How to wire a Lionel Gateman for DC ?

I did not include a buck and capacitor in the diagram for the left side because I'm thinking it will control the lamp and that won't need the buck. But if splicing those  - wires is bad, it's all moot.  I think the ground wire split over to the right hand bridge is ok (?), but splicing those negatives from the two sides......I don't know enough about DC. Would it work?

1. Yes.  I don't know if the manual says so, but I've taken a 153IR apart to see what's under the hood and it will most certainly operate on AC or DC.  I actually wrote up a teardown on OGR so it's out there with plenty of photos which I can dig up if you're interested.  To wit, the "AC" power inputs to the 153IR immediately go to a bridge rectifier to convert to DC!  And of course you can feed a bridge rectifier with DC and DC still comes out.

I was just drawing this up when you posted:

In fact, another all-DC option is to power the 153IR from a DC-brick adapter (wall-wart) with suitable power (several Amps to drive the coil).  These run between $5-$10 on eBay free shipping.  If the brick supplies, say, 12V DC you then wouldn't even need the capacitor/buck regulator though the bulb would then run at 12V DC which may or may not be bright enough.

2. Yes.  And wow!  I'm astonished (that's a compliment) that you'd come up that scheme!  An EE might "pooh-pooh" it is because the 2 "-" outputs from the 2 bridge rectifiers will be at slight different voltages.  Yet, you then tie them together (creating a "short circuit") at the common/shared terminal of the gateman.  This is generally considered bad practice.  But in this application it would work because of the modest levels of current/voltage involved.  

Stan,

Much appreciated! I love trouble shooting. My background is definitely not electronics, but science is. I was a Biology teacher.So,  I love the process. During my Yoga class tonight while I was supposed to be involved with the Asanas, I was mulling through the two possibilities. 

So, looks like I've got two options......DC into the 153 or my Rube Goldberg. Which should I do?  If something goes wrong with #1, I toast my 153.  As for #2......the buck is producing around 12V (at least until I do that fine tuning you requested), the other side would have closer to 16 (that's about what the KW throttle is set to for that quadrant of my layout).

Easy for me to say but you will not toast your 153IR by powering it from DC!  I suggest you use the diagram I supplied with the single bridge rectifier before the 153IR.

As you point out, both options will supply unsmoothed DC voltage to the lamp starting from 16V AC.  This lamp voltage of course is irrespective of whatever voltage you tune the buck regulator for coil purposes.  If anything you would need to lower the lamp voltage.  Obviously you have diodes and bridge rectifiers lying around.  Simply put one or more 1-cent diodes in the path to the lamp which will drop the DC voltage by about 3/4V per diode.  If you don't have diodes lying around you get a bag of 100 on eBay for 99 cents (free shipping from Asia).

Stan,

Well, it looks like the easiest wiring/connecting is option 1.....DC into the 153.  All of that circuit is already connected. I would just have to separate the bridge and put it on the other side of the 153.   I think I just have to be very careful which terminal gets those - and + wires in the GM.    This afternoon, I consulted the owners manual to see about DC operation and there was no mention. But I'm not surprised that you went in there on a component level and figured it out. 

Tomorrow I'll draw a diagram that includes yours and exactly how it connects to 3 terminals in the roof, so I'm sure I got it right. If that's ok. 

Oh.....one more question. I was doing all this so that my newly acquired 69 signal would be heard over the buzz. Since there's a coil in there operating the bell, I'm assuming that it also would work on DC power.  If that's the case, I have to determine which terminals that would connect to as well. 

This is VERY cool

For those who are following this thread, but don't want to do all the wiring to quiet the Gateman, below is the photo of the spot in the frame where a piece of foam can be glued to cut the vibrations pretty drastically.  The screwdriver points to the spot.

I just realized.......it's not going to work, Stan. Making the 153 work off DC can't be the plan.  The 69 signal......won't work on DC. I thought.....this thing will make one "ding" and that will be it. Ran downstairs and tried it.....sure enough.  No 60 Hz vibration. One magnet pull.....one ding. So.....going to have to go to plan B. A bridge for both outputs. Maybe use 2 bucks to match the voltages.  Have to think about this.          Tomorrow.....

