Atlas Switches burning up with DCS Remote and AIU

the thing with atlas switches is they must throw immediately if they hang up the coil can burn out you must check all your switches and move them by hand you should move them in both directions and should move easily with your hand if there is any binding at all there susceptible to burn out if energized to long the coil starts to melt and will shrink around the piston and will then cause binding. I would check all your switches manually and if they move freely then energize one at a time and watch closely that they throw both directions immediately! I have had this issue my self. also sometimes some coal or little pieces of debris will get between the switch rails and interferes with the switch movement! keep us informed and let us all know how you made out.  

Is the AIU set for accessory.  You may be getting constant voltage and not momentary pulse?

Alan  All switches are smooth by hand with NO binding.

Bryant They are on the Switch part of the AIU not the Accessory side.

As stated above the other 8 on the other table work flawlessly. These are wired the same but out of the 7, 3 have been replaced with one being done twice. Please read #3 above especially the note about not being used.

I do want to add one more question. What voltage are you using if you have Atlas switches to power them? Atlas states 12 to 18 volts.

If a switch had no manual binding, but still the motor won’t move it and heats up, then the motor must be binding internally?

I use 18 VAC on my Atlas switches. I have 11 and have yet to burn up a switch motor. I don't have an AIU though, all mine are run with the Atlas switch controllers that came with the switches. I will probably add an AIU one of these days, but the switch controllers are serving me well for now.

As said above, the AIU should only put out a pulse of electricity to operate the switch. It would be a momentary contact. That would be like 1/2 to maybe 1 second or so at the most. (I think it's 1/2 sec. for the AIU, but can't remember without looking it up) The power should then be removed (turned off by opening of the AIU relay). If the power is left on or sticks on then you burn up the switch machine. This is true for all twin coil switch machines, which is what the Atlas machines are.

If the switch itself is sticking then it just doesn't throw all the way. It should not burn up because the controller should remove the power after the momentary contact. At least that has been my experience over the last few years with my Atlas switch machines. 

Some of my switches were purchased used and the switch machines would not fully throw the switch. The switch itself worked fine with no switch machine connected to it. I got new springs, throw bars and other parts for the switch from Atlas and it made no difference. So I then replaced the sluggish ones with new switch machines and all was well. They have been fine ever since then.

As cjack said above, they could have been binding internally, but I never figured it out? So I don't really know why the older used switch machines were so sluggish? One test I tried was using 18 VDC on them. That did seem to improve their operation somewhat, but not like the new switch machines. That was a few years ago and I forget what else, if anything, I tried when fiddling with them. 

Keep in mind that you cannot set up non derail on atlas machines using insulated rail. Every wheel in the train will be electrifying the machine causing over heating.

RTR12   You are correct in your assessment of the time and purpose of the AIU. As I stated the other table works excellent. Also have had an experience with switches being sluggish and have changed motors which corrected the problem. That is not the case here. As for binding neither the track or the switch motor is binding. This is why I am so confused. Even changing motors makes no difference as it will work, then not work, plus it seems the problem moves among 3 switches only. Even had local hobby shop repair man come look. He is just as stumped as I am.

Romiller49 _ we aren't talking with trains. This is just trying to switch directions on the switch with NO TRAIN ON THE TRACK.

I appreciate all the suggestions. I learned years ago sometimes the simplest thing is overlooked. Hence why I have taken so much time rechecking everything several times. I had a guy search for days on a problem of no power on his layout. Finally I went over to check and did the very first thing you should check which he constantly skipped. I plugged the power in. I told him don't feel bad it happens to everyone. You get wrapped up looking for something complicated and it is staring you in the face. Been there plenty of times. That is why I love this forum. There is no wrong answer just ones to check. Thank you everyone for helping. I appreciate fresh minds.

I guess without seeing it I am also pretty well stumped. It is a bit puzzling to me too. Not that seeing it would help, but as you say could be something overlooked that someone else might notice right off? (I have done many things like your friend with the power not connected.) I'm no expert, but the switch machine wiring is not terribly complex, that is without the non-derail or any thing else in the mix. Something has to be leaving power on to the switch machines to burn them up. 

