You have to look at online ordering solutions for small stuff like that. Some useful small connectors are available from various hobby suppliers.
And for sensors liek this, where you have a lot of wire runs, telephone/network cable comes in handy for reducing the wire mess. It's a bit larger than #30, so easier to handle, and you get 4 pairs in each cable - so you could effectively have one cable connecting 4 sensors, instead of 8 individual wires that you need to bundle up.
--Randy
Modeling the Reading Railroad in the 1950's
Visit my web site at www.readingeastpenn.com for construction updates, DCC Info, and more.
What I did when using crimp terminals on small bell wire was to take a short 'blind' length of stranded 18-gauge lamp cord (which used to be easily available in hardware stores) and strip/retwist ready for crimping, then insert the stripped end together with the bell wire into the connector. This crimps all the copper together for massive contact area inside the terminal base even if crimping is a bit 'impromptu' or the wire end isn't tinned.
Then you can make up a strain relief with the usual kind of materials at the 'back' of the terminal -- this doubles as a handle to help separate them. By joining the little piece of 18ga to the bell wire this greatly aids against kinking where small-gauge solid copper goes into a crimp (or any other) type of terminal.
Any other kinds of stripped wire ends will likely work as well. It looks a bit strange but I assure you it works.
Mentioning modular plugs and such:
A lot of the connections I have on the layout, especially where the wiring goes past a layout section, have solderless slide connectors on them. Where I have the problem of using this idea with the wiring on the crossing sensors is that the wire is 30 AWG. I don't know of any wireless connectors that go that small. All the rest of the wire is about 16 AWG. (Wiring purchased from my local NAPA as no other place in my little town has wire that small.)
Indeed. Solder at the bench, but solder a length of wire to each sensor and put a plug on it. And solder wires to the points on the crossing controller at the bench, and put matching sockets on it. Or you can alwyas use screw terminal strips - solder wires, no plugs or sockets, at the bench, under the layoutm insert the bare wire end into the screw terminals.
Either way this will be a MUCH more reliable connection than any of the wire glues or just wrapping the wire. It should be fairly straightforward - do the sensors first, as while they can be damaged by holding the iron on too long, it's a REALLY long time - it's a pretty easy solder joint with limited risk to damaging anything via poor technique. The circuit board side needs you to solder fairly quickly - it's not liek if you take 2 seconds it will damage the board because you had to keep it under 1 second, but you can't hold the iron on for a minute at a time. Using screw terminal strips, you will only need to solder one pair of wires to the board - the parallel wiring of the sensors occurs at the screw terminals.
I will say, in many years of this, I solder most everything, including under the layout, and I've never dropped solder on myself. You only get a big blob that falls on you if you feed in far too much solder.
Two tricks to best success - use 63/37 solder, not 60/40, and make sure your soldering iron tip is shiny clean. When it gets blackened, it slows heat transfer which means you end up holding it on far longer than you should to get the solder to melt. When clean and shiny, it's nearly instant with small thin wires or components, a little longer with thicker wire. DO practice a bit if you are unsure. There are numerous videos to watch. Some of the things I see are a little "how ya doing" as our fruiends Down Under would say, but they still get the job done. Strive for a neat clean joint, but it doesn't have to be absolutely perfect to work. An average solder joint will be better than wire glue.
FRRYKidThe main reason I don't solder anything is that I'm not very good at it and I don't want to solder over my head.
Now, it does have to be said that I have never used anything but downhand position for soldering and would be very, very circumspect trying to do it 'over my head' But that is precisely why the idea of modular plugs and wire harnesses was invented ... and why learning how to sketch wire layout can be important. You do all the soldering and subsequent shrink-tube and paint-on insulation in the harnesses, at the bench, including any pigtailed plugs from things like the physical crossing lights that will go under the layout, and you color-code and/or tag all the wires and plugs so you can tell 'what's what' under the layout. All that happens subsequently in installation, maintenance, and troubleshooting involves easily-identified, easily-accessible plugs and touchpoints that can be field-separated and reinstalled -- ideally without tools.
rrinker I still suggest testing each of those 12 sensosrs one at a time, if they each work individually, then none are damaged (and it's not likely they would be damaged anyway, they're not particularly sensitive components). But if it turns out one doesn;t work, or requires a vastly different adjustment to work compared to the other 11 - it could be one of the wrong value that got into your oder by accident. They do have ratings, generally a dark resistence and a resistence at a certain light level, and yes it will matter because the rest of the circuit is designed around those being some particular range of values, and you can get the sensors in various values. Mixing just one of the wrong range in a bunch of parallel connections can mess up the operation. SO test them one at a time, if all else is the same except for the sensor, you shouldn;t have to adjust the potentiometer each time - set it for one sensor to work, then swap out the sensor keeping the light source, distange, angle, etc all the same. Any sensor witht he same value should work at the same setting or just a tiny tweak. If one requires you to move the potentiometer more than a fraction, something's up with that one. --Randy
I still suggest testing each of those 12 sensosrs one at a time, if they each work individually, then none are damaged (and it's not likely they would be damaged anyway, they're not particularly sensitive components). But if it turns out one doesn;t work, or requires a vastly different adjustment to work compared to the other 11 - it could be one of the wrong value that got into your oder by accident. They do have ratings, generally a dark resistence and a resistence at a certain light level, and yes it will matter because the rest of the circuit is designed around those being some particular range of values, and you can get the sensors in various values. Mixing just one of the wrong range in a bunch of parallel connections can mess up the operation.
