the problem is not the 18" curve per se, but the crossover that has two of them in an s curve without any straight track in between. my grandfather used to run a mallet 2-8-8-2 and a 2-10-2 through 18" curves. but there were no s curves without at least 9" of straight track between the curves.
they looked horrible but ran pretty well. i ran u30c's and sd45's around 18" curves successfully as long as the first car wasn't shorter than 50'.....
once again, no s curves without an intermediate straight....

n scale is great for scenery. you can build mountains that look like mountains that the railroad had to tunnel through. operations, though much improved, still leave something to be desired if you like switching. most of the cars have truck mounted couplers, most of the locomotives body mounts....

HO on the otherhand, runs like a fine swiss watch. because of its size and weight, electrical contact problems are minimized. most cars and locomotives now have body mounted couplers, and switching is a breeze. it is still small enough to build respectable scenery, yet large enough to detail....

sacles larger than HO, you get all the detail at the expense of the scenery....

HO is the best compromise, but a quick check of  any hobby shop will show slightly lower prices for n scale cars and locomotives....

wiring a layout is not hard. once you get the hang of it. it is basically the same 3 or 4 circuits repeated over and over.

the tyco and ahm stuff, like most train set equipment from the 1970s, were truly dismal performers. and yet.....i still have some around. with a little investment to replace the wheelsets and the truck mounted couplers you could make some nice running cars out of them. some of the stuff they had was truly unique, like the 60' hi cube boxcar, or its little brother, the standard height 60' boxcar. ahm had a model of a tie car, sort of a bulkhead flat with a bar cage along the sides. or the 6 dome tank car. all of these i would consider upgrading simply because of their uniqueness.
and back in the days before the bowser cabooses, the old tyco streamlined caboose could be kitbashed into a reasonable model of a pennsylvania N8 class caboose.....

metal wheels and body mounted kadees make a world of difference.

i would not use the snap switches or any other switch with an 118" radius curve as part of a crossover between your two loops. why? well unlike most switches, the curved route maintains that 18" curve all the way from the points to the end of the switch. using two of them in a crossover is like making an 18" radius s curve with no straight between the opposing curves. this is a HUGE mistake as explained in any book on track planning.
no wonder your 2-8-4 has trouble with that crossover.

numbered switches, even #4, have a short section of straight through the frog, and thus avoid the s curve that wreaks havoc on your equipment. any number of switch will be a major improvement, the larger the number the better. but even if you are short on space, using a #4 crossover will use up only slightly more room than what you now have, and you'll like the decrease in derailments that will result.

atlas has a number of layout plans covered in their books, of which previews are available on their web site. they also have a free track planning software you can download and play around with.

http://www.atlasrr.com

i wish bachmann would have a track planner like this for ez track.

that said, i find that i can usually come up with better ideaqs on my own rather than following somebody else's plan. other layouts are nice for inspiration, but there is nothing like designing and building your own.....

i don't know if there is a commercial lighting kit available for this set, but they are not that hard to make yourself. this is especially true if the layout you are running this train on is equipped with dcc.

check the model railroad sites for constant intensity lighting circuits.....

http://www.railserve.com

the resistor needed would be determined by the current you want to flow through the light bulb. what is the current rating of the bulb? how about the light function of the decoder? the lowest number of the two will be the determining factor.....

the appropriate formula is ohm's law,
basically voltage = resistance (ohms) x current (amps)
the voltage you would use would be the decoder output minus the bulb voltage.
current rating was covered above.

ohm's law transposed to determine resistance thus becomes:
voltage divided by current = resistance.

the 18" and 22" curves atlas makes are designed to be as close to 9" as possible and still have a reasonable number of pieces peer circle (12 for the 18", 16 for the 22")......if you are trying to figure out a grade, 9" per section is a reasonable estimate.

my setup is a 1k ohm resistor soldered to the rails of the programming track, with test leads connecting it to the zephyr. the test leads might be the problem, i will have to try a more permanent connection.

i was unaware of blast mode. since my layout still has the dc block switches, isolating all other locomotives is not a problem, i can simply turn off track power to any section that has a locomotive in it. i purposely left the dc wiring in place for ease of troubleshooting, and as a backup if anything ever happened to the zephyr.

atlas makes a series of electrical components to do just what you want to do. there is one called the selector, which also has the provision for a second power pack. using the selector, either of the packs can control any track connected to it. the selectors can also be ganged together to control as many tracks as you wish. instructions are included in the package. atlas also has a wiring book for their components which i think does a much better job of explaining wiring a layout than any of the kalmbach books.

Quote from: scott V. on April 20, 2009, 04:56:20 PM
I tried the 1000 ohm resistor on my programming track with the pr3 and it didn't make any difference.  I used JMRI and also tried the soundloader software on 3 different engines with bachmann decoders.  It would write to them but not read.   I'm going to change them out for digitrax decoders.

Scott

this has certainly been my experience with the dcc onboard locomotives. the 1000 ohm resistor may work sometimes, but it is not reliable. i have not heard of any other workaround. you'd think bachmann would have come up with something but nobody here or anywhere else can tell me anything other than the 1000 ohm resistor.....

i have a friend do all my programming  for me on his nce system. once programmed, the locomotives work fine.....

are you absolutely sure it is decoder equipped? dcc ready is not the same as having a decoder inside. that buzzing you are hearing sounds to me like a dc locomotive on dcc powered track.

most should be 33. the 100 ton hoppers may use 36, because that's what the real ones use.

i am assuming that the hook and loop couplers on the hogwarts set are the same ones used on the thomas cars? if so, i don't think that the coupler you  linked to would do the job. however, you may be able to take a standard HO car and add a hook and loop coupler to one end of it. i don't have any of the hogwarts cars to look at but you mentioned they are truck mounted couplers. the thomas cars i have are all body mounts, and it would be difficult to add a knuckle coupler to one of these.

it would also appear that the coupler height on the hook and loop type couplers is not as critical as it is with knuckle couplers, so i would use a standard ho boxcar modified with one of these:
http://store.bachmanntrains.com/index.php?_a=viewProd&productId=30

you will have to carefully cut off the original coupler mounting, and drill new screw holes for the hook and loop but with a little patience you could get it to work. we used cars like this with a horn hook on one end and a kadee on the other before knuckle couplers became standard. they come in handy when you need to mix your equipment