Forgot about the 69 bell.   Not familiar with that accessory but found this video:

But, as a science teacher, here's a high-school science project for you.    Hook up the "raw" (no capacitor) output of the bridge rectifier to the bell.  This applies pulsed DC which I believe will operate the bell about the same as AC.  You still have your option of using 2 bridge rectifiers after the AC-powered 153IR.

OTOH, this being a discussion forum, if the 69 bell responds corrects to pulsed-DC, yet another option would be:

The 153IR as shown is being powered by pulsed-DC.  This pulsed-DC goes to the bell when the 153IR is triggered.  Pulsed-DC also goes directly to the bulb.  A new diode must be added before the capacitor otherwise the capacitor would smooth the pulsed-DC and the bell would ding once as you discovered.  The diode isolates the pulsed-DC from the "somewhat smoothed" DC that is needed by the buck regulator.  

I don't know what mix of diodes and bridges you have lying around, but if you only have bridges (which are of course 4 diodes packaged in a useful configuration), you can access just one diode as shown in the inset box.

But, as a science teacher, here's a high-school science project for you.    Hook up the "raw" (no capacitor) output of the bridge rectifier to the bell.  This applies pulsed DC which I believe will operate the bell about the same as AC.  You still have your option of using 2 bridge rectifiers after the AC-powered 153IR.

So, to see if that works, connect a bridge to my AC source and then take the two DC leads and connect it to the bell?  Correct?

Also.....I do have another buck if I have to go with the two bridge option. I could match the voltage of the two sides if that would help. But I don't have another 450uF cap. Does the other side need a cap and if it does, can it be smaller?  I'm thinking that the side that goes to the lamp wouldn't need it.

ROGER1 posted:

But, as a science teacher, here's a high-school science project for you.    Hook up the "raw" (no capacitor) output of the bridge rectifier to the bell.  This applies pulsed DC which I believe will operate the bell about the same as AC.  You still have your option of using 2 bridge rectifiers after the AC-powered 153IR.

So, to see if that works, connect a bridge to my AC source and then take the two DC leads and connect it to the bell?  Correct?

Correct.  NO capacitor.

Also.....I do have another buck if I have to go with the two bridge option. I could match the voltage of the two sides if that would help. But I don't have another 450uF cap. Does the other side need a cap and if it does, can it be smaller?  I'm thinking that the side that goes to the lamp wouldn't need it.

Right.  The side that goes to lamp does not need anything added other than the bridge.  The lamp would be receiving pulsed-DC which you would not see any difference in a filament bulb.  The purpose of the 2nd bridge is not so much to equalize the voltage to coil and lamp as much as to equalize the DC- voltage.  May seem like semantic gymnastics but also trying to keep it simple with a mind toward additional downstream modifications should need arise.

If lamp is too bright with this method you could use a 99 cent buck regulator...or typically simpler just to insert some 1-cent diodes which would lower the average DC voltage to the lamp in 3/4V increments.

I just saw your post as I was drawing the below diagram. First of all.....some good news. Just tested the 69 with the bridge alone (pulsed) and it works. In fact....sounds better than AC. So.........I drew up the following diagram.   I did not include a cap on the left (I don't have a 450 but I do have smaller). Should I leave it out or put the smaller one in there.  And is my diagram correct as drawn?

If it's a go on the wiring, the interesting part is getting 2 bucks under the roof (it will be tight). It's the best place to put them as I can adjust them easily. If I had them under the layout, I'd need a buddy to help with dialing it in. 

After I use this second buck, I need to replenish.  Would these be a good choice?

f:0" target="_blank">https://www.ebay.com/itm/10X-R...SwceNZXCOo:rk:6f:0

ROGER1 posted:

I just saw your post as I was drawing the below diagram. First of all.....some good news. Just tested the 69 with the bridge alone (pulsed) and it works. In fact....sounds better than AC. So.........I drew up the following diagram.   I did not include a cap on the left (I don't have a 450 but I do have smaller). Should I leave it out or put the smaller one in there.  And is my diagram correct as drawn?