Other things I can think of: Can you try swapping one problem switch with one that works properly? Or swap the AIU outputs on them. Does the problem move with the switch or does it stay with the AIU output? Recheck the AIU setup for these switches. Can you hear the AIU relay switch on/off momentarily when you work the switch? Are the Atlas switch controllers wired in with the AIU relays so you can operate the switch from either one? 

Romiller has a point if you have them wired for non-derail, something could be wrong there? I got a couple of 6924s to do that myself, then decided the switch's springs were good enough for what I was doing and never installed them on the layout. Kind of like I did with the AIU. I did fiddle with the 6924s a little, got some good help here on the forum on their proper wiring and had them working on the bench. There is a lot of extra wiring needed for those.

If the first table works and all but the 3 switch machines work on the 2nd table as well, there has to be something different with the 3 problem machines. Wiring, AIU relays, wire crossed, mis-connected, anything else that might be in the circuit, etc. I know you already know all this, just trying to think of anything that might be in the way or different somewhere, anywhere.

Have you tried taking a switch out (if possible) and bench testing it with no AIU, then with the AIU, but still on the bench? If you do the bench testing, try upping the voltage, if you are at less than 18 volts now. I have not tried more than 18 volts on any of mine so you may want to limit things to no more than that.

Also if you used multi conductor wire, the inner wires could be shorted inside the outer jacket somewhere between AIU and switch machine. We had this happen to us a couple of times during my working life. Very ugly and difficult to find. It usually caused intermittent problems which are the worst kind.

Maybe this will offer up a couple of more ideas for things to check? Then again maybe not? Anyway, Good Luck and please keep us informed on the problem and how you fixed it if you find the problem. 

As rtr12 points out, the AIU SW ports apply power for exactly 1/2 sec when triggered - no more, no less.  It seems to me if you are seeing burning or heating in the solenoids, they are operating for extended intervals.  

You could attach a current meter(s) to the suspect switch(es).  The idea is to confirm power is applied to the solenoids only in response to DCS/AIU command.  But watching a meter can be a bother when you're operating trains.  And the brief current surge during normal solenoid operation can be difficult to catch on a meter - that is, even if power is available for 1/2 sec, the solenoid should automatically stop drawing power after successfully toggling direction - which should take less than 1/2 sec.

Do you have a CT donut lying around?  They are also available on eBay for about $1 or so (free shipping from Asia).  Put a pair of LEDs on the CT output.  Then run the AC power wire to the switch thru the donut.  A typical switch solenoid will draw 1-2 Amps when fired.  A typical CT might be a 1000:1 ratio so the CT output would be 1-2 mA.  This can easily drive any LED.

Obviously this is not an Atlas switch but illustrates the concept:  

Since this is something that appears to take some time to rear its ugly head, a tool like this (or a meter, or some other widget) would allow you to continue operating your trains requiring the occasional glance to confirm solenoid power is being applied only when commanded.  

RTR12  Atlas switches are 6072. No there is not any other controller hooked up just the AIU.  Ironically I have switched position of switch on AIU. The results were the switch worked ok SHORTLY then went to same pattern and when switched back again worked shortly then back to heating up. As for the AIU I have removed it and the local Hobby Shop has it as he wanted to take it and do extensive testing on it. NOW REMEMBER I HAVE SWITCHED THIS AIU OUT AND THE NEW ONE IS DOING IT ALSO. Just adds tot he confusion and frustration. Interesting you mention voltage used to throw switches. When I was doing the original table my local dealer suggested moving the spring in the switch to the  furthest position (this lengthens spring and tension) and to use 20 volts (which I am using an old TYCO HO Transformer for (accessory side)). Now remember this was done first on the original table side and is still working great. I also considered the multi conductor wire I am using and wondered if this ATLAS wire is at default with an internal short but thought hey it's Atlas's wire from a 50' spool. I checked with an ohm meter and came up empty no shorts. My dealer has told me if he finds nothing wrong with the AIU he has a relay device that he wants me to install that will only let the voltage thru monetarily I appreciate your help. I don't know how people survive without this forum.

Stan2004 Man am I impressed not only with your answer but also the pic and VIDEO (which speaks volumes). As noted above I am now in a hold till the dealer gets done but definitely looks like I will be buying parts to manufacture this simple device.

