Railstars: Gauging interest in decoders for EMUs

[mwsteele]

It's so hard to find a company that caters to every need...

 

Digitrax has good support with Hobby Stores and great technology in their items but again act like the Apple of the Model train world...

 

I'd be up for purchasing some of these if you were to get this going I'd pre-order 6 most likely.

[CaptOblivious]

Because I doubt capt wants to try and obtain every model released for cv's  

 

Says you!

 

It's so hard to find a company that caters to every need...

 

True, but I'd really like to cater to this group's needs. :D

 

Here's where things stand. I've got a prototype design completed. Features:

• size: 7mm x 11mm (thickness, not sure yet)
  
• Output: 1A continuous, 2A peak, shut-down on outputs short to frame (as opposed to "self-destruct on outputs short to frame" :D)
  
• Can control either a motor, or a standard EMU light board without modification (although it will still have to be isolated from the track, just like a motor—but you won't have to cut it up.)
  
• You will probably have to solder a resistor to the outputs to get CV readback from a cab car installation, unfortunately. Resistors that would provide the necessary 60mA load for CV readback would take up half of an entire side of a decoder this small. If I do build a "black box programmer", I'll make sure that you don't have to do that, though, I promise ;D
  
• Firmware factory upgradable (and, perhaps in the future, upgradable on the track…working on it, gonna take some time)
  
• And if the general preference is "who cares about firmware updates?", then I can shave another 2mm, making it 7mm x 9mm
  
• MSRP is looking to be about $35
  

 

I'm getting a quote on having prototype boards made up. I'll need a few weeks to write up the firmware (so much harder than designing hardware!), but after that I'll need beta testers—that means you! Who wants to help out? Compensation: a free decoder!

 

I'll put together a pre-order page in the next day or so, so you can register your interest with your wallet and help me make this happen!

[keitaro]

i don't have dcc yet so count me out 

[CaptOblivious]

keitaro, that's OK! I appreciate the input regardless!

 

Pre-order page is up:

http://railstars.com/store/#ecwid:category=2581469&mode=category&offset=0&sort=normal

 

I anticipate being able to fulfill those pre-orders in about, let's say 6 weeks to be conservative. Will hopefully be sooner than that, but I'd rather be wrong and ship sooner, than be wrong and ship later.

 

If we can make it to 50 pre-orders, this is a done deal. Very excited!

[Guest Closed Account 1]

I'll take a bunch to test.

 

Waiting to convert to DCC:

 

Kato bullets:

 

0, 100,700, 800, 800,

[Martijn Meerts]

I'd be happy to buy a set for testing, but the price + shipping + import tax would make them more expensive than silver mini's in this part of the world to switch over completely. If you end up making a motor decoder WITH function outputs, I have a good loco to see if it really does fit in even the smallest ones :)

 

 

BTW, if Digitrax is the Apple of the DCC world, why are their controllers so ugly? :)

[mwsteele]

I'll put my pre-order through next pay!

 

Is Railstars your company Capt?

 

I don't know too much about soldering decoders or anything yet so I'll be coming to you guys for help! 

[The_Ghan]

Cap'n,

 

Am I reading your post correctly? $35 for the single decoder? $105 to do a consist?  Now I'm confused.  If I use Digitrax DZ123's for cab cars and a DZ125's (with BEMF) for the motor car it is costing me $52 to do a 3 decoder install on an EMU, and that includes Transponding in the decoders.  I sometimes use a DN135 instead because it's dimensionally different to the DZ125 and DZ123.  It's also only about $15.  So, in my mind, price is something you need to consider further.

 

Cheers

 

The_Ghan

 

[whoops! Sorry about that! I cut off a bit by accident! Capt]

[i added it back in.  The_Ghan]

[Do that again and I'll report you to admin ...  ... er ... um ... I see ... The_Ghan]

[mwsteele]

Thanks!

 

Any chance that these would fit in one of the Kato Portrams?

 

PS: you edited Ghans post instead of posting your own, was kinda confusing.

[CaptOblivious]

Cap'n,

 

Am I reading your post correctly? $35 for the single decoder? $105 to do a consist?  Now I'm confused.  If I use Digitrax DZ123's for cab cars and a DZ125's (with BEMF) for the motor car it is costing me $52 to do a 3 decoder install on an EMU, and that includes Transponding in the decoders.  I sometimes use a DN135 instead because it's dimensionally different to the DZ125 and DZ123.  It's also only about $15.  So, in my mind, price is something you need to consider further.

