R8249 Decoder : electrical potential modification

[Gleurp]

 

Hi everyone,

I am new to DCC so be indulgent.  ;-)

I bought my fist decoder (R8249) and I need help :

When I test the decoder, I measure 21V between the white wire and the blue wire (I would have preferred 12V but it's ok for me)
BUT
there is (at the same time) 5V between the yellow wire and the blue wire. 
I would like to have 0V (and not 5V) between these two wires.   Is it possible? How?

Thanks for your help

[2e0dtoeric]

Probably silly questions in response to yours -

What power source are you using? Is it something 'from the junk box' or a proper DCC controller?

If a proper DCC controller, you should be using the AC side of your test meter, and expect to see around 16v. This voltage is constant, no matter where the 'speed controller' knob is, as there is a digital output from the controller that operate the decoder chip.

The yellow wire is a switched on/off output, as is the white, and the blue is the common.

Have a look at this circuit diagram -

/media/tinymce_upload/3b83561c90ba7b2e36ac020a744b10f4.jpg

[Chrissaf]

You're making measurements 'off load'. Off load readings in electronic circuits can be misleading. Decoder function outputs do not have selectable voltage outputs, well certainly not a basic decoder like the Hornby R8249. The 5v reading is most likely being measured on what in electronic terms is called a 'floating' output. It needs a load attached to get an accurate meter reading value. It may also be a 'constant current' output - see further below for the significance of this comment.

.

The white wire and the yellow wire are used for directional lighting. Thus when one is high, the other is low. High and low in this scenario relate to the attached lighting LEDs being either 'on' (high) or 'off' (low). Which wire is high and which wire is low is determined by the selected loco direction. The actual measured 'off load' voltages are somewhat immaterial for the reasons documented above.

.

None of the wires on a Hornby R8249 are true 0 volts. The 0 volts rail is internal to the R8249 circuit and not extended to the outside world via a wire. This is the reason why fitting a 'stay alive' capacitor to a R8249 is more difficult - it can be done, but you have to solder a 0 volt wire directly to one of the rectification bridge diodes on the R8249 PCB......there are posts on this forum showing the diode 0 volt connection point location.

.

This link will give more information on the function of the decoder wires and how to use them. Look at the diagram for a 4 function decoder.

.

http://www.brian-lambert.co.uk/DCC.html#Basic

.

As the diagram (on the clickable link above) shows no resistors being used to protect the attached LEDs, then the output to which the LEDs attach are using a 'constant current driver' circuit. This would also account for unexpected 'off load' voltage values being measured. However, it can not be assumed that all decoders include 'constant current' outputs, I would include 1,000 ohm LED protection resistors in series with the LED to be safe (belt 'n' braces).

.

Note: 'Constant Current Driver' this statement is of a general nature and not specific to the Hornby R8249. A 'constant current' output monitors the current being passed through it and maintains that current at a constant (pre-defined by the circuit designer) value. The output voltage in a 'constant current' circuit is a variable and not fixed. The load (resistance) is fixed and must be present when the circuit is being tested, the current is fixed, thus the voltage must be variable to maintain the current / voltage / resistance relationship defined by Ohm's Law. It is a law of physics.

.

If you are trying to use the white and yellow wires for something other than LED based directional lighting, then please explain what you are trying to achieve and a more tailored advice response can be given.

[Flashbang]

Hi

Measure function outputs on Blue wire which is Positive to White, Yellow Green and Purple wires with the multimeter set onto DC volts. 

I would add a load - LED and 1K0 resistor or a grain of wheat etc 12v filament lamp between Blue and the appropriate function lead to be tested, then connect meter to this circuit and turn on that function.

What I feel your measuring possible without a load connected is just leakage and disappears as soon a load is applied.

Of course I may be wrong and the actual decoder itself is defective?

 

EDIT... Snap.. Chris posted just in front of me!

[Chrissaf]

Plus Flashbang's post supports my comments about 'loads' being necessary. Note that Flashbang is the author / owner of the site I gave the URL link for.

.

PS - Flasbang's reference to a 1K0 resistor is just another way of representing a 1,000 ohm resistor. They are both the same resistive ohmic value.

[Gleurp]

Hi 2e0dtoeric,

My equipment was bought secondhand but seems in excellent condition :
- Transformer ROCO 10725 (I tested : 16.4V AC in output)
- DCC ROCO 10764 (I tested : 28.0V AC in output)
- ROCO 10792 : LokMaus 3 

My circuit is the same that your diagram.

My loco move without problem.

The LEDs (wires yellow/blue and wires white/blue) switch "ON/OFF" depending on the direction of the loco
BUT
when the LED are "OFF", they are always a little lit (because there is 5V between white and blue wires  :-( 

What is wrong?

 

[Gleurp]

Hi Chrissaf,

Thanks for your reply.

I think that my measurements are 'ON load'.
I have 2 LEDs in serie + 1Kohn resistor between white and blue wires
I have 2 LEDs in serie + 1Kohn resistor between yellow and blue wires

When one is high (I measure 21V on the circuit), the other is low (I measure 5V on the circuit).

My problem = when it's low, the LEDs are always a "little on" 

[Gleurp]

Hi Flashbang,

Thanks for your reply.

I have measured the voltage with a load : 2 LEDs and 1Kohm resistor.

I tested the voltage because when the statut was LOW, the LED were always (a little) "on"

PS : sorry for my bad english.

 

[Chrissaf]

Your English is fine, better than some posts I read. Try increasing the value of the resistor used with the LED.

.

Also, don't put the LEDs in series. Put them in parallel, each with its own series wired resistor. See my diagram below for details. The diagram represents directional lighting with both white and red lamps that change subject to loco direction. Red & White LEDs have different forward bias voltages and work better this way. The references to 'Front' & 'Rear' relate to the front and rear end of the loco.

.

/media/tinymce_upload/54ed276c7482e10cfe4ec0cef7575449.jpg

.

What I would ask however, is that you try and avoid using the white arrow in blue button. This is not a 'reply to this post' button. Just write your reply in the 'reply text box' at the bottom of the page and click the green 'Reply' button. It is not necessary to continually duplicate previous posts (particularly my long ones). If you want to indicate whose post you are replying to, just mention their user name preceded by the @ symbol at the beginning of the text i.e @NAME that way you can also condense multiple replies into a single post.

.

for example:

.

@user name A

Reply text specific for that user

.

@ user name B

Reply text specific for that user

.

@user name C

etc etc

[Gleurp]

When I increase the value of the resistor (6,8Kohm instead of 1Kohm), the LEDs are less luminous in high level but there is few change in low level : the LEDs are always "little on"    :-(

[Chrissaf]

See my later edit with the diagram.