How to wire a Lionel Gateman for DC ?

Would a buck converter connected to the coil contacts under the roof work to control the coil noise with DC current?

Whatever you end up with, keep in mind that DC at the same voltage through the coil will produce a lot more heat, there's no inductance issues to block the current.  I'd use a buck converter and crank the voltage down as low as you can go and still have reliable operation.

I use this from radio shack. Ground from track is isolated. I have 4 pre & post war models I use on the children's layout at TCA meets. All run on track power triggered by insulated rails. Quiet and smooth.

I agree these are very noisy. Its the only post war accessory I used a diode bridge on. A bridge alone will reduce noise about 80%. You will need to add a good size cap to get close to pure DC. These draw a lot of current some some ripple will always be present. Buck convertors are cheap so its good way to go. Alternatively use a separate transformer and crank down the voltage or get a DC supply. Desktop computer supplies are very cheap or free if you remove them before recycling them. 

Pete

John and Pete,

I've got a separate transformer for my accessories. Each arm (KW) controls the 2 halves of my layout's accessories. So, I don't want to go with another power source. Any time I've had to reduce the voltage to an individual accessory (the Oil Drum Loader for example), I've used a buck.  So......wiring this ........should I have the two leads from DC out going to Coil "hot" and coil gnd?  I'll probably take my power "in" leads and split it to one goes to the buck to control the GM coil and the other half controlling the lamp and the 69 signal.     Yes.....the GM is very noisy. It was very quaint, but now that I've got that bell signal in there, you can barely hear the bell when the 153 IR triggers it and the GM.         Will wiring the buck in in the above fashion effect the NO lead from the 153 in any way?

Oh....one other thing......is the cap in the buck converter going to be sufficient? I think it's 100uF, but I'm not sure. Been awhile.

Roger

Roger, the cap size will depend on the amount of residual buzz is OK. If 100uf leaves it still too noisy then add more. You really can't hurt anything unless you hook it up to line voltage.

Pete

GRJ, in this earlier OGR post you warn about magnetizing the plunger from DC operation thereby causing the mechanism to stick.  Do you have any additional info on how quickly this occurs?  Also, in the linked post de-magnetizing techniques are proposed - curious if anyone has actually implemented such a scheme.

I've seen that happen to E-Units, and a couple people mentioned the gateman having the same issue.  You can degauss them by using AC on them for a few cycles, at least that worked with E-Units.

Another thing that works with E-Units is a bumper inside the coil at the top, it would probably work for the gateman as well.

gunrunnerjohn posted:

...

Another thing that works with E-Units is a bumper inside the coil at the top, it would probably work for the gateman as well.

Are you referring to dulling the sharp impact when a plunger in a DC field re-aligns its magnetic domains (i.e., the high-school science project on how to magnetize a nail)?  I'd think this would be the simplest if some simple rubber washer, bumper, or whatever could soften the blow so to speak.  

gunrunnerjohn posted:

... You can degauss them by using AC on them for a few cycles, at least that worked with E-Units.

… like back in the day when we periodically degaussed the TV screen or tape-recorder heads.

For the gateman, a momentary 2-position DPDT button switch could be used.  It would normally route the 2-wire DC voltage to the 2 coil wires.  When the button is momentarily pressed it instead routes 2-wire AC voltage to the coil.  I don't know how often you'd have to do this but I suppose a good time would be whenever you see the gateman starting to stick!  If going under the hood to modify the gateman anyway, I'd think this would be prudent low-cost insurance.

I think I may be doing something wrong. Correct me please if I am.  I took out a buck converter and set it up to output 12V. Wired it with alligators into the GM and it all worked, but I didn't notice any big change in the noise. So.....I started small (and cheap) with a 100uf cap added in downstream of the diode between it and the buck. Not much change. Added a 450 and it was noticeable but still loud. Then I added a 4700 cap and the sound was less but definitely was not quiet. Can that be right?  It's the biggest cap I've got on hand.

Roger

So you're using a single diode (vs. a bridge rectifier) to convert AC-to-DC before the buck converter?  In which case you are only using 1/2 of the available AC voltage.  You'd need a relatively large capacitor (thousands of uF) to capture enough energy on the active half-cycle to supply power during the inactive half.  If you don't have a bridge handy, do you have 4 loose diodes?  You could wire them up into a bridge configuration.

Also, exactly which buck converter are you using?  Also, please confirm your capacitors have sufficient Voltage rating (like 35V).

Stan,

I'm using a 3 amp diode just for testing purposes. I do have a bag of bridges if that will work better. I did attach one to the GM earlier today just to see how it would work. But using alligator clips to hold that together and then alligator onto a buck was real cumbersome. But I'll get one out and wire it in. But I think I'm wiring the capacitor wrong. Where should the cap I add go exactly and how should the polarity on it connect?     I admire how you guys know which electronic component is necessary as well as the size. I'm real good at soldering this stuff in, but knowing WHAT to use and HOW to connect it is always a source of head scratching. I've got a boatload of electronics I've added over the years to my layout, but definitely needed you electronics guys for your input on what to use. A ton of fun.