 

The_Ghan, I hear you, and I sympathize with your concerns. Mind, though, that the DZ125 has an MSRP of $25; likewise, if I can get my dealers on board, you won't be paying the full MSRP of $35. In the end, a consist with my decoders will cost you about $84.

 

Also keep in mind that Digitrax has their decoders manufactured by the thousands. I could get my prices much lower if I could manufacture in the thousands too, as the bulk of the costs are in manufacturing. But right now that's not a possibility. I'm still looking into ways to bring the manufacturing costs down, so that price might get revised downwards (and if it does, folks who paid the earlier higher prices will be getting a partial refund with their decoders!).

 

I hear you too on the price for the Kato replacements. We'll just have to see how that works out :D

 

As for shape, I've just tried to make it as small as possible; If there is a very popular model that it doesn't fit well in, I'd certainly consider trying a different shape.

 

I'll put my pre-order through next pay!

 

Is Railstars your company Capt?

 

I don't know too much about soldering decoders or anything yet so I'll be coming to you guys for help!  

 

Yep, she's all mine! I appreciate the support; in turn, I'm happy (as are many here) to get you running with decoder installs. It's not as hard as it seems, I promise! I got started in electronics by learning how to install decoders.

[CaptOblivious]

Thanks!

 

Any chance that these would fit in one of the Kato Portrams?

 

PS: you edited Ghans post instead of posting your own, was kinda confusing.

 

Thanks for noticing that! My mistake!

[Guest Closed Account 1]

Thanks!

 

Any chance that these would fit in one of the Kato Portrams?

 

Oh yeah!

 

I Forgot about these because they are so labor intensive. Crazy install.

 

Thats a niche to conquer Don.

 

Each decoder needs to be compatible with the 3.3v motors.

[CaptOblivious]

Doing the Portram will be tricky, but an interesting challenge.

 

The_Ghan, because you push me to try harder, I've managed to redesign the decoder to shave some serious money off the costs. I can now offer the decoders at MSRP of $25, and a pre-sale 20% discounted price of only $20. How's that sound? (What's the difference? For those who know, it's the difference between using a smaller BGA chip and switching to a slightly larger VQFN chip. Also makes it easier for me to assemble prototypes.)

[The_Ghan]

Doing the Portram will be tricky, but an interesting challenge.

 

The_Ghan, because you push me to try harder, I've managed to redesign the decoder to shave some serious money off the costs. I can now offer the decoders at MSRP of $25, and a pre-sale 20% discounted price of only $20. How's that sound? (What's the difference? For those who know, it's the difference between using a smaller BGA chip and switching to a slightly larger VQFN chip. Also makes it easier for me to assemble prototypes.)

 

$20 is on the money.  A consist is $60? Or will there be a separate decoder for cab cars?  Also, does the 1A / 2A rating make it HO/OO compatible?  It would be good if it did.  Will it work with Tomix track cleaning cars?  Bugger if it does - I bought 5 x custom decoders for that job in January.

 

Another idea / question: Could the same decoder be firmwared differently so that it will power Kato and Tomix points?  If so, it might increase production volume and reduce unit cost.

 

Cheers

 

The_Ghan

[keitaro]

how bout removing the VAT and GST chip 

[The_Ghan]

I'd like to see the Cap'n develop his own feedback system to compete with Railcom and Transponding.  Perhaps it would be called COMet, or something like that.  Make cheap to license, or royalty free, or open source to gain market penetration.

 

Cheers

 

The_Ghan

[CaptOblivious]

$20 is on the money.  A consist is $60? Or will there be a separate decoder for cab cars?  Also, does the 1A / 2A rating make it HO/OO compatible?  It would be good if it did.  Will it work with Tomix track cleaning cars?  Bugger if it does - I bought 5 x custom decoders for that job in January.

 

Yes, a consist would be $60, after the pre-order or dealer's discount. Plus VAT and GST ;) Same hardware for motor cars and cab cars, just different firmware. I don't see why it wouldn't work with the Tomix Track Cleaner. I'll send you my report when mine is converted :D

 

Another idea / question: Could the same decoder be firmwared differently so that it will power Kato and Tomix points?  If so, it might increase production volume and reduce unit cost.

