Hello all,
I am currently in the process of correcting some power pickup deficiencies on my spectrum 2-8-0. It is the version with the Soundtraxx TSU-BOE-A Rev B DCC sound decoder. The speaker wires broke off of the board while I was working on the power pick up. Does anyone know or have a photo of which speaker wire goes where or if it even matters?
If it's one speaker, the speaker polarity does not matter.
The + speaker terminal will have a small red dot near it or a + sign.
Usually the + speaker wire does have some special marking down it. It can be a simple stripe or line or a series of "+" marks real small.
It will work if hooked backwards. But it's performance may be suboptimal.
Don - Specializing in layout DC->DCC conversions
Modeling C&O transition era and steel industries There's Nothing Like Big Steam!
Does not matter with one speaker.
Rich
If you ever fall over in public, pick yourself up and say “sorry it’s been a while since I inhabited a body.” And just walk away.
richg1998 Does not matter with one speaker. Rich
I respectfully disagree. One way pushes a speaker outward. One pulls it inward. Depending where the linear magnetic gap is designed, and the bottoming out distance, you could end up bottoming out the driver on the magnet or pushing it outside the linear gap which would lead to a non 1:1 response to input.
Not for model railroading. This is not a home sound system. One speaker, not an issue.
This is not a cancelation issue due to 180 out of phase! This is a driving the cone in the wrong direction issue! It may not be designed to be driven in that direction Rich.Larger cones/drivers have the option of making it work equally well by creating a linear maget that surrounds the voice coil. But our tiny speakers have the magnet at the rear of the driver. Magnets are non linear in their distance response except in a tight area. Drive the cone outside that area and you are in non linear response (distortion/dynamic range compression)
Not an issue according to a DCC man. One speaker, not a problem.
http://mrdccu.com/curriculum/speakers.html
If there is an issue, just follow the manufacturers instruction.
Well Mr DCC is wrong. Feel free to consult any person here who's designed a speaker or amplifier from the ground up. :-)
Well, there is the added complication of what class amplifier is driving your speaker. Some classes go both positive and negative around the reference point which means the speaker cone needs to move in both directions anyway - so for a single spekaer it won't matter which way the wires are hooked up. Other amp classes only move in one direction above the reference point, so the speaker cone will only move one way away from the resting point, in which case it will matter which way you connect the wires so that the speaker moves in the direction it is most efficient in.
Odds are though, to keep the size and heat down and still provide a 1 watt or better amplifier in a tiny decoder, they are Class D or variations thereof, using PWM output, and thus the polarity will matter.
---Randy
Modeling the Reading Railroad in the 1950's
Visit my web site at www.readingeastpenn.com for construction updates, DCC Info, and more.
My shrink wrapped Tsunami decoders mentioned nothing about speaker polarity. Two purple wires.
Ricbh
I hope we can clarify this (here or another thread?). For instance, I've been using 1.1" HiBass speakers with enclosures, where I'd guess the speaker cone had lots of room to move in either direction. But the Knowles sugarcubes I've been using in tighter spots are different in looks. One has a rectangular foil looking front "cone" while the other has a rectangular front that looks like a piece of ceramic. And the LokSound Selects I'm typically using just have two identically colored speaker wires. So I'm not clear on whether to worry about it, or how to address it if I should.
UPDATE: From the LokSound Select manual..."The polarity of the speaker is not important for single speaker installation." This suggests to me that the decoders with twin, same color speaker wires are ok with any speaker, or the decoder vendors don't understand the issue for the type amps they have. Just a suspicion, can't prove it. Might be worth asking on the Loksound yahoo group page. And i'll attach a datasheet link for one of the sugarcubes, though I don't know how to interpret whether the speaker cares about which way the "cone" (membrane) is driven: flat to out, flat to in, out to in and back, etc, The devil's in the details in this hobby!
http://media.digikey.com/pdf/Data%20Sheets/Knowles%20Acoustics%20PDFs/2403-260-00071.pdf
Paul
Modeling HO with a transition era UP bent
If you check the tsunami installation manual you will clearly see they do indeed indicate which of the two purple wires is +. (The outside on the TSU-1000)
I could understand the confusion sets in. But as a rule of thumb the distance a cone moves is a function of field strength - surround resistence. (The surround is that rolled part that connects the cone to the spider frame basket.)
Field strength is a function of 1/d^3 (One over the power of the distance cubed)
You always want a linear response between the signal and the voice coil. (So many 1 microvolt = .xxx mm movement.) But since field strength is cubic and not linear, this becomes problematic design issue and a secret sauce to magnet design on a lot of speakers.
If you graph out 1/d^3 you see the graph quickly decays as you increase distance from your coil. Now remember if you reverse polarity on the speaker you are going for the sharp left side with step slopes. If you have correct polarity you are moving to the right where the slopes are more gentle.
So what we want is a slope that is consistent on the response with as little change to the slope as possible.
So if you take the first derivative of the field strength, you get the slope. -3/d^4
If you take the second derivative of the field strength you get the concavity change of 4/d^5. For unit distances > 1 the slope change is approaches 0 (ideal) But this is true if we go right on the graph. (Cone driven outward) Now if you go left, well then you are heading to a less ideal curve where slope changes quickly per unit distance. This is non linear in nature. And unless we have a surround that also responds in a similar non linear nature it's less then ideal.I will write Loksound and inform them polarity does make a difference.
Back to the original post. The speaker wires broke off the board, not the speaker and I cant tell which way to solder them back on to the sound decoder board. I know which one is positive and negative on the speaker. I did not see a dot or a plus sign on the board, but I will recheck.
So after calling bachmann, I got the same answer, it will cause the speaker to work in the opposite direction if the wires are swapped and you will get a bit of distortion. I took a guess and soldered the wires to the board. It seems to work okay. The power pick up issue also seems to be somewhat corrected, though the quality control on the replacement part I recieved was lacking (the wipers were not correctly bent/shaped/sized), but I put it in anyway.
I see another urban myth developing you might say.
I posted the comment in a SoundTraxx forum and got an unprintable comment from a former DCC company owner who works closely with SoundTraxx.
He is currently writing an article about DCC myths and will include it.
Bruce is a pretty smart guy, but there's no denying the response curve of a magnet - it's pure physics and like Scotty was fond of saying, "You canna' change the laws o' physics"
--Randy
Today a new old problem cropped up with this locomotive. I was looking at the valve gear and drivers and noticed a problem. The gear connecting rod was flipped up instead of down. I loosened the eccentric crank screw (its a screw instead of pin) and was able to move it. Now Ive noticed that the gear reach rod pin is missing from the reverse shaft arm. Does anyone know what to replace it with?