After a review of the 74461 I am not sure what disappoints me more, the product itself, the marketing or the documentation.

I fail to see any reason to buy and use this product.

---

The Product. Cat#:
- 74461
The Product Title:
- Digital Installation Decoder / For Maerklin and Trix C-Track turnouts.
Description:
- A Digital decoder for "under roadbed" Installation and connection to, Maerklin and Trix C-Track turnout mechanism (exception: Maerklin three-way turnout .. ref 74465)

Packaging:
- Good: Image of the triangular circuit board on the front so showing what to expect,
- VeryGood: on the obverse, a highlighted section listing (functional contents)
- - 1 x 74461 Digital Installation decoder
- - 1 x Connecting Cable Maerklin
- - 1 x Connecting Cable Trix
(MyWorld people take note of how "front of house" marketing should be done)

Contents:
- In addition to the information on the the box, the internal list elaborates/includes
- - 1 74461 Digital Installation decoder
- - 1 Connecting Cable - yellow, red and brown (for Maerklin C-Track)
- - 1 Connecting Cable - blue, red and white (for Trix C-Track)
- + Installation Instructions
- + Warranty Card
- Not mentioned, but included are a couple of sticky labels, to record the address on and attach to the decoder



Image showing BoM items (minus warranty and user manual documentation)
the 4 items (decoder, cables x2 and lables) come within the sealed pink antistatic bag

The Decoder itself
- The decoder circuit board is shown with the DIP switches visible - There is actually a hole in the circuit board and the switch block is soldered on the side with all the other components.
- The Decoder under review had no marking to indicate version. The only identification was "SE110726C"
- the observable difference compared to the 74460 is the power cable is pluggable - this allows for a choice of Marklin or Trix connection.

Technical Data:
- Load at lantern output <= 100 milliamps
- Load at turnout output max. 2 amps
- Voltage for the digital power circuit max. 20 volts eff.
- External voltage max. 20 volts DC. notes! below
- Electrical Strength max. 40 volts

Comments: at no point in the product description is there an overview of the following - when references do appear (at the installation stage) it is as though you are expected to already know what you are there to achieve.
- 1) the decoder has two solder tabs that are for the connection of an associated lantern in the turnout, The new items catalogue description does make reference to this and the fact that 'soldering skills are required, this tidbit is lost in the description of the item in the online data base. Other than the <=100milliamp reference above the closest we see to information about this capability are the words"Not included in delivery scope" adjacent to it in a picture.
- 2) also not found specifically , is any descriptive reference to the purpose of the 3rd (yellow) wire in the feeder cable. At first guess it might be thought that this would be for a separate power feed to the lantern so as not to draw power from the trains on the track - in which case those without any intention of using lanterns might be tempted to cut it off to prevent the exposed connector shorting things out : DO NOT DO THIS. As it transpires this same feed (Yellow wire) also provides the power for the solenoid operation. So while the commands are delivered via the tack (Red/Brown feed, the grunt for the devices, both Turnout motor and Lantern, comes from the yellow). In 'basic' connection mode (aka 74460) the yellow would tap into the red ("B") track connection , along with the red. In "Alternative Power Supply" mode, the product specs (above) state 20 Volt DC for the external voltage - keep in mind that this is provided by exactly the same (yellow) wire as you would connect to the track Red in basic mode - so it is unclear why DC is specifically mentioned but a separate comment under "Norm use" states that the decoders 'external' (yellow) connector may only be connected to a Maerklin or Trix switch mode power pack . And it must be reinforced that when using an external power source the blown/return needs to be tied to the Brown/"0" of the track feed.

Decoder Addressing.
Again , until product availability, the New Items description were at odds with those in the product database (and common sense). The confusion over dip switches use AND programmable addressing as well as the address ranges for each remained until the product actually hit the streets. Specifically not mentioned was the DCC address range which common sense would at least align with the CS2s 2000(+) range (now seen to be 2044 as above).

In the first instance it is noted that the DIP switches block has 10 bits (this is similar to it predecessor but the meaning is different).
- position 10 is for selecting fx/MM (off/0) of DCC (on/1),
- this leaves the 9 other switches for BINARY (not Trinary) addressing = 512 addresses
- NOT FOUND ANYWHERE in the documentation is the (now) observed state, that setting address "0" in the DIP switches is what sets "the decoder for programmable address mode". by address "0 I mean switches 1-9, switch 10 must be on for DCC
- so this leaves the decoder with a hard coding address capability of 1-511
- - the various sub-ranges are a subset imposed by the controller/MM architecture (e.g 6021 = 1-256 , MM architecture = 1-320, MS2=1-320),
- In DCC mode the current limits are 1-2044