In above diagram the right CAP will smooth the pulsed DC from the right bridge so the 69 will not buzz effectively.  Since it seems you have bridges lying around, you could simply add a 3rd bridge on the 153IR-NO output; the + and - outputs of the 3rd bridge would then drive the 69 (no capacitor).

It sounds like space is getting tight in the Gateman.  Likewise you don't need a buck inside the Gateman to lower the lamp voltage.  You could use one or more 1-cent diodes which obviously take up much less space; per previous post each diode lowers the lamp voltage by about 3/4 V. 

The advantage of a 3rd bridge is you can locate it outside the Gateman since it could be placed right at the 69.  So in diagram above, if pulsed-DC works better on the 69, add the 3rd bridge to the 2-wires labeled "AC to Bell".

ROGER1 posted:

After I use this second buck, I need to replenish.  Would these be a good choice?

f:0" target="_blank">https://www.ebay.com/itm/10X-R...SwceNZXCOo:rk:6f:0

Be advised direct links to the auction site could be deleted per OGR terms-of-use.  Anyway, that's a good choice.  Specifically it has a 220uF capacitor on the input.  Many of these $1 modules "only" have a 100uF.  You can simply read the numbers on the input-side capacitor.  Obviously having 220 vs. 100 directly affects how much additional external input capacitance you need to add for smoothing pulsed-DC.  Many applications of DC-in, DC-out converters present smoothed DC to the input so the size of that input capacitor is less important.  But for O-gauge applications where the DC input is likely to be pulsed-DC, why not get the 220 since they seem to be available at the same price.

Stan,

I think we're getting close here. I think I can get the second buck in there. If I can't, I've got all kinds of diodes.   But here are two basic questions that I tend to find confusing..

1. DC current runs from + to -.  I always get mixed up when connecting polarized components. Connecting the positive side of a bridge to + side of a cap or buck always seems counterintuitive.

2. If the branch out to the 69 is "upstream" from the cap, why would the output be smoothed? Wouldn't it only be smoothed from the cap onward?

Biology mind doesn't compute. LOL

Also.....not exactly sure which terminals in the top of the GM the + and - outs are connected to

1. Don't have a good answer for that.  Perhaps equally confusing is that many common "DC" components like a recharge-able battery or the smoothing capacitors used here have DC current flowing in AND out.  All I can say is it's by convention that we think of DC current flowing from a higher (more positive) voltage toward a lower voltage...sort of like gravity.   To your point, there is an argument rooted in solid science that current is just flowing electrons that go backwards to the convention; but that's a philosophical argument for another day!

2. Well, in your diagram you could have drawn the 2 wires going off to the 69 below the cap.  The performance would be the same even though the 69 is now "downstream".  That is, if I understood your question...

ROGER1 posted:

Also.....not exactly sure which terminals in the top of the GM the + and - outs are connected to

You mean for the 2nd buck for the lamp?  Your diagrams are correct for the 1st buck for the coil.  Note that for these buck modules, DC- IN and DC- OUT are the same.  The DC+ OUT of the 2nd buck (lamp) goes to where you previously hooked up ACC PWR (lamp on all the time).

I'm busy wiring up the left hand buck. My solution to get all to fit is to glue the two bucks upside down to the inside of the roof.....on either side of the peak. That way, the caps don't touch anything.

Now, about the bell. So......I should tap in for the third cap between the 153 and a bridge....not thinking it matters which one. Correct?

Lots of wires in there.....one bad connection or short......magic smoke.

This is one hell of alot of wiring and electronics to listen to a bell. LOL    Have to sign off till tonight.....we've got a party to go to.  Need a break anyway.    Thanks, Stan!

I'll get in one last question.....I've got the negatives from both bucks going to the same terminal. One is smoothed, the other is not.....will that matter?

ROGER1 posted:

..

Now, about the bell. So......I should tap in for the third cap between the 153 and a bridge....not thinking it matters which one. Correct?

I think you mean the third bridge?

ROGER1 posted:

I'll get in one last question.....I've got the negatives from both bucks going to the same terminal. One is smoothed, the other is not.....will that matter?

