Wheel Change

[wpsnts]

Have you looked into NWSL's products?

[peteski]

Most likely somebody doesn't like our "tone"...whatever that is...

I didn't see any problem with the "tone" of our posts. It was quite techincal, with some speculation sprinkled in.   

I upvoted your comments as I enjoy intelligent, well-informed comments as well as spirited and logical defense of opinions which I may or may not agree with...one of the things I really appreciate about TRW.

Ah, so that is what happened - now it made sense to do the same for you. 

As to the well-distributed opinion that Peco, Atlas, Shinohara, and other manufacturers of N-gauge track took British Railway's track standards as the example to model their track after...I did a bit of research a few years ago, and determined they don't comply with any prototype track standard.  I've lost the information in the years since then, but it wasn't difficult to find at the time.  Nope, Peco C55 track doesn't look much like prototype British Railway trackage...of any era.

My conclusion was (and is) that whomever designed the first N-gauge track designed it to look generally like railroad track, but to be sturdy enough to allow trains to run while the track was set up on green or orange shag carpet.     It's toy track...while N-scale Railcraft/Micro Engineering track is model track...and Atlas C55 is almost model track.

Not that there's anything intrinsically "wrong" with toy track, since a pretty strong argument can be made that ALL model railroading is just people playing with toy trains...in one way or another...

Interesting, and good to know.  Maybe there was also another reason:  the early N and Z models were sometimes oversize (to accommodate the smallest available contemporary motors and gearing) ,and rather crude and lacking details.  The track designers might have thought that an accurate depiction of the track would look out of place with those models over it (amplify the inaccuracies and simplifications of those early  models).  Just another speculation on my part.

[sd75i]

  They run fine on the Peco code 55 concrete ties!  This layout Im putting back together has Atlas code 55 all over it, so I wont be changing the track!  I was thinking the same thing NWSL!  I also like the way the Atlas code 55 track looks!  We conquered the powerpack here, so I know this will be conquered as well!

[nkalanaga]

Bob:  I agree with the "toy track".  I've long thought that N scale should be considered two "scales", like O - "tinplate" and "scale". 

They have Lionel and similar models, that don't look particularly accurate, but run well, are quite durable, and can easily be used for operations if one doesn't mind the appearance.  In other words, they're for people who want to run trains.

Then they have the true "scale O", intended for those who expect trains to look real.  Harder to build, and to make work right, because the tolerances are much tighter, and not as durable, because of the finer details, but much better looking, for those who expect their trains to look as good as they run.

And, just like O, N started with the "tinplate" trains, and the "scale" models evolved from them.  The main difference, besides the size, is that N started with the right gauge, thinks largely to Arnold.

[peteski]

  They run fine on the Peco code 55 concrete ties!  This layout Im putting back together has Atlas code 55 all over it, so I wont be changing the track!  I was thinking the same thing NWSL!  I also like the way the Atlas code 55 track looks!  We conquered the powerpack here, so I know this will be conquered as well!

As mentioned earlier, there was no doubt that those deep flanges would run on Peco C55 track (wood or concrete ties).  I have my doubts that NWSL will have direct replacement for those geared Fleischmann wheels.  You could probably have NWSL custom-make them for you, but that will cost a lot of money. The cheapest way to do this would be to file down the flanges yourself. Several people on this forum have done this - maybe they'll chime in.

I also still cringe at having a miniature disk-sanders grinding down the railheads. 

[nscaler711]

I've long thought that N scale should be considered two "scales", like O - "tinplate" and "scale". 

And, just like O, N started with the "tinplate" trains, and the "scale" models evolved from them.  The main difference, besides the size, is that N started with the right gauge, thinks largely to Arnold.

That is why I make jokes about some things being like N-9 kinda like O-27...
Like some MT products, (Sorry Joe) most look great, operate reliably, but something is always off... Like ride height of the cars... I suppose at least the cars aren't "semi scale..." Whatever that means... It's either 1:48 or its not lol...



Disclaimer: I actually like MT products, and a new set of trucks usually go a long way in lowering cars... Like BLMA trucks.. .... and its not just MT.... Bachmann falls in the N9 category more often than not....

[robert3985]

Bob:  I agree with the "toy track".  I've long thought that N scale should be considered two "scales", like O - "tinplate" and "scale"...

