Question About the RK Scale Pennsy Signal Bridge 30-11030

Not in this lifetime John! Mind you, I said that about SMT too! I think the limit of my programming is as it relates to Excel spreadsheets.

Rod

@Rod Stewart posted:

...

Does that help Stan?

Yup...it helps to confuse me even more!

I can't quite make out some of the printed-wire connections and can't see the markings/values of some components.  But above is a guess at the schematic (for one side of the two).  I purposely used fec_fan's original schematic to compare/contrast.

As drawn I can sort of see how this could work with the PRR 7-LED version...but I don't understand how this circuit drives the center LED all the time.

@Rod Stewart posted:

...

The upper sketch below shows how the leds are currently wired....

...

So, I did some searching and found this OGR thread which I see I even participated in.  In it, two guys opened up what appears to be the same 7-LED signal head we're discussing.  Here is a hi-jacked cut-and-paste of Danr's photos from that thread.

Somehow I got the impression there were ONLY LEDs in the signal head?

@Rod Stewart posted:

Here for comparison is a pic of the RK Cantilever Signal Head, the common red over yellow over green setup. And a close up of the same 6 pin connector as the Pennsy bridge head earlier. Note that these connectors are not keyed, and you can plug them in in any one of five positions. MTH must be counting on us being smart enough to know which way its supposed to look! Haha.

That is, you previously posted the above R-Y-G 3-LED signal head that uses the same connector.  If this was a photo of a signal head you have, does it plug-and-play?  You'd think it would!

So now it's like going full-circle...and I'm wondering where are we going with this?!

@Pennsylover posted:

I'm not a fan of the timed yellow position.  I know this was done so that a single block could control all three aspects of the signal, but it would have been nice if a jumper could have been included to allow the yellow to be controlled by its own block.

Dale

If the base circuit board is to be modified, whether using a micro-controller or not, then it makes sense to include a jumper to disable the timed-yellow feature and allow "direct" access to the 3 states - whether the 3-LED R-Y-G style or the 7-LED PRR style.

IF the base circuit board is modified...then, IMO, that would be the time to provision for a special mode that manages the changing color of center-LED.  In other words, kick-the-can-down-the-road.

I also ran across this OGR thread...where the MTH 30-11013 7-LED signal has 6 wires (vs. 5) going from the board to the signal head itself.  As GRJ reported, the stock MTH base circuit board may have some issues so while filling the kitchen-sink, a revised board could support this version too!

@Rod Stewart posted:

Not in this lifetime John! Mind you, I said that about SMT too! I think the limit of my programming is as it relates to Excel spreadsheets.

That was your start, time to take the next big step!

@stan2004 posted:

Somehow I got the impression there were ONLY LEDs in the signal head?

That is, you previously posted the above R-Y-G 3-LED signal head that uses the same connector.  If this was a photo of a signal head you have, does it plug-and-play?  You'd think it would!

So now it's like going full-circle...and I'm wondering where are we going with this?!

Stan; thanks for posting the above pics. They are really good and show the PCB and wiring in pretty good detail. My over-simplistic earlier idea did not show the diode and resistor, nor did it show how the wire colors are inter-connected. But it did show the pairs of diodes wired in parallel which is how they are wired. From your pix above I put together this circuit which I believe is correct. I plan to verify it.

NOTE: The drawing below is incorrect. The correct version is R1.1 and appears further down.

I show the internal wire colors above, but not which pin of the connector they go to. Hope to add that detail soon.

Overall this wiring does not look much like the 11036 wiring that fec-fan did up and you attached earlier. So I doubt very much that a 3 color head from an 11009 cantilever would be plug n' play. I don't plan to try it, and the connectors are oriented differently anyway. They are horizontal on the 11009 and vertical on the 11030, but I am sure you know that. Even if the wiring is compatible, which one of the five positions would be correct? (Since these connectors are not position-keyed)

To your comment: "So now it's like going full-circle...and I'm wondering where are we going with this?!" Good question. I don't see me going much further than clarifying the head wiring and pin details, just for our collective knowledge base on these signals. And I plan to convert the second head on my signal to red-yellow green like the first. And I think I'll just remove or paint the center yellow led black, so it more closely resembles a N&W CPL head, and leave it at that.

How do you see all this? Can you deduce any conclusions from the wiring diagram above?

Rod

Let the confusion continue!  Of course my schematic of the base PCB based on your photo is an educated guess...and if we want to get to the bottom of this there's additional work.  For example, I'd ask you to gently lift the large electrolytic capacitors on one side of the two and read the markings and photograph the traces on the PCB again.

So the problem is there is no way the 7-Led schematic as you drew above can work with the base schematic as I drew (again, my schematic is incomplete).

1) I believe the outer LEDs are reversed.