SO test them one at a time, if all else is the same except for the sensor, you shouldn;t have to adjust the potentiometer each time - set it for one sensor to work, then swap out the sensor keeping the light source, distange, angle, etc all the same. Any sensor witht he same value should work at the same setting or just a tiny tweak. If one requires you to move the potentiometer more than a fraction, something's up with that one.
Pending Mom's plans for Tuesday or even late afternoon or evening Monday, that is the plan.
The main reason I don't solder anything is that I'm not very good at it and I don't want to solder over my head. (The whole system is under the layout which is roughly 36" high.) I have used wire glue at points for some things, however.
Yeah don't worry about it, just a couple of old timers remembering past options. Twisting the wire around is fine to test, but once you get it working you're goign to want to solder those connections for reliability.
Overmod invest in a good wireweap tool and learn to use it well
invest in a good wireweap tool and learn to use it well
???? Never heard of that.
Have to find some of those old clip posts with wire wrap pins, wire wrapping around a round wire lead doesn't do much for it.
Randy, if he doesn't like to solder do you think we could get him to invest in a good wireweap tool and learn to use it well?
That could be part of it. Twisting wires around the legs on the light sensor is not going to make a consistently good connection, heat shrink or no heat shrink. They tend to not fail, and if you never soldered them to thw wire, even less likely that you have a bad sensor. But still, you can try them one at a time and verify each one works.
rrinker It's easy enough to test each sensor - just hook up one at a time and see if it triggers when the sensor is blocked. There's basically nothing to them that could break, unless you smashed it with a hammer, or held the soldering iron on so long it melted. And they aren't nearly as sensitive to that as say an LED, an LED will be much more easily fried by overheating with the soldering iron than the sensors. --Randy
It's easy enough to test each sensor - just hook up one at a time and see if it triggers when the sensor is blocked. There's basically nothing to them that could break, unless you smashed it with a hammer, or held the soldering iron on so long it melted. And they aren't nearly as sensitive to that as say an LED, an LED will be much more easily fried by overheating with the soldering iron than the sensors.
None of the connections are soldered. They are very tightly twisted and the connections on the sensors are also sealed with heat shrink tubing.
If you look at the prior post you referenced you'll see my long dissertation on how photocells generally work. Frankly, I have not encountered a photocell which has a higher resistance when light hits it and a lower resistance when light is blocked from it!
Here's an application note from Circuitron (http://www.circuitron.com/index_files/AN/an-0003.pdf) which covers multiple sensors going to a single detection input. You can see on the righthand side of the drawing that the two sensors are connected in series. That is consistent with the way our products work too.
Easiest way to start on the 'pot' question is likely to be taking a good high-resolution shot of the Circuitron board showing the current pot positions clearly, and people with experience will take it from there.
Randy and some others can tell you how to test these sensors -- I did not think it was hard, just dark resistance with a multimeter followed by readings in various degrees of light. If these are consistent across the batch you probably don't need a spec sheet from a source like Randy or Ed to compare 'factory' values.
You might want to ask a little more about how the Circuitron actually works to make better sense out of troubleshooting this configuration.
I not sure how exactly to describe how I have the pots set. I also don't have a real easy way to sketch what I've got but I will try to describe the way I have everything wired.
For the common (SD on the diagram), each run has the wiring starting on D running to A (L to R) on each track. The wires from the sensors drop down vertically to connect to a wire that runs horizontally. (They don't connect directly to each other.) Then on the right hand end, those runs are then connected with a single wire intersecting the three runs to connect the circuit to the board.
For each of the control circuits (D to A running again L to R), the other wire on the sensor again is dropped and connected with a wire running from back to front for each letter sensor (D3, D2, D1, etc.) and to the respective stub on the board. Again no sensor is directly connected to another.
I was told that for this particular use I couldn't connect the sensors in series as they wouldn't detect properly. (Prior post http://cs.trains.com/mrr/f/744/t/281908.aspx)
The only thought that I had with the sensors that don't sense at all (D3, A1, A2, and A3) is that something is broken in the sensor itself.
I hope this helps clear things up. If someone can suggest how I might describe the pots, I can post the locations of those as well.
What we need, quickly, is a sketch of precisely how he's wired the sensors together, and where he has the pots in the Circuitron set.
FRRYKid Got yet another one for my Forum friends: Is there an easy way to add further adjustment to the sensitivity of Circuitron crossing circuits? I am in the process of install a 3 track crossing on my layout. Two of the four circuits are too sensitive in that a shadow from 3 inches above the senors trip it. One another circuit two sensors trip but the third one doesn't trigger. And none of the sensors on the fourth trigger at all. (The 12 sensors are wired in parallel as needed.) As usual, any suggestions that can be provided would be most welcomed.
Got yet another one for my Forum friends: Is there an easy way to add further adjustment to the sensitivity of Circuitron crossing circuits? I am in the process of install a 3 track crossing on my layout. Two of the four circuits are too sensitive in that a shadow from 3 inches above the senors trip it. One another circuit two sensors trip but the third one doesn't trigger. And none of the sensors on the fourth trigger at all. (The 12 sensors are wired in parallel as needed.)
As usual, any suggestions that can be provided would be most welcomed.
On a couple of the circuits the adjustments are very close to the circuit being activated. I was wondering if a resistor could be put in the "legs" of the circuits to compensate a bit.
In the alternative, what system would work for this application?
Is that with the sensitivity maxxed out? Do they even support having 12 sensors? The combined parallel load of 12 sensors may exceed the limits of the circuit. It's nice to want to save money when there are multiple tracks, but the GCP isn't really designed for that.