Feb 19, 2007
One things missing local pd car.
Never enough when you need them, always there when you dont. LOL.
An update on what we've been up to lately. Wain, Rey and I finally completed our troubleshooting of the South Seattle Yard panels. It came to pass that our issues stemmed from a misunderstanding between developer and "builder". We initially built the boards with a negative leg common buss. That worked fine for the primary "Multipanel" board, switches 1-24. We also installed the "expansion" board switches, 25-30, the same way. Turns out adding the expansion switches to the buss created feedback "noise" causing the LEDs to rapid flash and the tortoise switch machines to continuously cycle back and forth. Big problem.
Last night we decided to take the expansion board off the buss entirely and create 6 individual negative leg returns to the board, no buss at all. Here is Rey working on the back of the panel to isolate each switch and wire in the individual returns
I was at the other end wiring the 3 cable connectors (yellow, orange, red wires) to the ribbon wire and installing them. They are marked SEG1, SEG2 and SEG3. Stands for Switch Expansion Ground. I also removed the buss assembly that fed the expansion board. You can see the open screw holes. Now it looks like this.
The result, 100% operation as intended. Eureka! Now we can install this bad boy and run the railroad. Whew. This will hopefully be the last photo of it on the workbench.
Next week we'll begin install. On other fronts, Ed and Doug completed their debugging of the Renton Yard turnouts. We had a short somewhere, but it has been resolved and except for 2 that still need some ballast cleanout, this yard will be ready to go. Scott also completed the power supply supports. He built three that will support our 3 pyramid power supplies. Basically getting them off the floor.
That's a lot in a few short words, but we're on to the next phase. Thanks for checking in.
Glad things are moving forward for you and the debugging is going away.
Whew. I hate it when headaches such as this one, occur. Can put a knot in your stomach, trying to chase down the gremlins!
It's been an arduous process. 57 email exchanges with the developer and we (RGW crew) made the determination to pull the buss. Just lucky it worked.
I can only imagine, or really, dream of the resources. Great to see the process.
Yes no kidding that is definitely a knotter.
Glad it worked out for you.
All I can say is WOW! I could never do that. My hat is off to you all! That is some amazing work.
I am glad you all were able to get it worked out.
I'll be honest guys, this was kind of a last shot for us. The night before I literally thought about boxing the entire production up and sending it back to England to let the developer figure it out. We had literally tried everything, and two of my guys are top notch Boeing engineers and they couldn't come up with any answers. I want to make it clear, the developer is not at fault here, so please do not misconstrue my comments. This was a puzzler and he gave me about 50 things to try. It's just that none of them panned out. The truly weird circumstance was/is that the main board (switches 1-24) works perfectly with a negative return buss. The expansion board (switches 25-30) wouldn't. It was getting feedback through the buss, nobody knows why or how, it just did.
The solution was kind of a, well, nothing else worked so why not this kind of thing. We put in temporary switch controls on switch 25 and disconnected everything else. The system worked perfectly. So the first thought was separate the buss into 2. That didn't solve anything. So with three of us sitting there scratching our collective heads, I just said screw it, lets add a 4th ribbon cable (prior 3 were full) and wire each switch back to itself, power to one stem, negative return to the other, just as if we were using 6 temporary controls. Voila, it works. I'll be honest, if it hadn't, it would be half way to Britain right now.
Thanks for all your support. M
Wow! That is interesting. I am not any type of engineer but the locomotive kind. That why my outfit is still in the stone age compared to yours Michael. Again I am in envy of the resources you have there. So glad for you this particular PIA is dealt with. Leaves you better prepared for the next dilemma LOL.
I certainly hope so. Thx for the kind words. M
Well, last night was a night to forget, two of us soldering for 4 hours, for nothing.
We continue to have a dimming issue with our panel LEDs. The lights are very bright on first push, very dim on second and it alternates every push.
We thought we had found a fix. We were using two negative return busses, one for switches 1-24 (Multipanel "MP") and another for switches 25-30 (expansion board "EB"). We took out the EB buss and the dimming stopped. We tested the MP board without the buss and it worked. So we elected to run wire from the MP directly to the panel without a buss.
I worked the MP side shown above. Dismantling the buss and soldering the 3 pin connectors to the ribbon cable. Rey worked the panel side.
Painstakingly taking apart the daisy chained buss wires and adding individual wires to each of the 24 switches.
Four hours of wiring later, dimming still persists.
(insert swear word here)
I need a drink and some ibuprofen.
Hmmm. Well, I certainly can't click the "Like" button, following along with such a frustration. I believe situations such as this are where old cuss words are worn out, and new ones must be invented!
Trust me I've experimented with that very thing.
When things look their bleakest, in rides the cavalry. Rey brought his son Dylan with him last night. Dylan is a techie, UW computer science major. After starting his career in pcb manufacturing, he's now a full time programmer. Rey felt he might be able to diagnose our "dimming LED" issue. We gave him free reign over the board. Here he is doing some testing on the MP/EB boards. His friend Christof from the Czech Republic is in the background.
The kid is a wiz at electronics. After testing the boards and doing some research on the pcb chips being used, he came up with a problem and a potential fix. A few Amazon orders later, we'll try and implement that fix by next week. More on that then. Fingers crossed.
Meanwhile Ed, Doug and Scott worked on turnout operation. Scott found one tortoise with an exposed edge connector awfully close to the front of the fascia so he fixed that by adding a protective piece of fascia.
That was it for last night. Hopefully good news next week. Thanks for "Czech"-ing in.
Following on our update of last week I want to thank Daniel (not Dylan, apologies to him for that mess up) for giving us his time and doing his best to diagnose our LED dimming issue. He did research on the chips running the boards and determined we needed a 47ohm resistor in the circuit to avoid taxing the chip. Later in the evening he reworked his math and decided we actually needed 91ohm resistors.
Of course the impatient yours truly had already ordered the 47s by the time he recalculated. So we will use 2 in series giving us 94ohm. We also determined the most expedient way to put them in the circuit was adding them to short male/female connectors. Here's what we started with:
Over the weekend I soldered most of them and finished up last night. Here is a pic of them in place:
So in the end did it solve our dimming issue? Nope. But Daniel assured us that if we didn't do this the boards were in peril, so here they are.
So why is the board here and not on the bench?
Simple answer, we made a gut call. You may recall the MP board, which controls 1-24 always dimmed. For some unexplained reason the Expansion board never dimmed to any noticeable measure. So we made the call to move 1-24 off the MP and onto the EB and vice versa. The result, the 24 LEDs on the EB maintained their brightness enough for us to be satisfied and amazingly enough the 6 LEDs moved to the MP dimmed, but not enough for us to get hung up about it.
So we are moving to install. The fellas spent a great deal of last evening testing tortoise machines and turnouts. We about to get back to running the entire railroad. That is the best news of all.
Thanks for checking in.
Getting to this point has been quite the adventure!
If by "adventure" you mean "ordeal", then yes it has.
We still have to address the dimming on interbay panel.
A brief update from last evening. Doug, Scott and I spent a good portion of the night running wires from the South Seattle Yard panel to each of the tortoise switch machines.
It took about an hour to get all 30 cut to length.
Scott and Doug then spent the balance of the evening stripping and tinning one end of the wire so they are ready to insert in the panel. We left the other end alone so we can make adjustments at or near the tortoise.
While they were doing that I moved the panel back into the garage to start applying more shrink tube to some of the new resistor components.
That was our evening in a nutshell.