Plastic vs Metal vs Wide vs Narrow N scale wheels

[peteski]

A quick update.  I am finding the shimming the bottom of the frog with Evergreen styrene is working very well on Peco Electrofrog turnouts.  Using 0.010" thick material.   Testing with a train of BLMA gons and I shove back and forth all day without a derailment (well maybe 8-10 times, till I get tired).

Also doing the guard rails with the same material.  Maybe I won't sell the "narrow tread" wheels I had set aside after all.

Bill Kepner

Good to know that the theory works in practice.

[jdcolombo]

Hi Bill.

When I was using Peco Electrofrog turnouts (for over 15 years, before starting my current layout in 2010), I routinely shimmed the guardrails with .010 x .060 styrene strip.  the .060 strip width is almost a perfect match for the plastic guardrails on Peco turnouts, so all I had to do was cut the strip to the appropriate length, and then glue it on with a couple of drops of liquid styrene cement.  I used wooden toothpicks to hold the ends and center of the shims against the guardrails while the cement was drying.  A little brown paint hides the white styrene shim (although you'll need to be careful when cleaning the rails next to the guardrails, or you'll scrape off the paint and need to repaint). 

I didn't have frog issues, but this was before the day of narrow-tread FVM wheels.  If the wheels are properly gauged, however, the guardrail shims are absolutely necessary - otherwise, wheels will routinely pick the frog point and derail.  This was on both their Code 80 and Code 55 turnouts.

John C.

[leedabsme74]

Excellent description, John.  I have done the same thing with success.

[drgw0579]

It seems there is some variation of how much gap there is at the guard rail.  In one case, an 0.010" shim was too thick.  I think the NMRA standards call for a 0.028" gap, in some cases it looks like out of the box, the distance could be 0.034". 

I went looking for something I could open up the gap some.  "Needle files" appear to be too thick (at least for those I found that actually specified how thick they are; those in the MicroMark catalog do not ).

I found a hacksaw blade, specifically a worn hacksaw blade is about right.  I also need to hunt down the set of clearance gauges I once had for setting the valves on my BSA, to be able to check width of the gaps.

Its funny how when you start looking at a problem like this, some suggestions from the community and a bit of experimentation turns up a good process.  Now it doesn't seem like rocket science like it did when I started this thread.

Thanks all.

Bill Kepner

[garethashenden]

It seems there is some variation of how much gap there is at the guard rail.  In one case, an 0.010" shim was too thick.  I think the NMRA standards call for a 0.028" gap, in some cases it looks like out of the box, the distance could be 0.034". 

I went looking for something I could open up the gap some.  "Needle files" appear to be too thick (at least for those I found that actually specified how thick they are; those in the MicroMark catalog do not ).

I found a hacksaw blade, specifically a worn hacksaw blade is about right.  I also need to hunt down the set of clearance gauges I once had for setting the valves on my BSA, to be able to check width of the gaps.

Its funny how when you start looking at a problem like this, some suggestions from the community and a bit of experimentation turns up a good process.  Now it doesn't seem like rocket science like it did when I started this thread.

Thanks all.

Bill Kepner

If you need to regularly file straight gaps of 0.028" you can get a nut file. It looks like a feeler gauge, but has teeth on the edges. They're normally used for making/repairing straight cut screw heads. https://www.amazon.com/Grizzly-T25464-Nut-File-0-028-Inch/dp/B00DQJOX2S/ref=sr_1_fkmr0_1?ie=UTF8&qid=1489506338&sr=8-1-fkmr0&keywords=slot+file+0.028

[mmagliaro]

Holy cow!  NUT FILES!   I never even thought about them for model work.  And they sell them all the way down
to .010" thick for about $14 each (check out the Stewart MacDonald (http://www.stewmac.com/Luthier_Tools/Tools_by_Job/Tools_for_Nuts_and_Saddles/Gauged_Nut_Slotting_Files.html )

I have had many occasions where I wished I could file an ultra-thin slot.  I just bought a Grobet "special thickness" file for this,
at .016", but it cost me 30 bucks.  Nut files are much cheaper! 