@nkalanaga ...Haha...yeah, I've been dividing up 1:160th scale model trains into two groups...  "N-gauge" and "N-scale" for over a decade, but I've never 'splained it, except once over at another forum, and almost got banned for being "elitist" and "denigrating".

"N-gauge" and "N-scale" are close enough that ya gotta specifically look for 'em and how I use the terms, but especially on the subject of track, and wheel profiles...I'm always using both terms...probably to only my own satisfaction! 

Cheerio!
Bob Gilmore

[randgust]

Meanwhile, maybe back on topic....

I have turned down flanges with some rather rudimentary tools -  a hand needle/flat file, a dremel, and a micrometer.

It's not as difficult as you make it.   It's a lot easier on small wheels than big drive wheels, and easier on split-axle types than regular.    Even on 'old style', if you have a normal, metal axle on plastic centers you can pull off one wheel at a time, chuck the other wheel and axle in a dremel with a speed control, run it at medium speed, and start gently running it into a sharp, flat needle/flat file  (the one I have is a transition from a sharp point to a 1/8" wide over its length).    The quality sharp file is the most important tool.  Don't use a cheap knockoff on this one.

As far as for dimensions; figure wheel diameter width at flange (to account for any taper - add 2X on your target flange depth (.024?   pick what you want...) set the micrometer as a gauge and let the file run down until you clear it on exterior diameter over the flange.   Then round the flange edges over.

If you've got one oddball unit like this and Atlas C55, its probably no harder to turn the flanges than trying to find replacement wheels and get them in the drive.  That can be really tough.

If you have a very robust mechanism, lots of time, and a really sharp file you can turn flanges while the mechanism is running rather than taking the wheels out.   I sure don't recommend it, only did it once on a brass steam locomotive that had no way to get the wheels out.

As far as for track, I'm a Peco c55 guy on all rework and new construction on what was Atlas C80.   

1)  Darn tough material, you'll never break it loose from the ties by accident
2)  Painted and in ballast, it looks just as good from the side as Atlas in photos shot at track level.
3)  If you have ties weathered more or less the same color as the surrounding ballast (which I do) then the tie size and spacing becomes much less relevant to the decision than if you had high-contrast ties on placed on light ballast where the ties are more visible.
4)  The C55 Electrofrog switches are really well-made, tough, and have drop-in geometry almost identical to old Atlas C80 so substitution isn't hard.  No switch issues here.  That also solves most of the old Atlas stalling-on-the-frog problems at low speed.   
5)  I can still run my legacy N - and visiting equipment - without issues. 
6)  I'm amazed how many low-flange wheels work fine as long as tread width is adequate.
7)  Peco manages to keep stuff in stock in the distribution chain.
  The finger-flip of turnouts is nice right out of the box until you decide to upgrade to something more, if you ever do.

One of the chief disadvantages is that cutting and fitting the stuff - including redoing the ties around every rail joiner - is a tedious process.   It takes me 3-4 times longer to lay the stuff and get it right than C80.

Yeah, Atlas C55 looks better.   The question in my mind was whether or not the appearance offset all the other issues I'd seen firsthand on other layouts.   I was kinda traumatized by what happened to Jerry Britton's layout - up to that point I was probably undecided.   I do accept that a) Atlas tried to make good  b)  Atlas may have made some fixes in the switches.... but yikes... that left a mark.

And Bob, I admire the livin' "L" out of what you do with handlaid stuff and custom switches and acknowledge it's way better.   Back in the day I handmade some crossings and even a turnout or two as an experiment.    Nope, not for me, everything I made turned into a maintenance headache.  I may be able to do .010 brass handrails in my sleep but for track components I still want to be able to take it out of a package.

[sd75i]

  That's good info to take in.  Not sure I understand everything about turning wheels and flanges down but I'll research more.  I was gonna buy a fast track jig for switches!  I haven't really looked into the peco switches yet.  I know when he had layout running, he had lots of problems with atlas code 55 switches.   Derailments can really make things discouraging!   Any videos out there about turning wheels?

[randgust]

Here, this is close enough to what I do - other than the fact I would chuck-up the target wheel just as close as I possibly could to the dremel rather than let an axle take all the bending load from filing.   I'm also checking my work with a micrometer to hit a target flange depth.

