  IE, there is a complaint of a loco being extremely quiet and distorted.  We get it back here and it works fine.  I actually had a customer who mentioned this might be a possability as he was at higher elevation.  He took it to his other place and it worked fine.  Off of that, we found that in some cases, drilling a small hole (like #78 or so) in the base of the speaker housing/baffle (not through the speaker itself) seems to have done the trick.  Others, the speakers appear to be shot after they go static.  I'm not sure if the speaker is backfeeding and overdriving the amplifier on the decoder.  My other thought was perhaps when they were programmed the full sized select settings were used so the amplifier settings were different.  Certainly just a possibility.   

I tried reloading the stock select file I had used countless times (75839 alco goodwin) with a minor decrease in master volume and I have one chugging away on a test track along with two previous counterparts.  With the new speaker (w/ hole drilled) and new sound file the decoder is still a little warm but not hot like the prevous two.  I'm also taping the decoder a bit different now so it's not just holding right against the speaker.  I doubt that is the issue, but I'm not ruling anything out.

What is maddening about this is it is completely inconstant and also doesn't manifest itself until the units have been run for some time.  Certainly everything gets tested that gets built, but not for 20 minutes, so it's actually quite difficult to catch something like this.

For the TLDR crowd, what I *think* appears to be fixing the issue is:

speaker- drill a small hole
sound file- update to 75839 or any known good select micro file
tape decoder a bit differently
if speaker still sounds bad, replace  (of course, if someone is experiencing this issue I'll make sure a replacement speaker gets sent)

These are just some of my observations as I'm still trying to come up with a more workable solution.  Maybe I'm missing something else here.

Never a dull moment!

Enjoy!

Kelley.
www.dufordmodelworks.com

[EL3632]

Thanks for the replies. To address some of the things mentioned:

(I'm guessing the OP bought one of the recent ATSF ones and used that since Kato didn't have any D&H units left when we got around to modifying PA1's).

Correct.

Previously before this run, we had done a bunch of DRGW PA1's with LokSound (select micros) and there was next to no issues with them.  So, logically, everything is exactly as before since the V5 micros won't fit (this is why the recent run of ATSF PA1s have select micros on them.)

The other unit in the video was a DRGW unit originally, good to know. The other two I have (no issues, with the original 244 sound file) are ATSF as well.

Now, troubleshooting down the line, I have seen some issues with speakers that are sealed having sound issues at different elevations.  IE, there is a complaint of a loco being extremely quiet and distorted.  We get it back here and it works fine.  I actually had a customer who mentioned this might be a possability as he was at higher elevation.  He took it to his other place and it worked fine.

I am at sea level, around 10 miles from the Atlantic Ocean, if that means anything. When I used sound at the highest elevation I have ran trains at, Altoona, with Atlas and Intermountain sound units, there was no noticeable difference.

...by changing the file in the decoder (essentially reprogramming it) and finding no improvement.

I changed the file as soon as I got the unit, so technically I did not do this yet, I will, though.

My next step would be to re-solder the speaker wires, as Steveruger45 indicated.  It could be a cold solder joint.

When troubleshooting this issue the first few times, simply moving the speaker/wire around fixed the issue, and then went back to static around 15 minutes later. Since I am test-running the loco now I will adjust this and report back.

Take a photo of the unit with the shell off and post it here - it may help with additional troubleshooting down the road.

Photos taken but on a different device, will post again after this one from that device with said photos.

I'm no sound expert by any stretch, but how much did you lower the master volume and do you know the wattage rating of the speaker?

Master volume is at 78.13% at 100 in Lokprogrammer, down from 150% or so (can't remember exactly); I do not know the wattage rating, nor do I have a means of obtaining that without figuring out what the speaker is and looking up it's stats.

When I first started doing my own installs, I had similar behavior with one install prior to the decoder failing entirely. I did not realize at the time that the default master volume setting with many/most/all? ESU sound files (192 I believe) is too high for many speakers that are in a size suitable for N scale.

I agree, but haven't gotten around to figuring out a sweet spot for me just yet. I primarily run at shows/with other people as well, so sometimes louder makes more sense for me. I will say that when I reprogrammed the sound from the 244 to the 251 in these, I was running a stock 244 and the reprogrammed 251 and I could not hear the 244 at all. I definitely believe that the sound should be lowered regardless, but the ex-DRGW unit was fine with the same volume, same speaker (I assume) and all.

Maybe you could remove the shell and try running it "naked"?

Did this while typing up the post and getting information. Started running the loco with shell off at 6:29 EST, at 6:32EST it had begun to get noticably staticky, without the shell on. Three minutes is not what I was expecting, but I will do more troubleshooting.
Edit: it is now 6:50PM EST and the unit is making more static sound than Loksound. This is consistent with how it behaved previously.

[EL3632]

Photos taken but on a different device, will post again after this one from that device with said photos.

Hello, from the other device!

[Steveruger45]

There is some great info in this thread, thanks Kelly, all good stuff to know.
@EL3632,yes pls do let us all know how the trouble shooting goes, but your noted “ When troubleshooting this issue the first few times, simply moving the speaker/wire around fixed the issue, and then went back to static around 15 minutes later.”  Would seem to me to give a good indication of a good place to start, that is on the speaker wires and solder joints for same. It does sound to me (pun intended) there is a bad speaker wire and /or solder joint if wiggling it cleared the problem at least for a short time.  Either that or it is a coincidence.  Good luck

[EL3632]

As a random update, since this seems interesting, at least to me: I put the other 251 unit on and turned the prime mover on, with the bad unit so staticky, the other unit completely drowned the static unit out, to the point it was inaudible (except for high pitched of static). They are on the same master volume, with no other differences between the two.

[Steveruger45]

I just see your post with photos, i don’t like those speaker wires positions. Make sure the insulation is intact and not damaged and keep them away from touching the frame and track power strip.

[EL3632]

I just see your post with photos, i don’t like those speaker wires positions. Make sure the insulation is intact and not damaged and keep them away from touching the frame and track power strip.

Agreed. They are not damaged, and are currently floating as I type this. Once/if the issue is fixed, I will reroute them as best I can to the middle of the frame away from the contact strip.

After messing around with the unit for around 30 minutes:
The decoder does not get hotter than the other unit's, so I think the decoder itself is fine.
Messing with the speaker wires does in fact fix the problem, in the time I was messing with the unit it happened twice and fixed the problem both times. I was able to trace it more specifically to the "front" wire, closest to the rest of the engine, by only messing with that wire and leaving the back wire alone (or as alone as I could leave them). The speaker connections on the decoder look solid, and do not have any visible issues.
I will try to resolder the wire, is just a quick in/out best to not damage the speaker or is it ok to leave in for a few seconds to ensure a good connection is made?