Fun with Proto:87, plus some video and sound

[ednadolski]

I've always been intrigued by Proto:87, and I finally got around to installing some P:87 wheels in a loco that I had sitting in the box for a long while.   The installation of the wheels was surprisingly easy -- pretty much a drop-in, and they run smoothly enough even on some Micro Engineering flextrack that I simply laid down along a couple of shelves.

The other fun part of this was the sound.   Instead of relying on the tiny (and tinny) 1" speaker in the loco, I took a Tsunami sound decoder and hooked it up to a set of high-quality bookshelf speakers.   The only additional parts I needed were an audio output transformer and a 1/8" audio plug from Radio Shack (as described on Lance Mindheim's April 20, 2012 blog.   Then I consisted the two decoders together and muted the loco's onboard decoder.  I also spent a little time tweaking the speed and momentum CVs in the loco's decoder to match the sound with the motion of the loco.

So here is a video of the loco in action: Not a valid vimeo URL.    The main thing to keep in mind is that this sound is not dubbed in, it is the actual sound in the room as it was was recorded by the camera.   What is cool about this technique is that it can be used in any scale, and is certainly an interesting approach for N scale since onboard sound in an N scale loco is a huge compromise at best.   I'll probably see if I can adapt this to my Tehachapi layout, perhaps with some additional decoders, or maybe a SurroundTraxx.


Here too are some pics of the P:87 conversion.   As mentioned, it was as easy as simply swapping the half-axle code 110 NMRA wheelsets with the NWSL code 64 P:87 ones.   In this case, the Kato loco takes the wheels with the 1.5mm axles.










Ed

Edit: Fixed Lance Mindheim link 01-05-2018, image links fixed 01-06-2018

[gregamer]

Ed,

Looks great. I did a Kato SD40-2 conversion also. I noticed a frog at about 28 seconds into the video. Are you building Proto:87 turnouts?

Greg Amer
The Industrial Lead
http://www.gregamer.com

[ednadolski]

Hi Greg, I saw your web page with your P:87 work, that's what inspired me to try this out!   Nice work! 

I do have some frogs for converting some ME #6 turnouts.  This is a WIP for me, as I have to also make some guard rails (easy enough) and then figure out the best way to install the frogs and guard rails (less easy.... maybe epoxy?  spikes?  solder to PCB ties?).  I did another video of a "flextrack test" to see how well the P:87 wheelsets track through a #6 turnout geometry:

Not a valid vimeo URL
(note, the centerbeam also has P:87 wheels)

I am also converting the turnouts to Andy's throwbars, which seem to work OK (as long as you don't try to pick up the turnout!  ).   I did have to drill an extra set of holes in the throwbars, in order to reduce the oversized gap between open point and its adjacent stock rail.  The other thing I want to do is solder a jumper wire on each point rail, for better electrical contact.

Ed

PS:  I'm really liking that Tsunami 645 sound!   

 

[bnsf971]

Hi Greg, I saw your web page with your P:87 work, that's what inspired me to try this out!   Nice work! 

I do have some frogs for converting some ME #6 turnouts.  This is a WIP for me, as I have to also make some guard rails (easy enough) and then figure out the best way to install the frogs and guard rails (less easy.... maybe epoxy?  spikes?  solder to PCB ties?).  I did another video of a "flextrack test" to see how well the P:87 wheelsets track through a #6 turnout geometry:

Not a valid vimeo URL
(note, the centerbeam also has P:87 wheels)

I am also converting the turnouts to Andy's throwbars, which seem to work OK (as long as you don't try to pick up the turnout!  ).   I did have to drill an extra set of holes in the throwbars, in order to reduce the oversized gap between open point and its adjacent stock rail.  The other thing I want to do is solder a jumper wire on each point rail, for better electrical contact.

Ed

PS:  I'm really liking that Tsunami 645 sound!   

You can change the setting on the equalizer in the Tsunamis, and turn down the higher frequency sounds on the shelf-mounted one. You can also turn off the lower pitched ones in the engine-mounted one.

[ednadolski]

(Updating this after a long hiatus.)

I've been trying to convert a code 70 Micro Engineering #6 turnout to P:87, but my attempt has met with failure.

The main part of the conversion actually went pretty well.  This involved replacing the factory frog with the P:87 frog from the P:87Stores, and guard rails made from code 70 rail.  I epoxied the new parts into place, carefully aligning them with the factory rails.  A P:87 truck rolled very smoothly thru both routes, aside from a very slight bump at either end of the frog because I did not get the rail/frog height perfectly matched.   (That could be resolved with a bit of shimming  when the turnout is permanently installed.)

The real problem with the ME turnouts is that the track gauge is overly-wide in the diverging route at the point/switch area (the gauge in the frog area is actually fine thru both routes). I measured the gauge as ~0.010" to 0.012"  wider than the maximum gauge allowed for P:87.  Presumably this is an attempt to accommodate long/rigid wheelbase locomotives made to traditional NMRA specs. But the P:87 truck with its code 64 wheels gets about 3/4 thru the point/switch area when the leading wheel drops into the gap between the open point rail and diverging stock rail.  This forces the opposite wheel to ride up and over the other point rail, putting the car onto the ground.

