BTW all the turnouts are Insulfrog

Dear ???

Sounds like some of your locos may have wheelsets which are out of gauge. Suggest getting hands on an NMRA standards gauge, and making sure all wheels are the correct distance apart.

Such a problem can create both electrical (shorting on turnouts) and mechanical (derailing on turnouts) symtoms, and would explain why some locos appear ok while others do not...

Happy modelling,
Aim to Improve,
Prof Klyzlr

Peco insulfrog points are, counter intuitively, more likely to produce shorts than electrofrog points. The usual reason is that wider wheel treads touch the tip of the other frog rail as they reach the point of the frog. I think the only remedy is to dab some clear nail varnish on the rails at the Vee where they nearly touch. Worked well for me but will have to be renewed periodically. Also see Allan Gartner's http://www.wiringfordcc.com/switches_peco.htm for tips on improving electrical continuity.

Cheers

Tony in Gisborne, Victoria, Australia

My solution to the same issue was first to paint the frogs - which worked fine until the paint wore off. Then I cleaned off the paint and applied very thin super sticky .03mm sign vinyl to the frogs and they have been no issue ever since.

from: http://model-railroad-hobbyist.com/node/17708

Painted:

Vinyl applied:

from http://model-railroad-hobbyist.com/node/17708?page=2#comment-145020

Fixing shorts on Peco Insulfrog turnouts permanently

Thx for all your input guys - I'll trouble shoot the issue the first chance I get and keep you posted !!

Doesn't increasing the length of that 'insulated' section of track (the frog )increase the stalling potential of locos?

You're keeping the outer part of the wheel tread from touching the opposite rail, not the flange and fillet.

I've done what Joe described except I just used a little cheap CA instead of a bit of plastic. Pour a bead, hit it with accelerator, and file it down. Problem fixed.

Scott Chatfield

How about Epoxy.. no plastic filler needed, just let it harden and file it down.

Also, possibly, Bondo... though it can get runny before it starts to set.

Or Urethane cement (Hate using it... hard to handle, but would be Permanent)

Even tapering the railhead  inside the red zone downward would help...  lowering the railhead this way would also do the trick,  as contact is near-impossible - but filling it as Joe describes is a bit reassuring!

G'day Richie,

I replied elsewhere on here to some one else having the same problem - Here is a "bashed version" of that reply to save re-typing everything. I have added some extra stuff at the back end of the post that should help you get started on the derailment problems as well. Please have a long slow read. If anything doesn't make sense to you, please ask another question. There is no such thing as a "dumb question".   

I note that you are reasonably new to the hobby so you may have have missed the relevant bit in the Peco instructions. (Yes it is buried a fair way down on the instruction sheet.) 

the Peco instructions for Insulfrog instructions are availalbe online at https://www.peco-uk.com/imageselector/Files/Instruction%20sheets/HO-OO%20Insulfrog%20Turnouts.pdf 

The bold emphasis is mine, but the key point from the Peco instruction sheet is 

Turnouts and Crossings since they are electrically self-isolating and ready for use. However, some incorrectly shaped metal wheels can cause a short-circuit when crossing an Insulfrog due to the wide tread of the wheel touching both frog rails at the same time. This occurs when a back feed takes place in a continuous loop. To overcome it insert a Peco Insulating Rail Joiner next to the frog rail of the track forming the loop (3). The golden rule of two-rail electrification is to ensure that current is fed to the track from the toe end of any turnout(s).

That golden rule applies whether you are using DC or DCC. It doesn't matter. For DCC and Elecrofrog tunouts all four rails heading away from the turnout need to be isolated, not just the two frog rails. 

You have the track down already, so the option of using the insulating rail joiners is probably off the table. But you can replicate this by gapping the track on the two frog rails away from the frog area. This is a far easier fix than "frog surgery" suggested in previous posts above. It also has the advantage that it is done on straight rail and can be done far enough away from the frog that you don't  have to worry about your thin Dremel cut-off wheel snagging an adjacent rail by accident while cutting. 

It doesn't matter whether you are running DCC or DC. What gapping these 2 tracks does does is ensure that you don't have a short-circuit path in the frog area if the wheels span the insulating block.