If I understand what you're saying, yes.  The "-" outputs of both bucks go to the same lower left GM terminal.

To an earlier point, note that the buck regulator boards include a capacitor already.  If using the type in your link, you already get a 220uF for "free" which should be enough for a lamp bulb.  The coil buck needs additional capacitance because of the high current a coil draws - more capacitance better smooths the voltage.  Whatever external cap you place before the buck is additive.  So if you already have 220uF on the board, adding the external 450uF makes for an effective cap of 670uF.

The caps in the two bucks I"m using (my last remaining stock), have 100 uF caps.  I'm going to order the replacements from that link. But they will be for future projects.

Well, I think I'm good to go with the final wiring. This afternoon, I finished wiring all the bucks, bridges etc.  Next is connecting this rat's nest all together and test it out.  We'll see.

A buck question, Stan. I was synchronizing the two of them just now and it revealed a question I've had about these things for awhile. I set them both (for now) at exactly 12.0V.  My meter was clipped to the "Out" leads. Does it make a difference that there's not "load" on those leads or is the meter enough of a load to get an accurate read of the actual output?

No load is required to set the output voltage on those buck modules.

Well......no luck. Can't get the coil to work off NO.  A zillion wires here so it's hard to keep track of what's what.  Alot of it is alligator clipped.  ON the positive side......with this wiring setup, the bell will ring when the train triggers it and the lamp is on in the building. But I can get the coil to trigger with the train. I get output voltage

Here's the weird thing. I'm getting 12V output on both bucks. It should be zero unless triggered by the train. So.....something's not wired right.  ????

I just tested the output from the AC wires from my 153. Clipped my meter leads to NO an GND and got 0. Ran a train past the sensor, and it jumped to 15V.    Then I connected NO and GND to the bridge leads from buck 1. 0V as it should be. Ran a train through and it was still 0. So.......I'm thinking that both bucks have be connected by more than the attachment of their common - outs. Am I off base here? 

GOT IT!    YESSSSSSSSSSSS!

I had to start ruling things out by going up the chain. Voltage going in.....voltage to buck.....voltage out. Tested everything piece by piece and got it to the point where I had 12V out on Buck 2 and 0V on 1. Triggering the sensor knocked the voltage on 2 to 12. So it was good to that point. Started changing up the connections from the Bucks down. Started on 1 and changed the connections. Suddenly, the sensor was triggering the door......properly and DEAD QUIET. But no lamp. Changed the connections on Buck 2 down and........Bingo!   It all works.  I'm posting below the original wiring diagram and then the new one which you can take to the bank!

One other weird thing happened. In trying to get the bulb to work, I connected the - lead and it went on and the trigger still worked. I was very happy at that point, to say the least. But then......I noticed a wire that was not connected.......+ from Buck 2.  Yet, everything worked. I took a chance and put it on 2 and suddenly.....the lamp got much brighter and the door had more snap.  I'll take it. 

If it's still working later (LOL) I'll put it all together and take a vid to post.

A bit of a glitch here that I didn't notice earlier. When the train triggers the sensor, the door snaps open, but the lamp gets dimmer. Before triggering, the meter measures across the lamp contacts, about 13.5V.  Almost 0 across the coil. When the sensor triggers, I get 12V across the coil (as it should), but the lamp reads about 8.5.  What to do about that one?

I think I understand what you're seeing but it's tedious to explain.  I suggest you go back to first wiring diagram and troubleshoot again using following steps.  OTOH if you're happy with the lamp dimming when the door opens so be it; I can rationalize why it "works" but I do not recommend this configuration.

Hi Stan,

Thanks for getting back to me. I've got some questions about the above 3 diagrams:

1. For diagram 1 and 2......are they connected to the GM or are the output wires from the bucks loose?

2. I'm looking to see if the two bucks read 12V (which is what I set them at). Of course, buck 1 would only read that voltage when triggered?

3. Is this fixable? If not, last nite I was thinking that I'd take the bulb out.....put a different socket under the roof of the inside of the GM's "room" and connect it to the AC feed upstream of both bucks (PWR and GND).  That way, the lamp would operate normally (on all the time).  

