1. Home
  2. Model railroading in the UK
  3. DCC saga continues.

DCC saga continues.

Having decided to return to the steam era (having sold off 45 locos some 18 months ago) I settled on the bachmann 57XX as a good representative loco. I had also read of the problems some folk have had converting the split chassis types to DCC. I thought "why not?" so into action.

All went well until two of my straight from the packet decoders refused to be read or addressed in any way. The 3rd went off tickity tock and all was well until after a night of repose it decided to strike.

It was soon evident that this chip was also deaf but also that the loco would no longer run on DC either. The fault was a sticky brush spring and quickly remedied. Whilst the decoder was seperated from the loco I put it to direct test using the Black & Red wires. The result was an open circuit. This threw a little as it had just tested out on another loco without problems. Evidently the little beggers need a motor load before they will talk to you! I shall have to try a 35 ohm across the grey/orange pair. Do all the programmers have this requirement?

The problem of D.O.A. decoders is mounting --- this now makes 5 out of 30 purchased and covers 3 makers. I note that the Bachmann 36-553 is still out of circulation.

One of the latest arrivals is a Hornby Cl 66 of very basic scantlings. I got it cheap because despite it's obvious new condition it was wounded having clearly fallen on it's nose from a height - it even cracked the motor mounting casing. This was all taken care of with the super glue tube but I did notice that whereas it had good traction in general it did tend to slip badly if the driving bogie was asked to tilt just a little. Is this normal for this model or is the bogie restricted in the tilt along plane by something yet to be discovered?

Reply to
Sailor
Loading thread data ...

I have decided never to try any more split chassis locos. Not only are they all old and lacking detail, they are a bugger to get the chip mounted safely. And most of them that I have seen have cruddy motors too. Guy

Reply to
Just zis Guy, you know?

I assume this is because the decoder wants to talk back to the programmer by pulsing the motor and thus creating a pulsed signal containing its message as a back emf that the programmer can detect. DCC decoders are nolonger perfect little children "seen but not heard" who don't talk back to you, By having no motor to pulse it bceomes a sulky teenager :-)

Alan

Reply to
Alan Dawes

Some confused terminology and umderstanding there.

Properly designed decoders have never been "seen but not heard" on the programming track. They don't "talk back" to the programmer but simply answer yes or no by pulsing the motor. It's nothing to do with back emf. It's simply the extra current drawn by the motor that is detected by the programming hardware. All that is called for is an extra 60mA for 5ms.

Assuming the programming track, wheels, etc are clean then it's posssible to write to a decoder with no motor or other load connected, you just can't check the results.

MBQ

Reply to
manatbandq

Some confused terminology and umderstanding there.

Properly designed decoders have never been "seen but not heard" on the programming track. They don't "talk back" to the programmer but simply answer yes or no by pulsing the motor. It's nothing to do with back emf. It's simply the extra current drawn by the motor that is detected by the programming hardware. All that is called for is an extra 60mA for 5ms.

Assuming the programming track, wheels, etc are clean then it's posssible to write to a decoder with no motor or other load connected, you just can't check the results.

MBQ =========================== So to read an address does programmer send numbers till get a yes ?

Cheers, Simon

Reply to
simon

In paged mode, yes. If the CV is set to 255 this is rather unfortunate. In direct bit mode it queries each bit of a byte which only requires 8 tests and is much faster. All modern decoders should support the latter but there are still some differences of implementation that can make life difficult.

MBQ

Reply to
manatbandq

So how does one "read" a CV value? ESP perhaps!

PA

Reply to
Sailor

So how does one "read" a CV value? ESP perhaps!

PA

See what the loco does in terms of the setting.

But it might be better if we knew what you want to do and with what ?

Cheers, Simon

Reply to
simon

As I said, the decoder can answer yes or no. You just have to ask it the right questions :-)

MBQ

Reply to
manatbandq

With an appropriate command station or PC/Mac connected to the DCC system decoders can be read by these systems asking the decoder for CV X what each bit is set to. If the bit is set the decoder turns the motor and the command station interprets this as the bit is set. So for each CV this done 8 times, 8 bits in a byte. For older decoders this is done up to 255 times. Note not all command stations can read decoders back.

Reply to
Chris

That makes good sense. With the exception of the functions commands the rest of the cvs all relate to motor activity. Equally the absence of ac on the programming track (Hornby Elite) infers that the sole power applied to the chip is during read or write activities so using the motor as a load would modify the line volts.

Reply to
Sailor

,

Using the motor as load alters the *current*. It's that which is detected to determine if the decoder answered yes or no.

MBQ

Reply to
manatbandq

PolyTech Forum website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.