Interestingly, I observe this with 33" wheels but 36" wheels make it thru OK. Evidently the difference in diameter is just barely enough to avoid the issue. But this hardly seems like a recipe for reliable operation in general.

There is no good way to fix the gauge issue that I can see. The diverging stock rail is securely embedded in the molded plastic tie base and cannot be moved without significant damage.  One would have to cut away many plastic ties and replace them with PCB ties, but doing so would effectively defeat the purpose of using a converted turnout in the first place.

Thus I reluctantly conclude that Micro Engineering turnouts are inherently incompatible with P:87 use.  Hand-built turnouts look like the only workable option at this point.   Live an learn I guess.

Ed

[GaryHinshaw]

I totally forgot about this thread - it wasn't until I got to the last post that I realized how old it was.  (Great sound BTW!)

Sorry to hear about the failed turnout experiment, but I recommend that you check out this thread:

https://www.therailwire.net/forum/index.php?topic=34762.0

That guy came up with a really cool method for hand-laying N scale turnouts.     Maybe it would transfer well to p:87?   (How is that project coming?)

[ednadolski]

It would probably work out ok, but (a) the drawing would have to be done basically from scratch; and (b) it would take up a lot of sheet area (even a #6) so the cost would be non trivial.  The reasons for trying the ME was that I just happened to have a few hand, plus it has nice details without having to hand-spike everything.

I'm hoping to get back to the N project soon.  I put it aside because I needed some materials, then I got caught up with a few other things.  One of those was (speaking of sound) upgrading the GP9 to a Loksound, which I like a lot better than the Tsunami.

Ed

[ednadolski]

I was able to get this to work, with thanks to some suggestions from folks on the P:87 group.

The fix took two steps:  first, I straightened the point rail for the diverging route, since that had some excess curvature.  Second, I corrected the overly-wide gauge by moving the closure rail.  I shaved off the molded spikes & tieplates from about 10 ties from the hinge, then bent the rail toward the gauge side until the P:87 track gauge showed a better fit.   I ran a test train thru the turnout about 20 passes back-and-forth, and it went smoothly without derailments.

The rail seems to be holding its place, but I will drill a few small holes in the ties and install some P:87 spikes to help keep it there.

BTW the test train included a 73' centerbeam and a Kato SD40-2 6-axle loco.  Nice to know the longer rolling stock can navigate the #6 geometry, so this need not be limited to 40' boxcars and tiny diesels 

Ed

[Mark5]

The other fun part of this was the sound.   Instead of relying on the tiny (and tinny) 1" speaker in the loco, I took a Tsunami sound decoder and hooked it up to a set of high-quality bookshelf speakers.   The only additional parts I needed were an audio output transformer and a 1/8" audio plug from Radio Shack (as described on Lance Mindheim's April 20, 2012 blog.   Then I consisted the two decoders together and muted the loco's onboard decoder.  I also spent a little time tweaking the speed and momentum CVs in the loco's decoder to match the sound with the motion of the loco.

I always thought that when I went DCC that this would be a better solution (for me).

How exactly does the sound "get piped" into the speakers from the decoder? (the lancelot link above is dead)

Mark

[tom mann]

Ed, do the Kadee code 88s have the correct wheel back contour now, or are they slab backs?  Or, are you using another wheel?

[ednadolski]

How exactly does the sound "get piped" into the speakers from the decoder? (the lancelot link above is dead)

The speaker output of the decoder goes to one side of the audio output transformer.  The other side goes to the audio line input of the computer speakers.

Remember that the decoder has its own (small) power amplifier designed to drive the 8-ohm speaker.  This can easily generate a signal that it too strong for the audio input line of the computer speakers, resulting in clipping/distortion.  I had to turn the master volume output of the decoder way down to avoid that.

Ed, do the Kadee code 88s have the correct wheel back contour now, or are they slab backs?  Or, are you using another wheel?

I'm using the NWSL wheels, but they have flat backs.  I'm not sure if there is any P:87 wheel that has the back side contours, but I would love to see that.

BTW, P:87 wheels are actually code 64.  Code 88 wheels still have RP25 flanges, which are over-scale just like regular HO code 110.


Ed

[Mark5]

Thanks Ed,

I get the part about the audio transformer and the speakers.

Trying to visual the wires that go to the audio transformer ... they are certainly not coming from the decoder! Are they connected to the rails?

[ednadolski]

Trying to visual the wires that go to the audio transformer ... they are certainly not coming from the decoder!

Yes they do.  It is a separate, stationary decoder.  You do have to power it by connecting it to the track power bus (i.e. the rails).

It's unfortunate that the LM link went dead, as that had a good picture of it.

Ed

[Mark5]

Hmmm, stationary decoder .... thanks, gotta research this more.

[ednadolski]

Hmmm, stationary decoder .... thanks, gotta research this more.

Not much more to research. Track leads go to track, speaker leads go to the speakers (thru the audio output transformer), motor & other leads left open.  Mount neatly (I was sloppy and used alligator clips), program to desired address (I used with advanced consisting), and reduce the master volume.  The decoder doesn't really care if it is in a mobile loco or not. 

Ed