I did one of the tests and found something strange. Diagram 1 with both bucks unconnected.  I measured 12V from buck 2.  But when I took the meter leads off 2 and put them on 1, I got a measurement of 1.9V (without being triggered) and that dropped to .8 after being triggered.  So, it appears something is wrong with the #1 side.   

I think I may be on to something, Stan. I have high confidence that my wiring of all the components is correct.  All the wires that had to be connected together.......leads to GND (3), leads to NO(2) I soldered together with a single wire coming out of the connected wires. And I color coded them....black for GND, red for NO etc.       But here's what I just discovered when I opened up the rat's nest of wires that I zip tied to neaten them.  I had the GND wire and the NO wire connected to the wrong input from the layout.    Should have been NO from 153 to the "NO wires" coming from the bucks. I had NO to GND and vice versa. Swapped them and tested the buck with a train triggering the sensor.......12V              I can't say, at this point that everything will work as I have to reconnect everything else. But that will have to wait till this afternoon. 

When I watched that bulb dim yesterday, it looked like a short.     Dumb, dumb, dumb

ROGER1 posted:

Hi Stan,

Thanks for getting back to me. I've got some questions about the above 3 diagrams:

1. For diagram 1 and 2......are they connected to the GM or are the output wires from the bucks loose?

2. I'm looking to see if the two bucks read 12V (which is what I set them at). Of course, buck 1 would only read that voltage when triggered?

3. Is this fixable? If not, last nite I was thinking that I'd take the bulb out.....put a different socket under the roof of the inside of the GM's "room" and connect it to the AC feed upstream of both bucks (PWR and GND).  That way, the lamp would operate normally (on all the time).  

1. wires from bucks loose.  I think you figured this out though.

2. Right.  Buck1 (coil) should read ~0 when untriggered.  Then 12V DC when triggered.

3. The "problem" with DC operation when starting from AC Accessory Voltage is the coil and bulb share that common lower-left terminal - internally connected within the GM.  So as was proposed early in this thread, if you can isolate the two bulb wires by adding a floating socket or whatever, that would be a simple alternative.

I'm suitably impressed that you guys made this into a 75 post thread!   I thought I was the master of stretching a thread, but I think I have to pass the baton to Stan and Roger, enjoy your championship guys!

Stan,

I don't know if you saw my posts that occurred around the time you posted. I found one of the problems. I had the wiring of the components correct, but the connections to the wires coming from the 153......PWR, GND, NO.....2 of them were swapped. I fixed that and tested it. I got the 12V (from buck 1) on trigger. Just alligatored all of the wiring as it is supposed to be and it worked MUCH better.   With the wires in the wrong place, the voltage reading across 2 and 3 was just about 14V......it dropped to 5.8 on trigger and the bulb was significantly dimmed.    Now, it reads about 14 and on trigger it drops to just under 12.  Much better and much less dimming.    Interesting though.....last night I measured the voltage (AC) across that spot with the original connections on there and on trigger the bulb does dim a bit....with about a 1 V drop. And that's with the PW wiring setup.

So.....with the way it is now, can I leave it?  If I do, what's the worst case scenario?   I probably could figure out a way to add a new socket to the GM interior and wire it to AC (but isolated). But if this current wiring isn't going to hurt anything, I'm thinking I'd just leave it in it's "corrected" state.  What would you do?

I'm suitably impressed that you guys made this into a 75 post thread!   I thought I was the master of stretching a thread, but I think I have to pass the baton to Stan and Roger, enjoy your championship guys!

Thanks, John!  A dubious honor to say the least. This has been a long, interesting haul. All to quiet the silly Gateman to hear a silly prewar bell. 

You guys are indispensable for these kinds of projects

Roger

Time for photos and a short video.  The first pic shows the 2 bucks Gooped into the roof of the GM. The second shows the various bridges and associated wiring scattered around the edges of the base. And a video showing it in operation. I probably should have done one with no train and trigger the sensor by hand so as to better hear the bell and the silent GM.

Wow, is that a gateman or an explosive device?

Seriously, good job. Btw, my favorite part of that clip is the PW Seaboard Switcher.