Stalling at Switches
- Thread starter varillon
- Start date
What kind of switches do you have? What scale? What kind of engines or is it every engine? Do you have feeder wires going to each switch or are you relying on rail joiners? Does the engines stall only on the diverging route or on both routes? Are you running DC or DCC? There are lots of reason for stalling in switches but I need more informarmation to give you an educated guess.
The switches are Atlas remote, though I haven't hooked them to electricity yet. It's HO scale. The engine I'm running is a 4-6-2 Pennsylvania steamer with DCC. I haven't run feeder wires to the switch itself. As it sits, the switches are connected to other track only on the straight side. I'm only getting the main line setup at this point and will run the rest of the track later. The closest place where feeder wires are connected is about 6 feet away, and the wires are high enough gauge. I've run 14 gauge under the entire layout with feeders of 18 or 20 gauge at less than 2 feet long, so no worry of power loss.
mike8693
Member
6 feet! I think you may have just found your problem or atleast part of it. I usually try to add a feeder ever 3 feet or so.
Thanks for that information. I assume these are Atlas switches with the black plastic frogs. Watch your engine carefully and see if it's stalling at the frog. If that's the case, see if the lead wheel is dropping down in the frog, causing the drivers to lift slightly off the rails and stall. If this is the case, a thin strip of styrene, cut to size and glued to the bottom of the frog, is usually enough to prevent the wheel drop.
The other problem is frogs where the gap is too tight, causing the wheels to bind. a few passes with a fine flat file will usually widen the frog enough to fix that problem.
If the engine stalls as soon as it reached the rail joint of the switch, the problem is electrical. Use a small flat blade screwdriver and move the rail joiner foreward or backward a little and see if that stops the stalling. If you have an ohmmeter, you should be able to see where the current drop to near zero on the other side of the joint also. If this is the case, either solder a jumper wire between the track and the switch joint or pull another power lead and solder it to the running rail of the switch beyond the joint.
The other problem is frogs where the gap is too tight, causing the wheels to bind. a few passes with a fine flat file will usually widen the frog enough to fix that problem.
If the engine stalls as soon as it reached the rail joint of the switch, the problem is electrical. Use a small flat blade screwdriver and move the rail joiner foreward or backward a little and see if that stops the stalling. If you have an ohmmeter, you should be able to see where the current drop to near zero on the other side of the joint also. If this is the case, either solder a jumper wire between the track and the switch joint or pull another power lead and solder it to the running rail of the switch beyond the joint.
Thanks for the tips. I had thoughts of wheels moving away from the top of the rail but didn't really get where it would occur. It does have the black plastic frogs, and I've read that there may be a momentary loss of power but not enough to completely kill the loco. I figured there had to be ways around even a minor loss of power. Definitely glad this place exists for some of those mind boggling issues.
BTW, now that I've had the chance to actually hear the sound coming from this engine, I'm astonished at the quality. This is setup in a spare bedroom, so the sound echos. I actually made my boxer howl when I hit the horn button on my NCE controller lol Have to keep him out of there when I do that.
BTW, now that I've had the chance to actually hear the sound coming from this engine, I'm astonished at the quality. This is setup in a spare bedroom, so the sound echos. I actually made my boxer howl when I hit the horn button on my NCE controller lol Have to keep him out of there when I do that.
Well, here's where I'm at. Before adding another feeder, I watched the engine roll over the switch. When the first set of large wheels reaches the middle of the frog, the loco stops. This was running at minimal speed. Remember this is a 4-6-2, so the first 4 small wheels aren't at the cause of the stall. After adding the feeder, I can cross the switch at full speed. Anything slower, and it completely stalls out. At this point, I took a dremel with a cutting wheel and lightly nudged the inside of the frog where it meets the rail. I felt just a touch of lip there by running the dull edge of an exacto blade through. No wheels are lifting or sinking. I think things have improved overall because the loco stalls after it's gone through the frog now, and then starts back up after a pause. Still, I would like to improve it if possible but will live with it if I must. I'll also add some more feeders as I get the oval completely laid.
Edit: I completely forgot to check the power. I'll do that next.
Edit: I completely forgot to check the power. I'll do that next.
Last edited by a moderator:
You should see a no load voltage of somewhere between 14 and 18 volts. What is the no load voltage on the track before and after the siwtch on both the main and diverging routes? Does the voltage change depending on the position of the points? If voltage is in the normal range on the tracks on either side of the switch, you've got an electrical problem with the switch itself. The switch itself may be bad so the first thing I'd do is try changing it out. If you still have the same problem,you may have to wire the frog for power with a short wheelbase steam engine. See http://www.wiringfordcc.com/switches_atlas_roco.htm for a very good discusssion of Atlas switches and some wiring solutions.
varillon,
sometimes the rivet in the pivot of the points on Atlas TO's does not make good contact. You may be able to get away with spraying it with contact cleaner or dousing it with a little alcohol. If not, you can either solder a small wire from that point to the main rail or I use a small shim of brass cut from a rail joiner and forced tight into that small gap between the pivot point and the main rail.
