EM13 pcb board dummy to simplify other decoder install

[Kiha66]

I'm fairly sure that the decoder shape is not copyrighted, as atlas recently copied their other decoder shapes for their own locos in recent releases after they stopped using kato as a manufacturer.  Seeing how these are for personal use and aren't being sold, I dont think they would be upset in the first place as it encourages people to buy their own products over competitors.

[chadbag]

Should be easy enough to check on a decoder and its packaging for any copyright claims.

[Ochanomizu]

22 hours ago, Kiha66 said:

I'm fairly sure that the decoder shape is not copyrighted, as atlas recently copied their other decoder shapes for their own locos in recent releases after they stopped using kato as a manufacturer.  Seeing how these are for personal use and aren't being sold, I dont think they would be upset in the first place as it encourages people to buy their own products over competitors.

 

Hello,

 

I'm not sure I'd agree.  The decoder shape is unique.  The way it works, in that it slips between the contacts is, as far as I know, also unique.  It is effectively a "Kato" product.  I don't see how it encourages people to buy their products.  In fact, it discourages the sale of Kato EM13 29-351 product.  As for it being for personal use, there is a lot of people downloading illegal music, video and software who say the same thing.

 

I just think this should be checked.

 

That said, I'm surprised Kato hasn't offered a Railcom version of the decoder, presuming the only reason the Kato one wouldn't be suitable is because of no Railcom support.

[kvp]

4 hours ago, Ochanomizu said:

That said, I'm surprised Kato hasn't offered a Railcom version of the decoder, presuming the only reason the Kato one wouldn't be suitable is because of no Railcom support.

Kato decoders are actually made by Digitrax. Digitrax has Digitrax transponding instead of Railcom and most Kato decoders could be set up for that. Imho this was a good choice as Digitrax is the most well known US maker of DCC products. Personally i like their decoders and i tend to put 6 pin Digitrax decoders into my european prototype Kato trains as they work great without any setup or fine tuning.

 

ps: Railcom uses decoders generated signals on the track, while Digitrax transponding just slightly changes the load (blinks the headlights), which is very similar how standard DCC program track acknownledgment works. The nice thing about it is that you don't need additional components for it to work (except an extra resistor on a motor only decoder) and the decoder could otherwise remain a recieve only part with no way to drive the track. This works especially well with sub $1 low pin count Microchip MCUs, that Digitrax tends to use. The downside is that only 1 bit, an ack could be sent by the loco and the current protocol doesn't allow data readback (although in theory it could, with a few dozen seconds / read back packet speed)

[Ochanomizu]

Hello kvp,

 

Yes, I am very aware of the Digitrax-Kato relationship.  It is the reason I chose to go Digitrax 10 years ago. 

 

What I mean is, I'm surprised Kato hasn't done a Railcom decoder with another supplier, unless their agreement with Digitrax prohibits it.

[kvp]

5 hours ago, Ochanomizu said:

What I mean is, I'm surprised Kato hasn't done a Railcom decoder with another supplier, unless their agreement with Digitrax prohibits it.

I think making decoders for an incompatible system, while selling a Digitrax central with Kato branding would look funny. Especially the support calls for those decoders that won't fully work with their own central and it's matching accessories. Not to mention, digitrax is one of the most common centrals in the USA, one of the main foreign markets of Kato.

 

Also adding Railcom would make that really complex, especially that you can't have multiple active Railcom decoders on the same DCC address in the same trainset without central side consisting, which won't stay with the train. Digitrax transponding actually works nicely if more than one decoder is responding in the same block to the same address. The same thing with Railcom would be a programming mess and then you might need light decoders in all cars to avoid the lights sucking up the decoder side signal before it has a chance to reach the detector. Imho the digitrax solution is much nicer from a plug and play standpoint. While Railcom has more features, it has to be set up to work and doesn't like consists based on shared DCC addresses.

[chadbag]

@Edos2300  Any update on V2 of the EM13 / D&H board?

[Edos2300]

Hi everyone,

Sorry, I've been very busy lately with my appointment to the management of my company 😊

I still took advantage of the end-of-year holidays to finalize this project, I received last week the new PCBs and this time everything is perfect. Considering the previous discussions, unless the administrators allow it, I don't make Gerber files available directly here but you can ask me by personnal message. Note that for the moment I have only finalized the PD05A version, if anyone is interested the DH05C version can be quickly available now that the exact dimensions are validated.