 

Why, yes it could. Keep these good ideas coming :D

 

I'd like to see the Cap'n develop his own feedback system to compete with Railcom and Transponding.  Perhaps it would be called COMet, or something like that.  Make cheap to license, or royalty free, or open source to gain market penetration.

 

That would be really cool…but really hard. It might be beyond my capabilities. It also might be impossible without running afoul of Lenz's patents. :( I'll have a look at the possibility, but it would probably take a couple of years, by which time I hope to have wireless decoders for N ready to ship anyway ;)

[Guest Closed Account 1]

Doing the Portram will be tricky, but an interesting challenge.

 

The_Ghan, because you push me to try harder, I've managed to redesign the decoder to shave some serious money off the costs. I can now offer the decoders at MSRP of $25, and a pre-sale 20% discounted price of only $20. How's that sound? (What's the difference?

 

Next payday.

 

Still open to test a decoder.

 

Are you designing just decoders or light boards that accept decoders? 

 

Like a retro daughter board ? 

 

Oh that's a sweet idea. Give me some credit for this one.

 

We send you our light boards and you solder on 6 pin connectors. We slip in our own nem651 decoders.

 

Naturally you will have bought a bags of boards ahead of time so turn around time is minimal.

[CaptOblivious]

Are you designing just decoders or light boards that accept decoders? 

 

Like a retro daughter board ? 

 

Just a decoder; there are too many designs of light boards to make that route economically feasible, although I might consider it for a small handful of the most popular models.

 

Don

[KenS]

Don, if I've followed this, what you're planning as the first product is a set of three decoders for $75 ($60 for preorders) that include two that work with third-party lightboards (isolate the lightboard and wire it to the decoder) to provide directional cab lighting, and one that wires to the motor to provide motor-only control with BEMF. Is that correct?

 

With that, count me in for five sets. Without BEMF it would be a no-go for me. 

 

I expect I'd buy more eventually, but I can't swing more than five sets up front.

 

I don't think firmware upgradability on the track is essential, but it would be nice-to-have.  I don't see much value to factory-firmware upgrades on a wire-in decoder, although it may be useful to you so you can always sell the newest firmware without having inventory of the older stuff to clear.

 

One suggestion, if it doesn't raise the cost or size, put an extra pair of solder pads on the motor decoder so I can connect non-controlled lighting to the track power without running additional wires to the pickups or doubling-up on one set of pads.

 

Also, consider offering the motor decoder separately. It would be useful in those Kato subway trains that take Kato's cab decoders but not motor decoders.

 

On the topic of "why buy Railstars", it would be nice to have preloaded per-train configurations, but here's a simpler idea if you have the memory for it: define a number of pre-built speed tables that provide a reasonable starting voltage, good low-speed control, and a predictable ramp to a limited top speed (so a DCC throttle setting of 50% means something close to 50kph, for example, but some models stop at 100kph at 100%, others go to 120kph.  Then, when a model comes out, people (or you) can experiment and say "set the CV to table 14 for best results on this model".  It doesn't need to be exact as long as you allow people to program a full table if they want.  And (almost) anyone can set one CV to an integer.

 

Those of us who care for precision or who want to MU powered trains will do that eventually.  But for most people it's going to be enough if the same throttle setting makes two trains run at close to the same speed over most of the range. The big problem with most decoders is that they come "blank" and you either need to run them that way (with unprototypical and erratic results) or be willing to invest a lot in learning how to customize them, which isn't everyone's cup of tea.

 

Also, consider offering a Digitrax-style "no questions asked" decoder guarantee.  There's a cost to that you'll need to bake into your pricing, but I think that has served them well in good word of mouth over the years.

 

As for a Railstars equivalent of Transponding/Railcom, please don't. What we need is a reverse communications protocol that "nobody" owns (an NMRA standard or similar, where the intellectual property is open for anyone to use) and that multiple vendors support. If someone comes up with one, by all means be one of the first to implement it, but you don't have the size to break trail on a new protocole and build the interest to get other vendors on board.

[CaptOblivious]

Don, if I've followed this, what you're planning as the first product is a set of three decoders for $75 ($60 for preorders) that include two that work with third-party lightboards (isolate the lightboard and wire it to the decoder) to provide directional cab lighting, and one that wires to the motor to provide motor-only control with BEMF. Is that correct?

 

Yes, correct.

 

With that, count me in for five sets. Without BEMF it would be a no-go for me.

 

Very nice. BEMF only requires two additional resistors, and some additional firmware. Really, there's no excuse for not offering BEMF in a decoder.