But the screwiest is:

2) I believe the Green and Black wires get swapped in the connector.  So green wire at the base PCB becomes the black wire at the signal head PCB. I just can't imagine why MTH would do this but if this is true then I can see the circuit working to keep the center-yellow on all the time.

Haha; what a gong show! Welcome to trainworld, which is kind of like the Twilight Zone most of the time!

Before wasting anymore of your time, I am gonna confirm the head wiring and circuit again with one of mine, and get the connector pinouts sorted out. Further investigation of the base circuit board is difficult because I have since packed my spare one away again. ("Away" translates to deep under a big pile of boxes.) Unfortunately I forgot when I had it apart to ID the board wire colors to connector pinouts. I can't imagine though that MTH would have changed wire colors between the base PCB and the heads, but who knows? This is trainworld.

Rod

OK corrections are in order, per below R1.1. I did not realize that in the side by side pictures that you posted earlier of the board front and back Stan, the back is rotated 180 from the front view. This means that all the diode orientations were backwards in my original circuit posted earlier. Bummer. Also I added pinout connections referenced to the bottom view of the 5 pin connector, and the board wire colors. Hope this all makes sense.

You'll have to check Stan and see if your assumed board circuit makes more sense now. I tested all leds using a 5vdc supply and a 200R resistor, and they all light in agreement with this revised circuit. 

Rod

So now that you have assigned pin numbers to the 5-pin connector on the bridge, is it practical to check continuity to the base circuit board to see which color wire goes to which pin?

I'm a bit confused; I understand your "new" base PCB is packed away in a box...but isn't that same PCB in the base of the bridge?  Or is it difficult to extract and examine?  If it's difficult to get to then it will be difficult to map the 5 colored wires to the 5 pins on the bridge connector.

Also, when you say you used 5V and 200 resistor to test the LED, do you mean you tested the LED right at their terminals of the LED...or apply the 5V to different combinations of the 5-pin connector?

@stan2004 posted:

So now that you have assigned pin numbers to the 5-pin connector on the bridge, is it practical to check continuity to the base circuit board to see which color wire goes to which pin?

I'm a bit confused; I understand your "new" base PCB is packed away in a box...but isn't that same PCB in the base of the bridge?  Or is it difficult to extract and examine?  If it's difficult to get to then it will be difficult to map the 5 colored wires to the 5 pins on the bridge connector.

Also, when you say you used 5V and 200 resistor to test the LED, do you mean you tested the LED right at their terminals of the LED...or apply the 5V to different combinations of the 5-pin connector?

The problem Stan is that the bridge on the layout is all fastened down and wired up underneath, so kinda difficult to extract the PCB. The spare one is packed away again, but it would be the path of least resistance for sure. Seems to me though that it's likely the wire colors to the pins in the head should be the same as from the bridge socket to the base PCB.

My testing was using the pins on the head connector. The only one that didn't work like that was to the cathode end of D1 diode, because that tie point does not terminate at the connector. So to test the red leds I connected +ve to P4 and -ve to the cathode of D1, and it worked as expected. Testing from P4 to P5 causes both the red leds and the center yellow led to light.

Finished the 3 color mod of the second head and its back on the layout working just like the first head, so I am satisfied with the project overall.

Rod

OK.  I get it about access to the base PCB on the actual layout.  Makes sense now!

How about this though:

I realize the 3 wires (red, green, black) going into the base PCB might disappear thru a hole under the layout.  But it might also be ACCESSORY AC common or ACCESSORY AC hot?  I believe the black wire (of the 3 input wires) is the SAME as the black wire going to the bridge.  In other words you can answer my nagging question by checking for continuity between the thick black wire and pin 3 of the 5-pin connector.  Or is their continuity between the thick black wire and pin 5 of the 5-pin connector?

I think a useful outcome would be to be able to instruct fellow OGR'ers as to how to wire up the signal head to bypass the timed-yellow feature.  I've seen this request more than once on OGR for block signaling.  It's not as simple as it seems with the way the 7-LED signal head is wired...but the win would be if the signal head itself does NOT need to be modified...instead just the base PCB would be ripped out...and there would be a modestly simple circuit to directly turn on either vertical, diagonal, or horizontal...3 LEDs at a time.

Stan, I just ran a quick test, and can confirm THERE IS CONTINUITY between the thick black base wire and pin 3 (black wire in the head) of the head plug. So that should answer that nagging question hopefully. There is no continuity between the black base wire and pin 5 (green).

Rod

OK.  I guess I've got to know when to hold 'em, know when to fold 'em, know when to walk-away and know when to run!

Bottom line is you got your application solved.

I just don't see how the signal head works unless Black and Green are reversed...but again that's based on an incomplete schematic of my own doing.

As my parting shot, it's strange that even the signal head PCB suggests something odd about Black and Green.

