Problem with Fleischmann decoders on a Digitrax system?

[gavino200]

I have a Digitrax system. It works fine with all decoders that I have from a range of manufacturers. But I'm having an odd and similar experience with two Fleischmann decoders. One is a Function decoder. The other is a motor decoder.

 

I got the function decoder first. It comes pre-installed in a Fleischmann Inter-City "Steuerwagen" - a driver cab at the back of the train, without an EMU. When I start the system I'm able to turn the lights on and switch directions (red/white). But after a while it just randomly changes. Sometimes, it goes off or changes directions. Sometimes it just stops responding. I'm not able to change the address using JMRI. It's still set on 03.

 

I just got the motor decoder today. It's for a Fleischmann BR101. It works at first. Forward/reverse, lights on/off. But after about half a lap it stops responding. If I power off/on it works again, but the same thing happens.

 

I've done a search for compatibility problems between Digitrax systems and Fleischmann decoders. I've found a couple of posts where people ask about Fleishc/Digi compatibility problems. One person said he heard that there can be problems. No answers confirming this. However, I was only able to find answers where people said that they must be compatible, NMRI, etc. But I didn't hear anyone saying from first hand experience that there's no problem.

 

Anyone able to make sense of this?

[kvp]

Fleischmann decoders are DCC compatible. The only problem i've found is that they have a very high analog conversion mode threshold. If the DCC voltage on your layout gets below 12V, they stop responding to DCC signals. One thing you could try is to disable analog conversion mode. For programming, just try to use the H0 voltage levels option on your Digitrax central as many of these decoders seem to be designed to work on H0 systems. On the other hand, i was able to program them using program on main mode too.

[gavino200]

Thanks kvp. I'll give this a try.

[gavino200]

You were dead right. They both work perfectly when the voltage is set for HO/OO/12V. I was able to change the address programming on main using the Digitrax hand cab. JMRI doesn't work with these decoders.

 

I haven't yet worked out what CVs to change to turn off the DC detect function. 

[gavino200]

So I'm still working on this problem. According to the Fleischmann 685403 decoder manual there are two CVs responsible for the DCC/analog/current detection function - CV12 and CV29. I've changed these according to the instructions. No response. Still the same problem. Either I'm doing my binary math wrong or there's something else I'm missing. The decoder works fine when the command center is set on HO/12V.

 

https://www.fleischmann.de/cz/product/179330-0-0-1-3-410-0-002002/products.html

 

First CV12. 1 for DC operation on. 0 for DC operation off. I set it to 0 for off. I got a "good" on the hand set. 

 

CV29 is a bit more complicated. I've never set CV29 before. Values for bits 0 to 5 are given. They are as follows:

Bit 0 = 0 for forward motion. Bit 0=1. For reverse. I chose the default value of Bit 0=0

Bit 1 = 1 for 28/128 speed. 0 for 14 speed. I chose the default value of Bit 1=1

Bit 2 = 1 for DC possible. 0 for DC capability off. I chose Bit 2= 0.

I left Bit 2, 4, and 5 at their default value of 0. They related to railcom, 3-point curve, and additional addresses respectively.

 

Now the math. My understanding. A zero for any Bit equals zero. A "1" for bits 0, 1, 2, 3, 4, 5 equal 1, 2, 4, 8, 16 and 32. 

My only "1" is for Bit 1.

Therefore (I think) my value for CV29 is 02

So I programmed CV29 to 02.

No response. 

 

I'm confused. On the other hand it works fine on HO settings. I'm curious, if I run this at the same time as a non Fleischmann train at 12V will the slightly higher voltage do any real damage to the non-Fleischmann/non German loco?

[kvp]

N scale DCC voltage is 14V, while H0 voltage is around 18V. You can't measure this with a simple multimeter as the DCC waveform is square.

 

For a japanese loco made for 12V, the decoder uses up around 0.7*2+0.3*2=2V of the 14V input so drives the motor at 12V. For 18V, it would be 16V. Ok for most Kato motors, but way too much for any Tomix ones.

 

As for the programming, your math is ok, but maybe the decoder is a slightly different type as Fleischmann makes way too many with different CVs.

[gavino200]

On 6/25/2018 at 12:20 AM, kvp said:

 

 

As for the programming, your math is ok, but maybe the decoder is a slightly different type as Fleischmann makes way too many with different CVs.

 

Thanks kvp.I'll try communicating with Fleischmann. I was hoping to avoid that, as my ability to express my meaning about this, even in English, is limited.

[gavino200]

A little update on this. The Fleischmann decoders (motor and function) both still only work in HO voltage. But there is a second anomaly. When the layout is turned on first, both decoders don't respond at all for maybe 30 seconds. Then they respond in jerks an stutters. Then eventually they work fine. Both Fleischmann decoders seem to go through the same convulsions at the same time, and both end up working fine at the same time. Weird. 

 

So I think at some stage I'll swap out both decoders. I'm giving Fleischmann my own special "World's worst decoder" award. Congratulations Fleischmann, you deserve it!! 🙂

[gavino200]

I swapped out the loco decoder with a TCS 6 pin. It's working fine now with normal n scale voltage. 

