I spent some time down on the layout this morning, so let me update my findings.
Regarding the volume itself, I played with the speaker and the baffle. The speaker fits snuggly in the speaker mount, but it is not tightly sealed against the baffle. When Intermountain sent the 3-part kit (decoder, speaker, and baffle), the baffle was referred to on the invoice as a "QSI style" baffle. It is identical to the baffle found in my other Intermountain F3A loco with factory installed sound. Either the speaker is slightly too big to fit inside the baffle or the way the wires are soldered to the speaker prevents a tight, sealed fit. That said, I can hold the assembly together in my fingers for a tighter fit and produce higher volume.
The bigger issue, though, is the sudden loss of sound in certain instances. When the sound is lost, all of the lights (headlight, Mars light, number boards) remain on and the loco continues to run. But all of the sound (whistle, horn, prime mover) cut off, and the only way to restore sound is to lift the wheels off of one rail and then back down on the rail. That loss of sound can happen anywhere on the layout, and it only happens when I press the F2 button. When I apply a different button for short beeps, there is no loss of power.
One curious thing is that if the loco is sitting still and not moving, the sounds remain on no matter what. So, even if I press the F2 button and leave the horn blaring, the sound never cuts out, only when the loco is in motion.
Mark mentioned the possibility of a corrupt sound file. If this were the case on this decoder, would the situation that I am experiencing be the consequence?
Clearly, the pressing of the F2 button is causing a complete loss of sound when the loco is in motion.
What the heck is going on here?
Rich
Alton Junction
I went down to measure the ohms. Before doing so, however, I noticed markings stamped on the back of the speaker. It reads 4 ohms, 2 watts
Well, while that doesn't explain your other issues, Rich, it is an important question answered. That said, it might be worth measuring anyhow - just for confirmation. Not like things have been incorrectly labeled before...
Tom
https://tstage9.wixsite.com/nyc-modeling
Time...It marches on...without ever turning around to see if anyone is even keeping in step.
Tom, I got a reading of 2.6 ohms.
Sounds like it's in the ballpark, Rich. Now onto solving the other issues...
I was beginning to think last night that it was the speaker, but I am no longer so inclined. Could there be a defect on the decoder? I hope to hear more from Mark about the possibility of a corrupt sound file. What else is there besides the sound file, speaker, and/or decoder to explain this?
Hmm, yes I agree, I think Mark had it right with the corrupt sound file.
Call Intermountain again and have them re-program the decoder with the new sound file.
Michael
CEO- Mile-HI-RailroadPrototype: D&RGW Moffat Line 1989
If you are getting proper motion, lights and sound, I'd say the functionality of the decoder is good, but something is corrupt in the file causing the problem.
Personally, I don't like using the OEM files as they are usually configured differently than the way ESU does their regular files. If it's an option, I'd have them load file number 73436 which is Loksound's 567 16 cylinder non-turbo.
You might even consider sending it to one of the members here to load the file as you may get a quicker turn-around.
Mark.
¡ uʍop ǝpısdn sı ǝɹnʇɐuƃıs ʎɯ 'dlǝɥ
Mark R. If you are getting proper motion, lights and sound, I'd say the functionality of the decoder is good, but something is corrupt in the file causing the problem. Personally, I don't like using the OEM files as they are usually configured differently than the way ESU does their regular files. If it's an option, I'd have them load file number 73436 which is Loksound's 567 16 cylinder non-turbo.
In other words, would a corrupt sound file cause the behavior that my loco is exhibiting?
A corrupt file can do (or not do) anything imaginable. All it takes is one bit of bad information and something will not function correctly.
Based on what your decoder is exhibiting, functionality is ok - motor runs, lights work and sounds work, so all hardware is good. The decoder DOING something wrong then, would be a software issue.
A small power glitch / connection during the download may not be enough to abort the download but may be enough to miss even a fragment of important information - and that's all it takes.