The board has letters etched near the wire attachment pads.  R has a Red wire.  Y has a Yellow wire. W has a White wire.  BUT.  B has a Green wire.  G has a Black wire.  I don't get it! 

I don't get it either Stan. I also noticed the inconsistent color coded wires. It's the same in the head I modded yesterday. I can only conclude that there is more to the base circuit board than we think we know. There is something we are missing. I wish I hadn't been quite so anxious to get my spare bridge packed away again.

But since we have the head wiring pretty much nailed, is it now possible to design a new base circuit that allows separate timing for the yellow, like you were hoping?

Rod

@stan2004 posted:

I just don't see how the signal head works unless Black and Green are reversed...but again that's based on an incomplete schematic of my own doing.

As my parting shot, it's strange that even the signal head PCB suggests something odd about Black and Green.

The board has letters etched near the wire attachment pads.  R has a Red wire.  Y has a Yellow wire. W has a White wire.  BUT.  B has a Green wire.  G has a Black wire.  I don't get it! 

Stan; STOP THE PRESSES! You were right all along.

Today I had to dig through the same signal box for an unrelated reason, so out came the spare Pennsy bridge yet again. First thing I did was hook it up and verify that it works like the other, and it does exactly. Then I took the head apart and verified the internal wiring; sure enough, black and green are swapped compared to the B and G on the board. The other 3 wires match their marks.

Next I traced the wires from the head socket down to the PCB, and guess what? They are all the same color for color with the head wiring, except BLACK AND GREEN ARE SWAPPED!!! Pin 3 on the socket becomes pcb green wire, and Pin 5 on the socket becomes pcb black. So you nailed the problem. Why in gods name this was done is completely unknown.

Here's one theory: Suppose the bases were built by one vendor, and the heads by another. Or maybe just different crews in the same shop. Now suppose that the bases were built first, and after about 5000 of them or so were done, a routine QC check discovered that black and green had been incorrectly wired backwards at the head connector. What to do? Well going back to fix the bases would be a big exercise, because the wiring to the connectors (or at the board) is difficult to get to. But swapping 2 wires in each head, not so much. Or maybe the heads weren't even built yet? Or maybe it was easier to trash the original heads and build new ones with black and green switched. Just a theory, but the real reason must be something weird like that. We'll never know.

So armed with this information your board wiring diagram is looking pretty good!

Rod

Whew!  What a relief!  I was starting to question the student loans to get an engineering degree!

OK.  Now that 2+2=4 again, I'll ponder what makes sense for the guys that don't want the
"contrived" timed-yellow feature.

It will not be a pretty sight since, as you observed yourself, the MTH PRR/Pennsy 7-LED signal head uses inverted polarities for driving the vertical vs. diagonal aspects.  This is techno-speak gobbledygook, but is the harsh reality. 

While we are on the subject of signal head wiring, below is the internal wiring for the red/yel/green signal heads for the RK 30-11009 Cantilever Signal Bridge. Fairly simple.

Rod

I don't think there's much interest in component-level assembly but here's one approach to eliminating the timed-yellow using what I call the "Dollar Tree" AC relay (because it costs about $1 in parts).  The circuit has 4 input wires:  Acc AC, Acc common (tied to outer-rail), and 2 insulated-rail triggers.  One trigger section powers the STOP (red or horizontal), the 2nd trigger section powers the SLOW (yellow or diagonal).  If neither trigger section is active, then circuit powers PROCEED (green or vertical).  Alas, relays are a bit bulky so this would not come close to fitting in a signal bridge base.  I sketched out some discrete component alternatives (transistors, diodes, resistors, etc.) that would fit...but while inexpensive, started to add up in unique parts so I gave up after about a dozen different parts were required.

This should work for either the PRR 7-LED signal head or the 3-LED red-yellow-green signal head.

Am I missing something?  Don't real signals go from green to red when the train passes, and then after the train has proceeded a specific distance from the signal, it would then go yellow until the previous train has cleared the block?

@gunrunnerjohn posted:

Am I missing something?  Don't real signals go from green to red when the train passes, and then after the train has proceeded a specific distance from the signal, it would then go yellow until the previous train has cleared the block?

LOL.  You talk about real signals.  I prefer to live in a fantasy world where "all the women are strong, all the men are good-looking, and all the children are above average"

Yeah, I'm sure my diagram is off, backwards, upside-down or otherwise indecipherable.  The idea is/was to propose an alternative to the contrived "timed-yellow" method of controlling a 3 aspect signal.  If there is a train occupying the block directly ahead, then display RED/Horizontal.  If there is a train occupying the block two blocks ahead, then display YELLOW/Diagonal.  If there is no one in the two blocks ahead, then display GREEN/Vertical.

Well, that description is correct for prototype operation, in reading your previous post again, I see that operation would be possible with the suggested mods.  That's not to say the suggested mods are going to get done by a lot of people, but that's another issue.