The AIU may be cycling through more than one intermittent/momentary power, which is enough to fry an Atlas switch motor.  Some have changed the Atlas motors to DZ 1000's.  The DZ switch motors have a built in cut-out switch, that limits momentary power.  Atlas also provided, at one time, 6924 relay boards,  that limited momentary power.   

Atlas 6924 relay board.  Board can operate a switched crossover pair, and also allows for logical dead rail power routing.  There are other features, related to switching, that can be done with this board. 

Thanks Mike

This is all very puzzling? Only thing I can add is that I left all my springs at their default settings. I would have to get into one to say what that is for sure. As I recall, I never tried moving the springs on the switches with the weak switch machines. I have used 18vac from the time I installed them, still using the same voltage. I also have a few 6072s the rest are O-54s - R & L. I have not noticed any difference in operation between the two switch sizes.

Stan always has something up his sleeve to help with these things. I'm a loyal follower! He explains things very well too. He comes up with some very ingenious ideas and they always work great. I may also try these devices as I really like stuff like this and also enjoy fiddling with it. And you can also learn a thing or two. 

The DZ-1000s Mike suggested are also nice switch machines. I had one place where I could not fit my Atlas switch machine so I got one of the DZ-1000s. They are nice, work well with the Atlas switches and are not twin coil so they don't overheat. I think they are also now about $25-$26 each street price.  

I am a big fan of Atlas track and I'm anxious to see what your dealer finds and which relay device he wants you to try. And also the results you get from that relay test. 

Late winter, early spring 2014..... (I'm sure a few of you will remember me posting these pics and this issue back then).

About 6 months earlier (Fall 29013), I had hooked up my switches to one of 3 AIUs....this started happening....

What we figure is that I was using a Z-4000 for switch and accessory power and a ZWL for my 4 power zones......the transformers were out of phase....that was the only thing we could find.

I have switched all 21 switches to DZ-1000s. I switched accessory power to a Z which has been phased to the ZWL. .....No problems since.

Peter

Putnam  Looking at your pics they are obviously burned. Mine show no effect of heat other than the piston not sliding freely anymore after the problem surfaces. This means even if lifted out of the switch and connected to nothing. Diffidently a heat problem. Interesting the phase comment came up. I have all my transformers regardless of manufacturer phased together.

TO EVERYONE FOLLOWING THIS I have been racking my brain as also the local dealers brain. He currently has one of the 4 TIU's I have. Two I have switched and both gave the same results eventually.  After reading all suggestions, which most have been done (not the little device with the led's yet), Putnam caused me to think of something although I do not believe it is the issue. Table one has a TIU (auxiliary powered) that the 4 AIU's come off of. Table two switches are powered by a separate transformer (same make and model as table one) but it is on table two. Tracks on table two are powered from Z1000 bricks while table one uses two Z4000's. My wondering is if power from the switch track is somehow back-feeding into the switch motors although I can not see how. NO wait a minute I just remembered the track the switches are on are powered but have a toggle switch to turn them off and they are off when this happened the second and third time so that is not it. Sorry just thinking out loud.

Not sure it matters in this case, but, do your toggles have built in leds?

Not built in but they have LED's to show power on.

Probably is not the problem on your layout then. I have some rocker switches with built in LEDs and the common feeds back through the switch to the power rail when the switch is off. If a train crossed from a powered track to an un-powered siding it tripped my PSX-AC and a second or so later the PH-180 breaker would trip. This drove me nuts until I finally figured it out after a few months. I fixed it with isolation relays, pretty easy fix, but finding it was a real job for me. 

I kept thinking I hadn't changed anything because I had plain old toggle switches (no LEDs) in there before. All I did was change to the rockers with LEDs. I finally figured out that did count as a 'real change' and a big one at that! Sometimes the simplest things... And that is also what I get for not making a wiring drawing first (or last) and just wiring things up anyway. 

I was just thinking you could have something similar with possibly a different wiring setup that could be causing your switches to heat up? My setup could not cause problems with the switches themselves or their switch machines as it was. But, I am never sure how a different configuration might cause problems elsewhere. Anyway, was just throwing that one out there...