 

I had the PCBs manufactured at JLCPCB (I have no stake in them), the price is very interesting (10 PCBs cost me 25.00 USD including shipping to Switzerland )and the quality is good. Wherever the PCBs are ordered, it is essential to select a thickness of 0.6 mm and a electroless nickel immersion gold (ENIG or Ni/Au) surface finish.

 

 

[chadbag]

I have not tried them yet, but my boards came today.  (Should have been yesterday but there was a typo in my address and I watched the DHL truck drive right by without stopping).

 

There seems to be some sort of intro offer for new customers as I was able to get 10 of each for $39 delivered to the US by DHL "super fast service".   I ordered the PD05A ones first, and they showed up as $18 for 10, and then I added the 10x DH05 version, and they were $21 for 10.  If I made the qty of each higher the pricing changed to not reflect the special deal.  However, If I go in and try and re-order 20x of the DH05 it is giving me a price of $24.64 but the website is stuck and does not offer me any shipping options even though there is a shipping address given.  But I suspect the shipping cost would be low.

 

Anyway, here is an image and I look forward to trying these soon.

 

 

[chadbag]

I test fit one of the PD05A boards (without decoder -- no time yet to solder any) into an E7 and it fit perfectly!   I'll test fit a DH05 board in something soon.

 

Thanks again!

 

[Ochanomizu]

Hello,

 

Nice work.  I'm surprised NGDCC hasn't done the same thing.  After all, he's done custom shaped decoders for the Tomix Track Cleaning Cars.  I would have thought Kato compatible decoders would be high on his list.

[chadbag]

53 minutes ago, Ochanomizu said:

Nice work.  I'm surprised NGDCC hasn't done the same thing.  After all, he's done custom shaped decoders for the Tomix Track Cleaning Cars.  I would have thought Kato compatible decoders would be high on his list.

 

What would NGDCC bring to the table that the EM13 itself does not do?  I am not seeing a value proposition for him.  Maybe I am overlooking something.   But that might be a reason.

 

The difference with these is they use a Railcom compatible decoder so if you are not using Digitrax transponding, but rather Railcom, you have an option for KATO DCC with Railcom (for better or for worse).

[Ochanomizu]

Hello,

 

I didn't mean NGDCC needed to load the board with a decoder, just supply the board.

[gavino200]

35 minutes ago, Ochanomizu said:

Hello,

 

I didn't mean NGDCC needed to load the board with a decoder, just supply the board.

 

Yes, it is a bit surprising. He must have considered it. Perhaps he's afraid that Kato would go after him if he did. He would be directly competing with them on a big selling item. But that's pure speculation.

[Kiha66]

I suspect NGDCC would have a very hard time competing with kato.  Since DCC in japan is much less common than in the US or Europe and the EM13 is already so economically priced I cant think what better reason people would have to buy almost the same product for more.  Seems NGDCC does amazing work on the models which dont quite fit existing decoders, rather than competing with already established brands.

[chadbag]

14 hours ago, Ochanomizu said:

I didn't mean NGDCC needed to load the board with a decoder, just supply the board.

 

I don't think there is enough money in such a product in Japan to make it worth the effort.

 

You might be able to make a business case for it in Europe with the KATO Europe stuff that runs on EM13 (vs an NEM socket).   But I suspect that the ICE4 ZIMO decoder replacement for EM13 should also work in other KATO trains.

 

For the hobbyist it is a great idea.  The hobbyist is not trying to make money, but rather trying to accomplish a goal (spending money).

 

Edited January 21, 2019 by chadbag

[Ochanomizu]

Hello,

 

The fact that Railcom equipped manufacturers haven't made an EM13 compatible decoder leads me to suspect copyright is at play.

[chadbag]

1 hour ago, Ochanomizu said:

Hello,

 

The fact that Railcom equipped manufacturers haven't made an EM13 compatible decoder leads me to suspect copyright is at play.

 

Not me.  It makes me think the market is not large enough for them to make it worth their while.

[Ochanomizu]

Hello chadbag,

 

Well, you've bought pcbs for 20, for a start!  I'll bet there'd be a need for thousands.

[chadbag]

1 minute ago, Ochanomizu said:

Hello chadbag,

 

Well, you've bought pcbs for 20, for a start!  I'll bet there'd be a need for thousands.

