My Experience With DIY DCC.

[Railway42]

On the left in the blue case is a DCC controller designed by Bill Cuthbert, next in the black case is Bill’s bluetooth DCC command station, an app is available for this from Bill.
Then is the red DCC++ Throttle by Train Electronics.com. The housing was designed by me which I then 3D printed. I used a single 3.7v li-ion battery as a power source and have incorporated a charger circuit.
The blue box is the DCC++ base station which goes with the throttle. The box under the throttle is the high power DCC++ booster, both designed by Train Electronics.
Next is the Massoth commercial units which I use for my railway.
I only use Massoth decoders for my Locos whether using sound or not and all are programed with the Massoth Service Tool.
All the output stages of my DIY units are run on 19volts. My first experiment was with DCC++ and I was disappointed with the test. The loco control is fine but the function control is not and did not respond to key functions.
I then made the bluetooth DCC command station and installed the app on an old phone. This system worked fine. All the sounds were correct plus I had the control of my LGB point motors, but again not to my liking. I found my touch screen too slow (may be an old phone is the problem?).
As Bill Cuthbert’s system works with NMRA compliant decoders (as stated in his write up) this seemed to fill my needs for a DIY unit.
The next to be made was his DCC controller with the output stage running on 19volts and set for 3amps on a heatsink with a fan. This will run 3 LGB locos with sound and all the functions working fine.
I needed to build a booster unit to give more power. I again looked at the one from Train Electronics.com. Their DCC++ booster seemed to fit the bill (also the only one I could find) so one was made. Again I have found it works with the loco’s motor control but when calling up the sound functions it is not correct. I have used a DCC++ controller and a DCC controller as the input and get the same effect.
I can only think that DCC++ is not able to work with NMRA compliant decoders.
As this is just my finding, please let me know if I am going wrong or if I am incorect with my assumption.

 

[PhilP]

Have you considered that if ALL the supposed NMRA Compliant systems work (or rather, don't work) with the Massoth decoders, it could just be the Massoth decoders that are not NMRA Compliant?

The problem being, that although there is a 'standard', it is then up to the manufacturer, to read, understand, and comply with said standard.

I have found anomalies trying to drive Massoth decoders from some commercial DCC systems, in the past.

 

[Railway42]

Have you considered that if ALL the supposed NMRA Compliant systems work (or rather, don't work) with the Massoth decoders, it could just be the Massoth decoders that are not NMRA Compliant?

The problem being, that although there is a 'standard', it is then up to the manufacturer, to read, understand, and comply with said standard.

I have found anomalies trying to drive Massoth decoders from some commercial DCC systems, in the past.

Thanks for the feedback you may be correct. I wanted to try to make my own DCC system and to run the system I had already (Massoth/LGB) and am very pleased to say I have succeeded. There is alot of DCC++ DIY circuits out and wanted to relay my findings. The wireless DCC++ throttle is very good and probably only costs £35 to make. The DCC++ would be the one to use if it would work with my system.

 

[Greg Elmassian]

I have a number of questions to help clarify in my mind what you learned.

I will ask them one by one if you don't mind.

In reference to this statement: "I can only think that DCC++ is not able to work with NMRA compliant decoders. "

and this one: "My first experiment was with DCC++ and I was disappointed with the test. The loco control is fine but the function control is not and did not respond to key functions. "

So, this is very strange... perhaps your DCC++ system was set for parallel functions, and your decoders were set to serial?

Can you explain and clarify the situation more?

Thanks, Greg

 

[Railway42]

I have a number of questions to help clarify in my mind what you learned.

I will ask them one by one if you don't mind.

In reference to this statement: "I can only think that DCC++ is not able to work with NMRA compliant decoders. "

and this one: "My first experiment was with DCC++ and I was disappointed with the test. The loco control is fine but the function control is not and did not respond to key functions. "

So, this is very strange... perhaps your DCC++ system was set for parallel functions, and your decoders were set to serial?

Can you explain and clarify the situation more?

Thanks, Greg

Hi Greg
On finishing the DCC++ system and running it on my LGB track I found the motor control superb but the sound system did not respond to the function keys either to switch it on or off or to call up separate sounds.
On building another unit which the designer said was DCC and NMRA compliant everything works. Good control over the motor and all the functions for the sound work as they should.
Then on building a booster unit again DCC++ when feeding the control through this no sound control is working with the function keys. This is if I feed it with the DCC++ or the working DCC system.
All my locos are either LGB (Massoth) or Massoth units all on parallel and all working with my Massoth system.

Alan

 

[Greg Elmassian]

That is very strange, the function commands are basic DCC commands.

I would suggest making a packet analyzer, I believe they can be made very cheaply.

Many people use DCC++ and have no problems with function controls, if this was a common problem, it would be all over the forums.

Another issue that makes no sense is the booster that won't pass function commands, even through a working system.

The common part is your decoders and your track. I would guess you have some noise in the system sufficient to "mess up" the signal bad enough that it interfers with decoding the commands.

Nothing else really makes sense.

Do you have a scope to do a basic check on the waveform to check for excessive ringing/distortion?

Greg