 

/>
It looks like he's working on an old Atlas wheel, which is probably a great way to practice.    I worked on a couple old Roco GP40 wheels to practice.   There's really only two things you can actually do 'wrong', one is take too much off, that's hard to fix, the other is to leave a sharp edge on the flange that will pick switchpoints and rail joints.   Heck, I've seen factory wheels (particularly Japanese stuff) with that problem.

His file looks just like mine.   And that's one of those tools I jealously guard as it's turned out to be the best little file I've ever seen. (Xuron?)

[mmagliaro]

Some comments, having had to do this many times.

What that fellow is doing in the video will work, but be warned that wheelsets often have the axle pressed into a plastic or other
insulating bushing at one end so that you don't have a short across the wheels.  Because of that, the axle is never as tight inside
that end as it is in the end where it's pressed right into the metal wheel, and when you are running that file on the flange,
it can start slipping the axle inside the wheel bushing.  If that happens too much, you'll heat it, bend it, or otherwise egg-shape the hole
and ruin the wheelset.  This is just something to be wary of.  Go SLOW, when you do this,  and only apply as much pressure with the file as you absolutely have to in order to get metal to come off the flange.

It's much better if you can pull one wheel off the axle, and chuck the axle itself in the drill (or Dremel), as Randgust suggests, so that the wheel is very close to the nose of the chuck, instead of hanging way out in space on the end of an axle.  If it's a geared wheelset, you may not be able to do this.  But if it uses a hollow axle tube with a gear on it, with the two wheels pressed in on each end via half-axles, then you have it made.  You can pull out the two half-axle wheelsets, chuck their axles in the drill one a at a time, and file away.

Finally, after you get the flange turned down to the diameter you need (stop often and check with calipers),
spin the wheel some more and use some 400 grit, then some 800 grit, and then some 1000 or even 2000 grit sandpaper on that
flange so that it is smooth.  The file will leave it quite rough (even a fine file), and you will eventually notice this on curves or turnouts.  A rough scratched-up flange makes for noisier running, more drag, and more dirt accumulation.

On diesel wheelsets, rolling stock, and pilot or trailing truck wheels of steam locos, this is an easy thing to do.
The complications go way up when you have to do it on steam loco drivers, because now you have to worry about slipping the drivers on their axles, getting them back on the axles in proper quarter and without being loose or wobbly.

[randgust]

One thing about working with a Dremel is that at low-medium speed, you don't have a heck of a lot of torque there, unlike a normal lathe.  So you're somewhat reminded by a drop in RPM that you're working the tool too hard.   That's what I meant about the sharp file - the sharper the file the better it cuts without putting a distortion load on the axle or creating heat to damage any plastic component.

When I blinded the flanges on my 4-8-4 it literally took hours as the rods and crankpins were going to take a terminal beating if I didn't take it really, really, easy with the file.     If you're seeing metal come off and the motor RPMS are relatively constant, you're doing it right and the file is sharp enough to do the job.

Some of the coolest little mechanisms out there from the oddball Japanese manufacturers still have absurdly deep flanges.   I've had to turn Tomytec wheels (which are split-axle) for customers running on Atlas C55, and at this point I consider it pretty easy to do.   

[mmagliaro]

One thing about working with a Dremel is that at low-medium speed, you don't have a heck of a lot of torque there, unlike a normal lathe.  So you're somewhat reminded by a drop in RPM that you're working the tool too hard.   That's what I meant about the sharp file - the sharper the file the better it cuts without putting a distortion load on the axle or creating heat to damage any plastic component.

When I blinded the flanges on my 4-8-4 it literally took hours as the rods and crankpins were going to take a terminal beating if I didn't take it really, really, easy with the file.     If you're seeing metal come off and the motor RPMS are relatively constant, you're doing it right and the file is sharp enough to do the job.

Some of the coolest little mechanisms out there from the oddball Japanese manufacturers still have absurdly deep flanges.   I've had to turn Tomytec wheels (which are split-axle) for customers running on Atlas C55, and at this point I consider it pretty easy to do.