 

One suggestion, if it doesn't raise the cost or size, put an extra pair of solder pads on the motor decoder so I can connect non-controlled lighting to the track power without running additional wires to the pickups or doubling-up on one set of pads.

 

Could you say more about this? I'm not quite following what you are wanting. An extra set of pads that duplicates the red/black track leads?

 

Also, consider offering the motor decoder separately. It would be useful in those Kato subway trains that take Kato's cab decoders but not motor decoders.

 

I'm only offering them individually right now; I might add "three-packs", but what you are asking for is no problem.

 

On the topic of "why buy Railstars", it would be nice to have preloaded per-train configurations, but here's a simpler idea if you have the memory for it: define a number of pre-built speed tables that provide a reasonable starting voltage, good low-speed control, and a predictable ramp to a limited top speed (so a DCC throttle setting of 50% means something close to 50kph, for example, but some models stop at 100kph at 100%, others go to 120kph.  Then, when a model comes out, people (or you) can experiment and say "set the CV to table 14 for best results on this model".  It doesn't need to be exact as long as you allow people to program a full table if they want.  And (almost) anyone can set one CV to an integer.

 

This is an excellent idea, and I think there should be room at least for a few of these tables. Each would be only 27bytes, and I've got a fair amount of EEPROM to play with…

 

Also, consider offering a Digitrax-style "no questions asked" decoder guarantee.  There's a cost to that you'll need to bake into your pricing, but I think that has served them well in good word of mouth over the years.

 

Done. I already offer a 3-year no-questions-asked warranty on all of our products. :D

 

As for a Railstars equivalent of Transponding/Railcom, please don't. What we need is a reverse communications protocol that "nobody" owns (an NMRA standard or similar, where the intellectual property is open for anyone to use) and that multiple vendors support. If someone comes up with one, by all means be one of the first to implement it, but you don't have the size to break trail on a new protocole and build the interest to get other vendors on board.

 

The NMRA has dropped over-the-rails feedback like a hot potato, and will likely not be picking it back up again. FWIW.

[KenS]

One suggestion, if it doesn't raise the cost or size, put an extra pair of solder pads on the motor decoder so I can connect non-controlled lighting to the track power without running additional wires to the pickups or doubling-up on one set of pads.

 

Could you say more about this? I'm not quite following what you are wanting. An extra set of pads that duplicates the red/black track leads?

 

Yes, an extra set of pads (on all decoders) that duplicates the red/black track leads, nothing more.

 

Basically, if I install lighting (e.g., find a way to mount a Kato LED and diffuser set), I need to wire it to the track somehow.  I can do that with separate wires down to the wheels, or splicing into the wires from the wheels to the decoder, or by soldering to where those wires solder to the decoder.  But my level of soldering skill makes two wires to one pad a recipe for trouble. Having a separate set of track-power pads simplifies that.

[CaptOblivious]

One suggestion, if it doesn't raise the cost or size, put an extra pair of solder pads on the motor decoder so I can connect non-controlled lighting to the track power without running additional wires to the pickups or doubling-up on one set of pads.

 

Could you say more about this? I'm not quite following what you are wanting. An extra set of pads that duplicates the red/black track leads?

 

Yes, an extra set of pads (on all decoders) that duplicates the red/black track leads, nothing more.

 

Basically, if I install lighting (e.g., find a way to mount a Kato LED and diffuser set), I need to wire it to the track somehow.  I can do that with separate wires down to the wheels, or splicing into the wires from the wheels to the decoder, or by soldering to where those wires solder to the decoder.  But my level of soldering skill makes two wires to one pad a recipe for trouble. Having a separate set of track-power pads simplifies that.

 

 

Shouldn't be hard, but I'll have to see: Space is tight! But, if it helps any, the track pickups are the outermost two pads…perhaps I could simply enlargen them…

[mwsteele]

Hey Captain, any way that you could just ship regular USPS to Canada? $15 is a little steep for shipping for something that will fit in an envelope.

[CaptOblivious]

Hey Captain, any way that you could just ship regular USPS to Canada? $15 is a little steep for shipping for something that will fit in an envelope.

 

I could have sworn I'd set them up for cheapo flat-rate shipping…because, yes, you are correct sir! Let me check the settings.

 

 

 

Try it now. Should return much more reasonable results, although I've found Priority Mail to Canada isn't always that cheap!