In cases like this, my first inclination to always try a fresh download. If THAT download does the same thing, I try a different file altogether that I know to be good. There have been at times, files posted on the ESU site that have had bugs in them as well. I've never used the file in question to verify if there is in fact a bug in it - not too many people use the OEM files for downloading as the regular ESU files are configured to more familiar configurations to us Loksound users.
Mark has some very valid points there.
I remember one of my early installs in a Athearn Genesis F-7 using aLoksound Select Direct. I was dismayed when the locomotive stuttered and sputtered on her maiden voyage. I thought the decoder was at fault and when I swapped it with a second decoder I had on hand I got the very same symptoms .
Well, after doing some more diagnosis I found that some of those Genesis engines were notorious for poor power pickup so I stripped and cleaned all the wheels, bronze axle frames and then soldered the feed wires where Athearn used the cheap plastic tabs on the side frame pickups.
It got better but what it really needed was a good running-in period. I'm not implying that this is your solution, Rich, but I can sympathize with your frustration.
On the plus side, going with ESU is a wise choice! I invested in a Lokprogrammer about a year ago. For me, the learning curve is a bit steep but, keeping in mind that you really can't screw anything up that can't be un-done, I continue to play with it and learn more of its capabilities. It is an amazing bit of hardware and I wish other manufacturers would have embraced the technology. I sure would like to replace the Paragon3 sound files in the BLI L1 that sounds like a clogged vacuum cleaner
I have a couple of Loksound Selects that had an EMD F-40PH file loaded on them that I was going to install in a pair of Kato's that I have. Instead, after getting the Rapido ones (with Loksound on-board, thanks Jason) I reloaded an Alco 244 sound file in them and used them in a couple of PAs. I really didn't know what I was doing but I followed the prompts for installing the file and everything went flawlessly.
Now, I'd like to learn how to add one of my own recordings. I have always wanted to have a communicating whistle cab signal. It's a little "tweet" peanut whistle that would sound when the conductor would pull the signal valve back in the train.
I can't believe any of the sound manufacturers have never offered this sound. We can get chicken and circus caliope music, radio chatter and hot box detectors in steam locomotives, but not a signal whistle!
I have one that I grabbed off an old NYC E-8 that was on a scrap line in Hollidaysburg, PA so I can make the accurate recording. I just have to learn how to add the file to the "project" as ESU calls it.
Well, I rambled on too much, but I hope you get your C&EI F unit sounding better, soon!
Happy Holidays, Ed
I have figured out one of my two problems, and that is the low volume. The speaker does not fit tightly into the baffle. Actually, it fits snugly, but it does not snap into the baffle. It simply slips down into the speaker housing below the baffle. Curiously, it fits snugly into the speaker housing, but then there is space between the speaker and the baffle. In order to get a tight fit, the speaker would need to be thicker from front to back. Very frustrating.
Rich:
Re the speaker not snapping into the baffle, perhaps Intermountain shipped you some wrong parts. I can't imagine that they are sending product out en masse without it working properly.
If you can't get satisfaction from Intermountain (which would really shock me given my own very positive experiences with them) then you can make your own baffle very easily by cutting off the bottom of a suitably sized prescription pill bottle. I have done this a few times and it worked great. I used Goo to attach the baffle to the speaker.
However, that still won't solve the problem if the software files are messed up. I actually think you have two things going on here.
Dave
I'm just a dude with a bad back having a lot of fun with model trains, and finally building a layout!
Rich,
Nothing says you have to use the speaker/housing they sent with the kit.
I have an assortment of speakers I keep on hand for various hood or cab unit diesels and steam applications.
Some of my best sounding speakers are from:
http://www.railmasterhobbies.com/Speakers.htm
Get the largest enclosure that will fit, even if you have to mill or file the frame a little.
All my Intermountain Fs came with factory installed sound so I never messed with trying different speakers but many of the BLI and Genesis Es & Fs and Alco PAs and FAs I have used various Railmaster speakers and enclosures.