As for the buck.......I'll have to go down and verify, but 7812 rings a bell. But I could be wrong. And all the caps I've been trying are either 35 or 50V compatible.

Yep, that's right.  I see Stan beat me to it.   Use a bridge rectifier and isolate the hot and ground of the gateman as you do loose the common ground.  Also, run at the lowest DC voltage that reliably triggers the gateman, as I said, you're pulling more current with DC through the coil at a given voltage.

7812 is a linear regulator, not a buck switcher!  That is likely to get really hot if you put any significant current through it!

I was mistaken.     DSN2596   DC to DC.

Two questions.....

1. How exactly do I isolate the hot and ground?

2. How does the added cap get wired to the bridge and buck

Thanks!

Roger, how many times an hour or minute will the gateman get activated? If its only every 5 minutes for 20 seconds all this extra circuitry is not needed. A bridge and a capacitor will do it. If this is on a 4x8 loop with gateman on 3-4 times a minute then you do need to consider heat build up.

Pete

Bridge rectifier, AC into the AC terminals, + & - terminals to to the cap and then on to the DC-DC supply.  Isolate the gateman by making SURE it doesn't connect to anything else on the layout, including the frame of the gateman.

Pete, it's also how long it's activated!  The issue is, if you happen to park something on the insulated track, you can burn it up.  I've seen it personally with a customer's accessory.

If you have a chance could you measure the Ohms of the coil (nothing else hooked up)?

Also, once you get the buzzing issue resolved (you're really close!) could you experiment with dialing down the voltage output of the buck regulator and letting us know how low a Voltage you can use to reliably trigger the mechanism?  When you find the magic voltage, I'd dial it back up, say, 2 Volts for operating margin.

John,

The only thing it's connected to is the 153 IR . Oh, and the 69 signal (but that is connected only to the GM.  Do I wire + out on the bridge to + in on the cap?  And same question for the cap to buck polarity.  Thanks

Stan,

I'll dial the voltage in and post it on here.   All of this to hear the silly bell on that prewar signal. Quaint comes to mind.       I will use the dialed down buck even though I don't use the GM much. I do it for friends and company and I'll leave it on for the session. Otherwise, the silly thing sits.

ROGER1 posted:

...Do I wire + out on the bridge to + in on the cap?  And same question for the cap to buck polarity.  

Yes and yes.

Perfect, Stan.  Exactly what I needed.    When this is done wrong, those caps last about a micro second before they burst. LOL

Well, hopefully your were protecting your good eye!  Fortunately, what I figure should be a suitable capacitor only costs a dime or maybe a quarter on eBay.  

I'm planning on wiring this thing up this afternoon, but I've got another question. I'm thinking I'd like to split the power feeds (hot and ground).....one pair going to the buck for the coil and the other pair branching off to control the bulb and 69 signal (as they would probably do better with AC). Will that work and will it effect the "isolated ground" or buck in any way?  Thanks

Roger

Why would a bulb have a problem with DC?

You could do that as long as the 2 wires to the coil are not connected to the bulb.  You could meter this to see.  Hard to tell from photos of underside of chassis posted earlier.

That is, if I understand what you're saying you want to have the 2 "yellow" AC IN wires in my diagram going to the bulb (and 69 signal)...but the 2 DC OUT signals on the right going to the coil.  The point being the metal frame of the accessory may be connecting one of the AC IN signals to the DC- OUT signal.  This would release the magic smoke.

Why would a bulb have a problem with DC?

It will be dimmer at the 12V DC feed

Stan,

I see what you mean......I don't like "magic smoke".  Maybe I can disconnect the bulb circuit and connect it solely to the AC feed

It should be the same intensity with 12VAC and 12VDC.

I've actually got around 16V AC going in. Coming out of the buck will be 12DC.  I guess I could live with the dimmer bulb, but if I can separate the two circuits, I'll do that.

I'll see how it looks at 12V.   If the bulb will be dimmer, maybe that's a better compromise than trying to isolate that bulb circuit. Alot of metal plates/contacts under there.

Here's what I'm finding. The noise is reduced with the caps but is far from silent. Even with the 4700uF cap. I think I'll use the 450 as it will be easier to fit in there. Most of this will not fit under the roof.   Second thing......a good part of the noise is mechanical vibration carried by the moving parts including the gate guy. Trying to fit foam and Goop in strategic spots.

Stan,

Trying to get a read on the resistance of the coil....a bit difficult as it's hard to get completely in there with all the wiring attached. It's hard to feed those wires in there and I'm hesitant to take them all out for a reading. I'm getting around 8 ohms. Can that be right?

Well, I had a Eureka moment. It was becoming more and more apparent that most of the noise was being caused by vibrations of loose parts underneath. I tried putting foam in all sorts of places to dampen them. Nothing had an appreciable effect and the ones that did, impeded the movement of the cam with the guy attached. Then I found it.  Where that cam hits the side of the frame (screwdriver tip pointing to it) was the sweet spot. I taped a piece of foam there and the difference was amazing. So.....I just Gooped a better piece of foam to the base plate in that spot. I also replaced the c clip with and e-clip (it's thicker) to reduce vibration from there. But the main spot is where that cam hits the side of the base. 