 

Swapping the function only decoder on the Steuerwagen isn't quite so simple. I tried wiring in a digitrax function only decoder, but I just get the blinking lights, short signal. There isn't a short. I'm guessing this board and a regular US style decoder aren't compatible. I'm not shocked. 

 

The lightboard is complicated and I don't want to mess with it. Instead, I'd prefer to just get a compatible function only decoder. Problem is, I don't know what to call this, in order to search and buy one. It's not a 6 pin, and "6 pin compatible function only searches get me nowhere". I'm thinking of just buying an ESU function only decoder and seeing if it works. But I'd figured I'd try knowledge before guesswork. Anyone got some?

 

Here's the Fleischmann light board and Fleischmann function-only decoder.

 

[chadbag]

Are you "programming on the main" or using a "programming track" to try and set things up with the cab car decoder?

 

On my Minitrix cab car, I had to do the programming on a "programming track" and not "programming on the main" for it to work with any semblance of sanity.  (I need to go back and work on it more, but I was able to get it to work enough to leave it alone for a while doing this).

 

[gavino200]

6 minutes ago, chadbag said:

Are you "programming on the main" or using a "programming track" to try and set things up with the cab car decoder?

 

On my Minitrix cab car, I had to do the programming on a "programming track" and not "programming on the main" for it to work with any semblance of sanity.  (I need to go back and work on it more, but I was able to get it to work enough to leave it alone for a while doing this).

 

 

No. I'm through with the Fleischmann decoders. They work fine and can be programmed, but only when run at HO voltage. I'm not trying to get this thing to "work". I just want to find a non-Fleischmann function decoder that's compatible with this light board without altering the lightboard (as that doesn't seem easy).

 

Connecting these wires (corresponding colors) to a digitrax function only decoder doesn't work.

Edited July 31, 2018 by gavino200

[chadbag]

So the cab car works, and can be programmed, when you run at H0 voltage?

 

OK, so the same as their motor one then.

[gavino200]

7 minutes ago, chadbag said:

So the cab car works, and can be programmed, when you run at H0 voltage?

 

OK, so the same as their motor one then.

6 pin motor decoders have no blue wire.

[chadbag]

4 minutes ago, gavino200 said:

6 pin motor decoders have no blue wire.

 

What I meant was that the cab function decoder seems to have the same problem as the motor decoder: needing H0 voltage to function correctly.

[gavino200]

3 minutes ago, chadbag said:

 

What I meant was that the cab function decoder seems to have the same problem as the motor decoder: needing H0 voltage to function correctly.

Yep. It’s a Fleischmann thing apparently. I switched it out today for a TCS 6 pin.

 

I have a 6 pin DH motor decoder. I don’t mind using it as a function decoder, but I don’t know what to do with the blue wire.

 

I’m also fine buying a compatible non fleisch function decoder but I don’t know what to buy.

 

 

[chadbag]

So BLUE is usually the common.  I am looking at the D&H FH05B instructions.   The 6 wires are as follows (I would assume other 6 wire function decoders are basically the same)

 

Für obige Variante 3 verbinden Sie die Drähte des Decoders oder Adapters nach folgendem Schema (s. auch Grafik auf der folgenden Seite):

roter Draht   mit dem rechten Lokschleifer

schwarzer Draht   mit dem linken Lokschleifer

weißer Draht mit dem in Fahrtrichtung vorderen Licht

gelber Draht  mit dem in Fahrtrichtung hinteren Licht

grüner Draht  Funktionsausgang AUX1

violetter Draht  Funktionsausgang AUX2

 

 

Which is basically  

 

RED -- right track

BLACK -- left track

white -- front light

yellow -- rear light

 

They give a handy graphic in the German manual   (the English manual seems to skip the FH05B)

 

We need to ask @kvp but in looking at things they seem to usually use one of the rail (red for example) to replace the missing blue since the idea is that the blue is a common voltage source for lights.   It appears in your picture that the blue is being used for power of something else as the two rail inputs would be powering the lights.  Hopefully kvp will be able to enlighten us.

 

 

[gavino200]

12 minutes ago, chadbag said:

So BLUE is usually the common.  I am looking at the D&H FH05B instructions.   The 6 wires are as follows (I would assume other 6 wire function decoders are basically the same)

 

Für obige Variante 3 verbinden Sie die Drähte des Decoders oder Adapters nach folgendem Schema (s. auch Grafik auf der folgenden Seite):

roter Draht   mit dem rechten Lokschleifer

schwarzer Draht   mit dem linken Lokschleifer

weißer Draht mit dem in Fahrtrichtung vorderen Licht

gelber Draht  mit dem in Fahrtrichtung hinteren Licht

grüner Draht  Funktionsausgang AUX1

violetter Draht  Funktionsausgang AUX2

 

 

Which is basically  

 

RED -- right track

BLACK -- left track

white -- front light

yellow -- rear light

 

They give a handy graphic in the German manual   (the English manual seems to skip the FH05B)

 

We need to ask @kvp but in looking at things they seem to usually use one of the rail (red for example) to replace the missing blue since the idea is that the blue is a common voltage source for lights.   It appears in your picture that the blue is being used for power of something else as the two rail inputs would be powering the lights.  Hopefully kvp will be able to enlighten us.