How this works:  

Peco Insuffrog turnouts ONLY power the route the turnout is set for. The other track is electrically dead if the turnout is fed from the toe of the turnout where the routes come together. If something gets across the insulation block then frog rail on the other route feeds via the wheel and you get a short circuit. The other route frog rail is being powered by the feeders on that other route NOT by the turnout. This is where the short-circuit occurs. By gapping both frog rails you ensure that the other route cannot backfeed the frog area from the route that is not set.. It doesn't matter if both frog rails are the same "DCC polarity" while wheels bridge the insulating pieces, because there is no current path across the gap to short circuit on the 'wrong rail" on the other route. On the route that is not set there is also nothing to bridge across that gap.If both frog rails are gapped and you have correctly wired feeders. Just don't park an electrically conductive vehicle (like a lit passenger car) across the gap. This will defeat the gap.  

For the route that is set:

• power through the turnout up to the frog gap cut is being provided by feeders at the toe of the turnout
• power beyond the frog gap cut is being provided by track feeders
• power to the frog area to the other route is isolated at the turnout by the Insulfrog design.
• power to the frog area from the track feeders on the route that is not set is isolated by the gap in the frog rail for the route that is not set. 

With properly coned wheels it should be impossible to bridge over the insulating blocks. BUT some larger steam locomotives (probably your Broadway Limited locos) and some early smaller fixed-wheelbase diesels had "blind drivers" that had a flat profile and were fitted with un-flanged wheels.  This allowed these intermediate wheels to handle tight curves by sliding sideways on the railhead without having to provide sideways slop in drive-rods and axlebox bearings in the prototype design. Blind drivers are also a trick that some manufacturers have used that allows us to bend our model trains around scale-dimensioned curves many times tighter than than the prototype was designed to handle. Please check that your most persistent "culprits" are not fitted fitted with blind drivers. Even if they are so fitted, gapping the frog rails at each turnout should fix the shorting problem, but you might have to move the gaps to a distance from the frog that is longer that the length of your longest locomotive or locomotive combination. By moving the gap out by  loco length or more any individual loco can't provide a short circuit path if it bridges the insulating blocks at the frog.   

As for the derailing locomotives, let's look at these separately - It sounds like they are steam locomotives rather than diesels?? But if we are to help sort out the derailment problems then we'll we need some more details to zero in on the problem area(s). Can you please confirm the following: 

1. Which locos are the culprits? (Steam or diesel, model and maker)
2. What radius Peco turnouts are you using and which ones are you have trouble with??
3. You said "when going in reverse". Does this occur on both routes or the diverging route only?
4. Do you have the same derailment problem when travelling from the toe end of the turnout versus from the diverging end of the turnout? (What happens if you physically turn the loco 180 degrees and back it up from the toe end of the turnout? Does the same part of the loco derail at the same place?)
5. Have you any obvious kinks in the track where you go from the Peco turnout to the flex-track? (This can create a localised tight radius that will kick your loco sideways, causing it to bind and derail when it hits the turnout) In this case, the problem might not actually be the turnout itself.

The suggestion has already been made to get an NMRA wheel gauge and check the wheels for gauge. Assuming the wheels are in gauge, have a go at answering the above questions. 

To help you sort out the second question, here's some data on Peco Turnouts that correlates Peco part number and turnout radius that I've posted elsewhere on here earlier.

First some info on Peco collated from the Peco catalogue webpages with effective radius and divergence (in degrees out of 360 for a circle) The E prefix in front of the number is the Electrofrog catalog number. No E = Insulfrog catalog number. 

For Code 100 rail 

Set track radius #2 (ST240-ST242) 438mm (17 1/4")  Insulfrog only. 22 1/2 degrees

Short radius (SL91/SLE91-SL92/SLE92) 610mm (24") 12 degrees 

Medium radius SL95/SLE95-SL96/SLE96 914mm (36") 12 degrees

Long radius (SL88/SLE88-SL89/SLE89 1584mm (60") 12 degrees

Peco Set Track Curve Radius

R1 - 371mm (14 5/8")

R2 - 438mm (17 1/4")

R3 - 505mm (19 7/8")

R4 - 571mm (22 1/2")

For Code 83 (US profile)

#5 - (SL8351/SLE8351/SL8352/SLE8352) 660mm (26")  11.4 degrees 

#6 - (SL8361/SLE8361-SL8362/SLE8362) 1092 (43") 9.5 degrees

#8 -  (SL8381/SLE8381-SL8382/SLE8382) 1702mm (67") 7.15 degrees 

If you know what the Peco catalog number was for the Insulfrog turnouts, then we can work backwards to get the radius for question 2 above 

That's probably way too much to digest in one bite for a relative newcomer to the hobby. As I said upfront, please have a long slow read and if in doubt, please ask.