The example above shows a Genesis chassis and I believe the 28mm enclosure. The top corners edges of the speaker housing were belt sanded to match the contour of the roof for a better fit and I used 3M tacky stuff for a vibration-free installation.
Many of the OEM speakers and even the ESU ones that come with the decoders I set aside and use only as a last resort. I look for a speaker that has a fabric surround or is sold as a hi-bass and try to use this as my first choice. Many of the BLI and Proto Walthers speakers are very low-budget types.
I use an old Sansa Clip MP3 player loaded with various railroad sounds to check out the speakers and enclosures on the bench at various sound levels before committing to installing it, then only to find it is not performing to my satisfaction.
Good Luck, Ed
hon30critter Rich: Re the speaker not snapping into the baffle, perhaps Intermountain shipped you some wrong parts. I can't imagine that they are sending product out en masse without it working properly. If you can't get satisfaction from Intermountain (which would really shock me given my own very positive experiences with them) then you can make your own baffle very easily by cutting off the bottom of a suitably sized prescription pill bottle. I have done this a few times and it worked great. I used Goo to attach the baffle to the speaker. However, that still won't solve the problem if the software files are messed up. I actually think you have two things going on here. Dave
I am confident that whatever the problem is, Intermountain will address it and fix it. In all fairness to Intermountain, I have not yet contacted them about the problem. Because of the holidays, they have been closed since Wednesday, so I will call them on Monday to discuss the issues.
Intermountain mentions on its web site that its HO scale F units use a 1.10" diameter speaker. And, on my invoice, the technician noted that the baffle is a "QSI style" baffle. The baffle that Intermountain sent to me is identical in shape and size to the baffle on my other Intermountain F3A with a factory installed sound syste. That factory installed sound system is on an older unit when Intermountain was using a QSI sound decoder. The only difference that I can see is the thickness of the speaker. The speaker is thicker on my factory installed sound system, so it is pressed into place between the baffle and the speaker mount by two screws that hold the speaker assembly in place.
I have thought about using some form of adhesive to hold the speaker to the baffle. The speaker does fit perfectly into the baffle but there is no way to secure the back of the speaker to the baffle. It does not snap or twist into place. Also, the front of the speaker drops perfectly into place on the speaker mount which is part of the chassis. The problem is that the space between the baffle and the chassis mount is deeper than the thickness of the speaker, so the speaker falls out of the baffle.
Mark R. A corrupt file can do (or not do) anything imaginable. All it takes is one bit of bad information and something will not function correctly. Based on what your decoder is exhibiting, functionality is ok - motor runs, lights work and sounds work, so all hardware is good. The decoder DOING something wrong then, would be a software issue. A small power glitch / connection during the download may not be enough to abort the download but may be enough to miss even a fragment of important information - and that's all it takes. In cases like this, my first inclination to always try a fresh download. If THAT download does the same thing, I try a different file altogether that I know to be good. There have been at times, files posted on the ESU site that have had bugs in them as well. I've never used the file in question to verify if there is in fact a bug in it - not too many people use the OEM files for downloading as the regular ESU files are configured to more familiar configurations to us Loksound users. Mark.
My concern is that the sound cutoouts out entirely after a second or two when the locomotive is in motion.
It makes me wonder if the sound file is corrupt, as you suggest, or if the decoder is faulty in that, perhaps, the current draw by the horn is more than the decoder can handle, so the sound system shuts down. Is that a possibility?
Also, I still have to wonder if something is wrong with the speaker. That's part of my problem. There seems to be three possible causes, the sound file, the speaker, and/or the decoder.
I was reading through the Intermountain web site section on sound decoders. I found the following excerpt particularly interesting.
Rich, you can use a small bead of silicone caulk to seal the speaker in the baffle.
I've had that problem with loksounds o nmy VO-1000 by bowser. Their amps sometimes burn out.