[John]

I somehow had a different image in my mind 

[Kisatchie]

I somehow had a different image in my mind 

Hmm... me too. I have
some cashews that need
a little filing...

[atsf3751]

It's always fun seeing topics like this because many have strong opinions. It's just like the last time I was in my LHS picking up track and one of the older employees was explaining to me that no one uses code 55 because it's so unreliable and how code 80 is the only way to go, lol.

Just imagine what he would say if he saw our code 40 on the club layout.

[AKNscale]

He would probably be flabbergasted, lol

[peteski]

I don't understand how the height of the rail somehow relates to the track reliability.  The wheels only contact the rail-head, regardless whether the rail is 40 thousandths or 80 thousandths high.

[robert3985]

I don't understand how the height of the rail somehow relates to the track reliability.  The wheels only contact the rail-head, regardless whether the rail is 40 thousandths or 80 thousandths high.

Code 80 rails are MUCH more reliable in N-scale than mere code 40 especially when you start dropping anvils and ball-peen hammers on it....

"Understanding" or "thinking" doesn't have anything to do with why some people think thicker rails are more reliable than thinner rails.

Cheerio!
Bob Gilmore

[mmagliaro]

I think the "reliability" thing is not so much related to the height of the rail as it is to the manufacturing process of the track.

I may be treading into murky political waters here, so I will try to be as factual and honest as I can.

Atlas code 55 is more fragile.  For example, I have had the rail pop out of the ties on two very generous curves (i.e 24" radius) and this is not on some NTrak module being tossed in the back of a van every week, or on a layout that has been torn down and moved, or sat on by little kids or a cat.   It is in a clean, temperature and humidity controlled spare bedroom.   I have repaired the failures by setting PC board ties in place and soldering them, and then gluing and spiking down those PC board ties to keep the track fixed in place.  This was in a section of hidden trackage, so the repair doesn't show.

Second:  We all know about the breaking Atlas code turnouts ( throwbars, point rails breaking or being out of gauge, throwbars that let the point rail tip creep up over the stock rail, contact bars under the turnout failing so part of the turnout goes "dead", etc)

So while I agree that the bias toward code 80 has nothing to do with the rail size per se, I can sympathize with some people getting that impression when they hear about Atlas code 55 "failing".  The message they hear, albiet incorrectly, is "code 55 is unreliable", when in fact, the correct message is that Atlas code 55 is fragile      It looks great and it performs well once it is tweaked, but it does have a higher failure rate than the code 80 stuff.

And to the Atlas company reps who frequent these boards, I apologize in advance if this sounds like I'm bashing.  But I have seen enough issues with the code 55 track that the only honest conclusion I can come to is that it fails more and has more functional problems than the code 80, and it has nothing to do with the size of the rail.

[atsf3751]

I don't understand how the height of the rail somehow relates to the track reliability.  The wheels only contact the rail-head, regardless whether the rail is 40 thousandths or 80 thousandths high.

Rail height is relevant on one point and that is the depth of the wheel flange. When we started building the layout (1981) MT pizza cutters were the norm and many engines still had deep flanges (Rapido engines were more than 40 thousands). When we tried flex track there were many cars and engines that bounced on the simulated track nails. This was not a pretty sight.

Soldering to PC ties was the solution, only Rapido equipment still had problems. Of course, we still fail the ball peen hammer and anvil test, but we are looking into it.

Marty Young
San Diego, CA

[peteski]

Rail height is relevant on one point and that is the depth of the wheel flange. When we started building the layout (1981) MT pizza cutters were the norm and many engines still had deep flanges (Rapido engines were more than 40 thousands). When we tried flex track there were many cars and engines that bounced on the simulated track nails. This was not a pretty sight.

Soldering to PC ties was the solution, only Rapido equipment still had problems. Of course, we still fail the ball peen hammer and anvil test, but we are looking into it.

Marty Young
San Diego, CA

That is a mechanical incompatibility issue. It has nothing to do with reliability - at least from my perspective. N scale wheel and track standards have been redefined over the years and the incompatibility of the old and new seems to have caused the mechanical interference.