 

Yeah.  Say there is a worldwide market for 20k of these per year (which I think is probably a lot more than there really would be).  You can probably sell them for $2-$3 gross profit per unit (which at the "sample" purchase rate, is 100-150% markup -- I don't know what the price would be to get 20K made all at once), which is $40k -- $60k gross profit a year.  For most larger businesses, that is probably not worth it.  And the estimates are probably high.  It might be interesting for a small hobbyist centered business but they may not have the reach to saturate the market.

 

I don't see a real business case for this as a commercial product given the low cost of the EM13, which will work for most people.

 

[chadbag]

Success!   I ordered 10 of each variety of board as "samples".   I have a bunch of DH05C and PD05A without leads on order from Lippe but wanted to try one out, so I took a DH05C with leads from my decoder stash and cut the leads off.  I attached it to the board, and installed in my Shiki-Shima motor car.  I tried to give it address 0001 (Long),which the programmer happily did.  But while the programmer could read all the CVs etc from it, the motor car would not move when I put it on the layout and programmed Address 1 (Long) into my throttle.  Nor would the "coontrol" mode of the programmer and programmer SW make it move.   I changed the address to 1001 and now it works just fine.  Spec-wise, address 1 Long should be ok, but vendors are allowed to fudge the lower "primary addresses" in terms of support for short and long on the same address etc based on what I read on the DCC Wiki.  Since the Shiki-Shima is train model E001, I wanted 1.  1001 will do and is not yet in use on any of my other locomotives/trains.

 

To attach the decoder to the board, I put a small amount of low temp solder paste (melts around 280 deg F)  on each contact on the board and then heated it up with my re-soldering air solder iron to melt away the carrier and flux and leave small pads of solder on the contacts.   Then I did the same thing to the contacts on the decoder.  Then I took my normal soldering iron (Hakko FX-888D) and set the temperature to around 410 deg F (for quick melt) and held the decoder against the EM13-style board, contact against contact.  Then I just touched the iron to the backside of each contact on the decoder until I saw the solder melt.  It went quickly.   I took my ohm-meter and tested all the contacts to make sure there were no shorts and also that the decoder had continuity with each of the proper arms on the board.  I then installed it in the Shiki-Shima and put it on the programmer and away we went.

 

(Originally I wanted to use a little solder on the opposite ends of the decoder and board where the SUSI outputs are to hold the board down but there was no easy way to melt the solder once applied to the board and decoder as the contacts on that end did not go through the decoder board like the normal ones)

 

 

 

[chadbag]

I did my first one of these with the PD05A.  It is a bit harder than the DH05C version in my book to solder together but both work fine.   I used a PD05A that had the ribbon cable attached so I could just cut the ribbon cable off leaving a mm or two attached to the chip so I could just solder those 4 wires (I snipped the 2 we don't use) to the board.  Easier said than done as the wire is so short that heating it up risks it coming off the decoder side, which happened to me on one of the wires.  I finally got it soldered back on.    I bought a bunch of the PD05A that have nothing pre-attached for my other ersatz EM13 boards.  At least I have the really long left over ribbon cable from the first to use.

 

I installed this board in a KATO/Lemke ET425 commuter rail of the DB (Women's Word Cup livery version) and it works just fine.  Along with 2x KATO FL12 for the front and rear lights.

 

[Melandir]

I see that another possible option for anyone that want Railcom (I'm one of them) to use the Zimo decoder set that will be released with the new ICE4 from Kato/Lemke

 

I'm waiting to see the price of the set

[chadbag]

I had some more made for the DH05C as I had used my test run of those up last year and I have a lot more trains that need EM13 style decoders.  (I still have some of the test boards that take PD05A decoders, and will use those up with my existing stock of decoders, but won't be making any more of that style as the DH05C ones are easier to work with and the decoder is more robust).

 

I made these one 0.8mm instead of 0.6mm as a test since the cost was less.  I will be testing them in the next few days to make sure the extra 0.2mm in thickness works ok.  I had a few more made than I will ever need so if someone wants some I'll have them available for 25 cents each (US).  I am not trying to make money off of them -- just pass extras on to interested folks.

 

These are the raw boards -- you need your own D&H DH05C decoder to attach to it.   My test batch from last year with the 0.6mm boards work great!  I have 10 trains working just fine with them.

 

Edited May 14, 2020 by chadbag

[chadbag]

Just as an update.  I have not put a decoder on the .8mm thick ones I just had made but I did slide one in to an EMU motor car and it fits (and there looks like enough space for the decoder to not interfere with the driveshaft.