Showcase Miniatures - redesigned signal heads for LEDs

[ryan_wilkerson]

I just received an email from Showcase announcing a new micro 3 color LED that fits inside a redesigned signal head. Sounds like great news.

"...now after much work and re-design, we have done it! Introducing our "no compromise" N Scale signals that will now accommodate a specially designed Micro Tri-Color LED right inside the signal head..."

Read more here: http://www.showcaseminiatures.net/n-scale-century-foundry-signals.html

[C855B]

Hallelujah! Glad I was dragging my feet on my version. Maybe a little pricey, but, hey, it sure beats hours of chasing down grains of salt on a microscope stage. Maybe my project provided a little bit of encouragement.

Big thumbs-up to Walter and Debbie! Thank you!

[Philip H]

all well and good . .  . .


I want hooded type D's!

[C855B]

Just ordered a bunch, so I should have something to show in a couple of weeks.

In the last exchange I had with Walter, he wasn't indicating a lot of interest in other signal heads. I'm also after the common "Type D" three-color head, which, frankly, has never been done accurately in N... and barely in HO (BLMA, with one version). Bob Knight (TrainCat) was on to something when he was working on the Pennsy PL heads. He formed the target out of thin printed circuit board, mounting 0402 LEDs directly to the board, and then it didn't matter what was behind it.

Yes, I hear you, Philip. I agree. Don't overlook there are different variations of snow (Vader) hoods, too.

Bob's inspiration is the ideal solution for Type D heads - LEDs and wiring on the target (in a couple of sizes), and then take your pick of backs and visors. I can think of at least five different backs that could be RP'ed and glued onto such a target. The trick is figuring out how to do visors - sure, they could be soldered on the PC board, but getting all three positioned accurately would be daunting.

Not that I've thought about this or anything like that. 

[sp org div]

Yay!
Im in to get correct color indications for any further installs on the layout....
Did I miss something, or is there not yet a posted LED package size?
Max current rating?
Extra leads on those ribbon cables are probably gonna come in handy.
Thanks Showcase Miniatures!

Jeff
http://espeeoregondivision.blogspot.com/

[peteski]

Max current rating?

Why is max current important?  Modern LEDs are very efficient and they light up very brightly with just couple milliamps of current.

[sp org div]

Why is max current important?  Modern LEDs are very efficient and they light up very brightly with just couple milliamps of current.

Silly me, I should be asking more important questions, like what are the color specs on them?

Not clear on your comment, but I do respect your knowledge base from previous posts there, Mr peteski.
The layout is far enough along, that just considering the overall dynamics, as I have more than a couple locations where I am considering multiple signal / fascia repeaters from the same output in parallel. 
Currently driving my signals now with near max (20ma) to get the desired effect... and we had to build buffers to get sufficient output behind the Se8 cards...  way back.

I wonder how many they purchased, as it is likely a production run that could go fast....

[peteski]

I imagine that Showcase had these LEDs custom-manufactured for them.  Those types of runs are in thousands of devices.

I don't know what size these are but I suspect these are SMD size 0404  (0.040" square).  Devices of that size are usually rated at 5-10mA of operating current. 20mA would really be pushing their limit.  Dealing with most modern LEDs where the LED chip is directly viewable, I am surprised that 20 mA would not be way too bright for an N scale signal lamp.  Maybe your signals use less efficient LEDs.

If you power the LEDs in parallel then the total current is the sum of currents going through each LED in the circuit.

[C855B]

I wonder how many they purchased, as it is likely a production run that could go fast....

And I helped. They shipped 15 to me yesterday, so I expect them in the mailbox tomorrow or Friday. Quantities should be large - confirming what Pete said, when Walter and I were corresponding over this, we more or less came to the informed conclusion that a minimum run of a custom chip like this would be in the 2000-2500 range. Chip production was "easy", the biggest problem was getting wires attached. Seeing that their supplier has evidently cracked that nut I have no reservation about the price. Soldering wires on 0404 chips is not my idea of a good time.

I would consider the example pictures as faithful renditions, especially since they've been down this road with the HO version. This is simply a smaller chip. The green is definitely signal green, at 525nm. The yellow is biased towards amber, and is probably close to 595nm. Red appears to be 630nm or so. These are commonly available dies.

Yes - these dies are bright. 5-7 mA is all you want. My proof-of-concept with RGB chips was driven at 4mA.

I hope to break away and get at least one of these in a head as soon as I get them. Beyond that, there's so much on my plate at the moment I don't know when I'm going to be able to build more signals.

[C855B]

Okey dokey... they arrived this morning. Yes, they are custom chips, but ever-so-slightly larger than 0404. Which, unfortunately, stops any testing at the moment - they do not fit in the "old" heads that I have drilled and milled for 0404s. The installation instructions discuss a new 2-piece head, so I'm sending a note to Debbie about getting new heads.

Planned test challenge - three heads on the mast. The twisted wire bundle (very nicely done!) looks like it is small enough, but if not, definitely within range of reaming the tubing. Heed the warning about carefully handling the wires - even just staging this picture I managed to put a knot in the fine-as-a-hair bundle.

[peteski]

Remember that you can eliminate 3 wires from the 3 target signal by using the mast as a common lead.  This is especially easy when using thin-wall brass tubing.

[C855B]

Yes, but I wouldn't want to unwrap the bundle. These must be #40 or 42 wires. I don't know if I'd want to put part of the circuit on the mast, anyway. These micro-wires are susceptible to being scraped at the slightest touch. One exposed conductor by itself is no issue, but exposed wire plus "hot" mast means a failure. Given that it would likely be a short, I can imagine that it would fail in a way to take other wiring out, too.

Debbie is sending new heads, as a matter of fact she just uploaded the head ordering information this morning.

[peteski]

No need to unwrap the bundle - just remove the single common wire from the bundle of 4.  But whatever works best for you...

[C855B]

FWIW, I was just able to get three four-wire bundles down the mast without reaming, so it's not going to be an issue. At least that part of it.

As I was stuffing the little hairs down the tubing, I was thinking-through your idea about separating the common on each and using the mast. The issue then becomes one of soldering the ultra-fine wires to the top of the mast given the assembly order of all the fine bits. Doable, but has perils. In the end I think I'm more comfortable with terminating all the electrical fiddly-bits separately from the mechanical considerations. We'll see after I get a couple under my belt.

[Sokramiketes]

Is it an RYG chip or just bi-color with a mixed yellow?