But it's sometimes best to make sure you have the right speaker with your model Rich. The older ones required a much higher resistence speaker. They weren't nearly as loud.
Don - Specializing in layout DC->DCC conversions
Modeling C&O transition era and steel industries There's Nothing Like Big Steam!
4 ohms is nominal. Most require 8 ohms nominal. So 2.6 is way too low. You'll likely overheat the amplifier and cause it to shut off. Something is wrong with the speaker. Try swapping it.
3M double sided sticky tape will take of your snug fit issue. That or hot glue...IF YOU ARE CAREFUL With it.
You can also stuff the back side of the baffle with polyfill or loose fiber glass. (Batting from a craft store) it helps absorb back waves which cancel out your sound. But there's a risk it could lower your overall volume.
DigitalGriffin 4 ohms is nominal. Most require 8 ohms nominal. So 2.6 is way too low. You'll likely overheat the amplifier and cause it to shut off. Something is wrong with the speaker. Try swapping it.
DG,
The newer Loksound decoders (i.e V4.0 & Select) take 4 ohm speakers. According to the earlier link I posted about how to test the impedance of the disconnected speaker, 2.6 ohm is about what you should expect from a 4 ohm speaker.
FWIW...
Just popping in to offer a comment:
if you're having problems both with low sound and the sound cutting out, maybe one of the wires to the speaker is poorly soldered, or the soldering didn't take as well as it should have.. Although, it would probably be more of an intermittent problem than maybe what you are describing.
- Douglas
DigitalGriffin I've had that problem with loksounds on my VO-1000 by bowser. Their amps sometimes burn out.
I've had that problem with loksounds on my VO-1000 by bowser. Their amps sometimes burn out.
tstage DG, According to the earlier link I posted about how to test the impedance of the disconnected speaker, 2.6 ohm is about what you should expect from a 4 ohm speaker.
According to the earlier link I posted about how to test the impedance of the disconnected speaker, 2.6 ohm is about what you should expect from a 4 ohm speaker.
I respectfully disagree Tom
4Ohm may be okay for the Loksound Select, but 2.6Ohms is way too low. Nominal impedence is with a constant voltage. The only time resistence drops is when there is a constantly changing voltage. Voice coils are essentially inductors, so when you try to change their current, by changing the voltage the change the resistence and voltage as resistence. (Inductors try to resist current change)
8Ohms nominal on this speaker comes out 8.1 to 7.9Ohms on the meter.
So, when the author of the link (see bottom of pg. 1 of this thread) says an "8-ohm speaker will read around 6.2 on the meter. This is completely normal." That's completely bogus?
When you are measuring resistance with a digital meter, you are measuring resistance to DC current. Speaker ratings are AC resistance - which IS different. Kind of the same thing as trying to measure DCC track voltage with your meter set to AC .... it's similar, but not the same.
Maybe this will help ....
The multimeter looks at resistance, not impedance. Impedance is the opposition to the passage of alternating current and varies with frequency. Resistance is the opposition to the passage of direct current. Resistance is normally lower than nominal impedance. For instance, an 8-ohm nominal impedance cone speaker would typically have a (DC) resistance around 6 ohm. See any DAS spec sheet for an impedance curve. Dedicated impedance meters for audio use exist; they typically use a sinewave generator to measure impedance at 1 kHz, so you need to look at the speaker`s impedance chart to see what value you should expect at that frequency.
Don,
I've looked at a few more sites since your reply and there seems to be a general consensus that an 8 ohm speaker should read ~6.2 ohms and a 4 ohm should read ~3.2 ohm:
Checking Speaker Impedance With a Multimeter
Resistance vs Impedance
Are they both wrong? That said, 2.6 ohm may indeed be too low on Rich's speaker.
[Edit: Mark's video is the exact same one in the first link above]
DigitalGriffinI would say rich is overheating the amp causing shutdown. Its potentially damaged now. Only way to find out is swap the speaker.