If you measure near 14 volts across the rail at the TO, then you don't need to worry about adding more drop feeds. 6-9 feet is plenty and I have some farther apart than that without trouble. You need to use a digital volt meter and set it on AC volts, 20volt scale. The voltage is not pure DC, but a bipolar squarewave.
sometimes the rivet in the pivot of the points on Atlas TO's does not make good contact. You may be able to get away with spraying it with contact cleaner or dousing it with a little alcohol. If not, you can either solder a small wire from that point to the main rail or I use a small shim of brass cut from a rail joiner and forced tight into that small gap between the pivot point and the main rail.
If you measure near 14 volts across the rail at the TO, then you don't need to worry about adding more drop feeds. 6-9 feet is plenty and I have some farther apart than that without trouble. You need to use a digital volt meter and set it on AC volts, 20volt scale. The voltage is not pure DC, but a bipolar squarewave.
Last edited by a moderator:
Well, the voltage issue is solved. I was using 2 meters and fudged something up when switching back and forth to AC and DC. I'm getting a steady 18.97 volts everywhere except at a crossing that is only connected one way. The connected route is at full voltage.
There's one thing that isn't making sense in my mind. Since the frog is plastic, how can adding power to it make any difference? The wheels will only touch plastic; therefore no electricity will reach them.
There's one thing that isn't making sense in my mind. Since the frog is plastic, how can adding power to it make any difference? The wheels will only touch plastic; therefore no electricity will reach them.
I did actually which made me write the question lol I don't see how driving a screw into plastic makes any difference (Option 1). The description for Option 2 says the frog rails and closure rails are one piece. If they touch, though, it would cause a short. What am I missing here? Do these rails change shape inside the frog so that a plastic buffer separates them? There's just no way they could stay the same size and shape without touching.
hamltnblue
Active Member
It's also very possible that your problem might be the engine shorting out at the frog. I have had problems with 3 different atlas switches that needed some tweaking. If you have a second engine run it the same time. If it is a short when the first engine stalls at the turnout the second one will stop also, no matter where it is. If you have a volt meter the voltage will drop when this happens. What was happening with mine is sometimes a wheel flange would touch both rails where they cross at the frog. This can be caused by 2 things. A wheel set out of gauge or a turnout that has a narrow gap. If it's the turnout (and it usually is) use a screwdriver to separate them some or a dremel to widen the gap just a little would help. Some have also used fingernail polish to pain the inside of the rail where they cross.
Your 6 foot spread on the feeders should be more than fine also.
Good Luck
Your 6 foot spread on the feeders should be more than fine also.
Good Luck
As I wrote: The points are picking up power in the pivots. These pivots don't always have good contact with the small brass plate. Its not a common occurrence, but it does happen. I believe that I have had two out of a couple of dozen do this. Beware of Dave Gartners suggestions. While very thorough, they are ridiculously complex and full of overkill. If you have to do all that stuff he suggests, its better to spend $14 on a new turnout.
One thing you can do on the turnout: remove power and connect one meter lead to one "point" and the other to its common rail. Put your meter on ohms. Measure for continuity as you manually move the points back and forth. You should maintain continuity during the entire move. If not then you have a bad connection at the pivot or there is a manufactured flaw in the built in jumper to its power rail. The only repair for the built in jumper is to make your own jumper wire and solder it in.
Along with the Atlas frog sometimes being too deep, it also has had a hump that lifts the wheels slightly. Watch the locomotive wheels carefully and move it with your hand. You should see or feel if this may be the case. Filing it deeper will eliminate this problem.
Your 4-6-2 has a long enough wheel base that an insulated frog should not affect it.
Your not making power as that is impossible. If you are supplying 14volts, that is all there can be anywhere. There is probably something wrong with the meter.
Blue gave an excellent suggestion. Are we only talking about one locomotive? If so, then wheel gauge could very well be the problem and you are actually momentarily shorting instead of losing power. What happens to the other parts of the layout when you have a stall? Does it go dead?
This made me think of something. Two things happen if I have a short. The light on the power panel goes out, and there's a hum. I let the engine stall, and sure enough I got both. I did the continuity test and don't lose any ohms throughout the pivot. I don't have a second engine, so that test is out. I'll have to examine the turnout with a magnifying glass to see if anything is touching with the wheels on the frog. I know there is a gauge for making sure the wheels are right, but does that gauge also shape the wheels if they are slightly off? Also, what gauge would you guys recommend?
Yuo need an NMRA gauge, which measures a number of things, including wheel guage and frog and guard rail clearances. You can get one at most hobby shops or order on on-line at http://www.toytrainheaven.com/.sc/ms/cat/Track Gauges--NMRA. If the wheels are out of gauge, you can usually just drop the wheel out of the engine and push or pull gently to move the wheels on the axle. Most wheels are on a plastic sleeve over the axle and can be moved a little either way with no damage. Just make sure that you keep all three drivers so they are not only in gauge individually but also in line with each other so you don't get the engine wobbling down the track.
I have made the suggestion before and Rex reinforced it to just swap out that switch for a new switch. Some Atlas switches just don't work right no matter how long you fool with them. They make tens of thousands of those a year and some are bound to be bad. If a new switch does exactly the same thing, then you have another problem but I suspect a new switch will solve your problems.
I have made the suggestion before and Rex reinforced it to just swap out that switch for a new switch. Some Atlas switches just don't work right no matter how long you fool with them. They make tens of thousands of those a year and some are bound to be bad. If a new switch does exactly the same thing, then you have another problem but I suspect a new switch will solve your problems.