 

 

 

Thanks. I'll wire in the DH motor decoder in the morning and connect the blue wire to one of the rails. I'll let you know if it works. 

 

Edit: I'll turn on the bat signal: @kvp

Edited July 31, 2018 by gavino200

[kvp]

Bat signal detected... :-)

 

Do not connect directly to the rails! If there is no blue wire, then you'll need to add two diodes to the two rail inputs pointing towards the place of the blue wire (the lights common).  This will form the missing part of the bridge rectifier. Omitting this will result in a flickering headlight that turns on only 50% of the time or less/more depending on if sx or dcc zero stretching is in use. Also without the diodes, any stray reverse voltage on the tracks could get through the leds, burning them out.

 

It's possible to use two diodes for all lights or give each led its own pair. If you are wiring in the decoder, you can also tap the decoder's rectified high voltage pad with a blue wire if it's exposed (and documented) somewhere.

 

Wiring diagram:

https://dccwiki.com/6_Pin_DCC_Plug

 

The blue wire (when present) connects to the two diodes aready on the decoder and usually has a bit of filtering too, so eliminates the two external diodes.

 

[chadbag]

8 hours ago, kvp said:

Bat signal detected... 🙂

 

Do not connect directly to the rails! If there is no blue wire, then you'll need to add two diodes to the two rail inputs pointing towards the place of the blue wire (the lights common).  This will form the missing part of the bridge rectifier. Omitting this will result in a flickering headlight that turns on only 50% of the time or less/more depending on if sx or dcc zero stretching is in use. Also without the diodes, any stray reverse voltage on the tracks could get through the leds, burning them out.

 

It's possible to use two diodes for all lights or give each led its own pair. If you are wiring in the decoder, you can also tap the decoder's rectified high voltage pad with a blue wire if it's exposed (and documented) somewhere.

 

Wiring diagram:

https://dccwiki.com/6_Pin_DCC_Plug

 

The blue wire (when present) connects to the two diodes aready on the decoder and usually has a bit of filtering too, so eliminates the two external diodes.

 

 

I don't recognize the components but if you look at the pic that Gavin posted of the cab car, the blue wire is not going directly to the lights but appears to be going to some other circuitry,  Can you recognize it by the pic?

[gavino200]

1 hour ago, chadbag said:

 

I don't recognize the components but if you look at the pic that Gavin posted of the cab car, the blue wire is not going directly to the lights but appears to be going to some other circuitry,  Can you recognize it by the pic?

 

I also suspect that the clever Germans may have included the necessary circuitry. It looks very complicated for just a red and white LED.

 

I actually wired the thing up and it works. Finally my IC is finished!!

 

Edited July 31, 2018 by gavino200

[chadbag]

So what did you end up with blue wiring wise?  I can't see it very well.

[gavino200]

9 hours ago, kvp said:

Bat signal detected... 🙂

 

Do not connect directly to the rails! If there is no blue wire, then you'll need to add two diodes to the two rail inputs pointing towards the place of the blue wire (the lights common).  This will form the missing part of the bridge rectifier. Omitting this will result in a flickering headlight that turns on only 50% of the time or less/more depending on if sx or dcc zero stretching is in use. Also without the diodes, any stray reverse voltage on the tracks could get through the leds, burning them out.

 

It's possible to use two diodes for all lights or give each led its own pair. If you are wiring in the decoder, you can also tap the decoder's rectified high voltage pad with a blue wire if it's exposed (and documented) somewhere.

 

Wiring diagram:

https://dccwiki.com/6_Pin_DCC_Plug

 

The blue wire (when present) connects to the two diodes aready on the decoder and usually has a bit of filtering too, so eliminates the two external diodes.

 

 

Thank kv. I decided to try it directly wired. The resource you linked is very helpful. I think I'll do that next time. I actually don't even have any diodes. What kind of diodes do you recommend I buy for an application like this? I'll need to get them for my jet train project soon in any case.

[gavino200]

4 minutes ago, chadbag said:

So what did you end up with blue wiring wise?  I can't see it very well.

 

The blue wire is on the left of screen. It picks up just next to the white wire and goes to the black track connection. There's a break in the blue insulation that hadn't been liquid taped yet.

Edited July 31, 2018 by gavino200

[chadbag]

Is that a PD05A decoder?

[chadbag]

11 minutes ago, gavino200 said:

 

I also suspect that the clever Germans may have included the necessary circuitry. It looks very complicated for just a red and white LED.

 

I actually wired the thing up and it works. Finally my IC is finished!!

 

 

Interestingly, if I click this photo, it opens a new window and goes to Hobbysearch to a Tomytec temple.  https://www.1999.co.jp/eng/10340728