I totally agree with everything you say, Randy.
And that's a good point about "hours".  Don't be shocked if you really do have to take 15 or 20 minutes, or even more, to do one wheel.  Slow and gentle  - it's crucial.
I've even had a few where I braced the engine upside down in a foam cradle, running on its back, with the file carefully positioned in an Xacto "Xtra Hands" jig so the file would just rest on the flange, and then just run it like that and walk away for a while.  That way, it could gently grind ("rub" is more like it) for an hour if it had to without putting any stress on the mechanism.

I don't think the OP is going to be in that situation here.  Those wheels are probably sturdier.

[robert3985]

Excellent advice and cautions from both Max @mmagliaro and Randy @randgust.  I'd like to add one more, and that is to be careful with metal shavings getting into the mechanism, including the motor, U-joints and gear trains.  When I've used the rotation of the wheels still in the model to file down flanges, I ALWAYS make sure I get the openings with the tips of the gears poking out shielded with shims and tape so they're not open to the small metal chips I'm going to produce getting stuck on their oily/greasy surfaces and causing problems. It doesn't take a huge chunk to cause problems, and the particles are difficult to see if they get embedded in the grease.

In this thread, I haven't said much about hand-laying track...but since the subject has been broached by both Randy and @sd75i, I'll say a few words about "rolling yer own" turnouts...

Randy...ya made "two" turnouts????  Everybody knows it takes at least three to five hand-laid turnouts to learn the "tricks" to get properly functioning, reliable and good-looking turnouts.  With Fast Tracks jigs & fixtures, it takes two. So, truth is, you just didn't like making them...and a lot of people feel the same way.  Truth is, there have been times over the past 40 years or so when I bought and used Micro Engineering C55 #6's to save time when I was under the gun.  ME #6's are by far the best-looking and best operating commercially made N-scale turnouts on the market, and like Peco turnouts, they have an over-center spring mechanism which makes them immediately functional without the need for a switch motor.  I would, under the right circumstances, purchase other turnouts of Micro Engineering quality if they were available...but they're not...so I make my own, which consist of #6's, #8's, #10's and #12's...with  three-ways, wyes and curved turnouts wherever I want/need them.  One of the advantages to making turnouts using the solder/PCB tie technique, is that pizza cutters run unobstructed, even on code 40 trackage...which is the main reason I chose to make all of my Park City Branch trackage using handlaid code 40 soldered to PCB ties (every fifth tie), since at the time, guests invited to op session had not yet converted to lo-pro wheelsets.

Properly constructed and proportioned hand-laid turnouts are much more attractive, much more reliable, at least as robust, and much easier to repair than commercially available turnouts.  Since you are not limited to the type, or angle of turnout when you roll your own, another big advantage to making your own turnouts is that the design of your layout's track plan becomes much less restricted.

Lately, after reading what several modelers do to fix their commercially made turnouts' problems, all that wiring and shimming is automatically not necessary if you make your own, and may make building hand laid turnouts only take a bit longer to construct than the time it takes fixing commercial turnouts' problems.

'Nuff said about hand-laid turnouts for now.

However, you can list all day and all night how to minimize or hide Peco/Atlas 80/Kato tracks' cosmetic problems, but in the end, all of them do not come close to appearing like North American prototype trackage...or European prototype trackage either.  Paint, weathering and ballast only distract or hide the problems.

The main problem with Atlas 55 track is the oversized "nubs" which hold the rails to the ties...both from a functional and cosmetic standpoint.  The turnouts have a proportion problem in that the closure rails between the closure point heels and the frog guard rails are much too short...MUCH, making an Atlas 55 #7 turnout be proportioned more like a #6.3.

So...IMHO, the very best combination of functionality, appearance and availability would be to use Micro Engineering C55 flex, Micro Engineering C55 #6 turnouts, Atlas C55 #10's, curved turnouts and diamonds (stay away from their #5's), and learn to hand-lay turnouts and diamonds for spots on your layout that commercial products are not available for.  "Fixing" Atlas track isn't so difficult and involves sanding down the "nubs" a few thousandths...which I do on my Micro Engineering C40 flex.  Using this protocol, you don't need to file any flanges on your old, pizza-cuttered treasures...just put 'em on and run 'em!

Cheerio!
Bob Gilmore

[sd75i]

  I'm getting some great info here and much appreciated to all!