Going to attach my wiring now. Alot of the parts will have to go in the base, but I plan on putting the buck under the roof, so it can be adjusted. 

If it all works, I'll post a vid.             If not.......I'll smash it flat. LOL

ROGER1 posted:

… Trying to get a read on the resistance of the coil....a bit difficult as it's hard to get completely in there with all the wiring attached. It's hard to feed those wires in there and I'm hesitant to take them all out for a reading. I'm getting around 8 ohms. Can that be right?

8 ohms seems reasonable.  Power in Watts = V x V / R = 12V x 12V / 8 Ohms = 18 Watts going into the coil.  I'd still like to know if you can dial down the 12V buck output voltage and get solid action with 11V, or 10V, or whatever.

Separately, one reason to drive the bulb with AC (rather than buck DC out) is this means the capacitor does not have to be as large.  Unfortunately, the effect is probably small.  That is, the bulb is probably a Watt or two (?) which is small relative to the coil power.  To confirm, you can simply disconnect the bulb and see if the buzzing changes when activating the coil without the added bulb loading.

Stan,

I didn't include the pic of the "sweet spot", but something else popped up that may make it irrelevant. Good news, bad news thing. I wired up the buck in the roof and I had the cap and the bridge under the base and connected them. Alligatored to my power supply and then weird things happened. Noisy operation of the door. Went back to straight AC and there was less vibration. In the pic below, the bottom two terminals are power in. The single one at the top is coming from the NO terminal on the 153.    I switched the red and black wires to the bottom terminals and got no operation of the door (but I still had the light). But here's the thing........I took an alligator and connected bottom right to the top and the door opened and.........SILENT!  No buzz.  So......how to get this to work now. Somehow, changing the wiring to DC has effected that NO terminal.  When wired for AC, it all works when the train triggers it. In DC, it is either very noisy when triggered or......if I switch the wires at the bottom, the train doesn't trigger it, but a jumper makes the door open and it's silent. Something about the wiring seems like it has to be different for the train to trigger and the door to be silent. Any ideas?

Have to add one thing here.....when I say connect the jumper to those two terminals, the door immediately opens and with no buzz. It's opening without the train triggering the 153......contact from the jumper becomes the trigger and the door opens.   ?????

I've got an idea about this......I think it's that wire from the NO terminal.   There are 3 wires coming in from the 153......Power, GND and the NO. The power and ground wires are going in to the bridge, cap, and buck. The NO wire does not. Can it be that it also has to go through that circuitry?  

I found this diagram from the Lionel gateman manual.  Your photo is "upside down" but will give my explanation based on your orientation.

ROGER1 posted:

... Alligatored to my power supply and then weird things happened. Noisy operation of the door.

You apparently have NO connected to upper right terminal.  Your have red DC+ to lower left and black DC- to lower right.  So the lamp receives DC+ and DC- and should light up.  When 153IR triggers, AC PWR appears on NO.  This means the coil is receiving AC PWR and red DC+.  This is bad!  That is the coil is receiving half AC and half DC.  This is actually worse (more buzzing) because this generally means the mechanism will vibrate half as often.

Went back to straight AC and there was less vibration. In the pic below, the bottom two terminals are power in. The single one at the top is coming from the NO terminal on the 153.    

That makes sense.  By hooking up straight AC, the coil now receives AC PWR and AC ground... in other words you aren't mixing AC and DC as in above case.  The vibration frequency is twice as fast but is effectively smaller.

I switched the red and black wires to the bottom terminals and got no operation of the door (but I still had the light). But here's the thing........

That makes sense.  This just swaps the valid DC+ and DC- wires to the bulb so the bulb still gets a valid DC voltage.

I took an alligator and connected bottom right to the top and the door opened and.........SILENT!  No buzz.

Right.  The alligator means the coil now sees DC+ and DC-.  So DC on both inputs.  No AC.  No buzzing.  

So......how to get this to work now.

Exactly.  And that's the problem.  When the 153IR triggers, it applies AC PWR to the NO.  This would mean the coil sees a mix of AC and DC which is bad.  

Somehow, changing the wiring to DC has effected that NO terminal.  When wired for AC, it all works when the train triggers it. In DC, it is either very noisy when triggered or......if I switch the wires at the bottom, the train doesn't trigger it, but a jumper makes the door open and it's silent. Something about the wiring seems like it has to be different for the train to trigger and the door to be silent. Any ideas?

My question to you.  I can't tell from the photos, but is it at all practical to separate the coil and bulb so that there are 4 wires going into the gateman.  In other words 2 wires to the bulb, and another 2 wires to the coil.  Presently the gateman has 3 wires/terminals because it shares one terminal with the bulb and coil as shown in the Lionel diagram above.  For example, is the bulb in a socket?  Can the socket be lifted off the chassis so it is electrically isolated from the metal frame (and hence not electrically connect to the coil on the common terminal)?

Stan,

I had a feeling that was the case.   I thought of two other options I'd like to run by you. Probably neither is ok and it's ok to trash them.

1. Will the 153 run on DC. I'm doubting it, but if it could, the electronics to convert to DC could be the feed into it.

2. Would this work......diagram below