Programming.
- The prime reason for programming of this device is to set DCC addresses between 512 and 2044. (although you can also program 1-511) For this purpose CV1 becomes the lower byte address (0-255) and CV9 comes into play as the higher value address byte (x256 multiplier) with a documented range of (0-7).
- - eg CV9=0,CV1=1 = 0*256 + 1 = 1
- - eg CV9=1,CV1=9 = 1*256 + 9 = 265
- - eg CV9=7,CV1=252 = 7*256 + 252 = 1044
- NOT FOUND ANYWHERE is a specific need to ensure switch position 10 is set to 1 (DCC mode) in addition to all other switches (1-9) off.
Note: the combination of 7(CV9) and 255(CV1) could give address 2047 but the last couple of address (above the documented 1-2044) must be reserved.
Attempt to load CV9 with a greater value than 7 fails to stick. Attempting address 2045-2047 cause CV9 to show(255)

Two additional programing points need highlighting
- CV36 is exposed as the means to disable (turn off) the lantern - This is a firm setting and does not appear to be linkable to a (separate) address to add value for layout operation in any way. There is certainly no reference (other than the CV) to this feature or why/when it might realistically be used. Maybe if there was a "dimmer" capability there may be some practical use.
Feb2014 update: The Jan2014 CS2 updated provided insight to CV119(for 74461) and CV125(for 74465) as a means to vary the brightness - this is a PWM function with the range 0=off to 255=100%
- In fx(MM) operation the programming (of CV8(reset) and CV36(lantern power)) are also documented , but given the only reason to do a rest would be to set the lantern state back to on, I see little chance of use. Programming can be set from a 6021 using "traditional methods", but with some complicated preamble/postamble steps added. As the fully MM architectural address range of 1-320 can be addressed by the dip switches there is no need to resort to a programmable option. If all address switches are left off in MM mode (i.e. switches 1 to 10 all off) the default address appears to default to #1, regardless of CV1 value.
- even if the decoder is in DCC "hard mode" using the switches, the CVs can be read and while the address cannot be changed CV36 can.

Miscellaneous:
- Improvement: The DIP switches are addressable from the underside without needing to remove the circuit board and risk damaging the mounting studs
- Bad: On current (relative pricing) schedules the 74461 works out at over 3x the per/address compared with a 60830 - the previous 74460 price differential had stabilized at just over 2x
- The instruction manual includes an address chart for the switch setting of the 1 to 511 permutations. as is Maerklins want they include a column for "keyboard", with two values - presumably the keyboard(6040) unit number followed but a button number (1-16). This might not be too bad except thay imply that keyboard address 17 to 32 are valid for addressing the 256-511 range - this is news to me! It would also have been nice (confidence check) if the example picture of setting the DIP switch address and the adjacent image of recording the number on a lable refered to the same address -the image of the decoder is a lazy copy of the 74460 , showing only 2 power connection wires.

Summary:
- Price aside I have only ever seen the 74460 as having any use in a child's carpet-bahn environment, where removing of excess wiring during assemble/dis assembly is desirable. In any type of fixed layout , concealing the electronics (from future maintenance or readdressing) has scared me and something to be avoided. This feeling was exacerbated by the loss(failure), over only a couple of years, of 10 such devices from the first batches.
- The Lantern support being added to the decoder is a red herring as far as I am concerned, even if it added the "non-flickering" feature compared to wiring direct to the track. However if you were concerned about this at all you would have already been running a different power feed.
- If, as may be a modellers want, a separate power supply has been installed for the lanterns in order to vary the brightness(mood) this capabilty can not be used now as there is a need for the 20V to drive the solenoid. It is noted that the reference is a "MAX 20V DC" - no reference is made to a minimum !
- By sourcing turnout power from other than the track is a good thing but this has been achievable with k83s on a separate bus or k83 look-a-likes from other vendors for years
- Of real concern however is the inadvertent reprogramming of DCC address (and the horror of having to reprogram them by isolating each individually). While people with a "programming track" option like a CS2 "should" not be driving their main layout from it - I have seen people do this forgetting they had another loco on the track other than the one they were targeting. The MS2 however has no such "programming track" so everything goes out to the main line.
- Also of note(concern?) is that , in my trials, CV programming became questionable with continuous usage (testing numerous reds/writes - using MS2 at the time) - This may be a MS2 trackbox or a one off decoder issue - but I did note the decoder was warm to the touch (no actual turnout switching had been taking place). Turning power off , to let it cool down, did help.

If/When Maerklin post the user manual online it will probably have the following URL ...
http://medienpdb.maerklin.de/product_files/1/pdf/74461_betrieb.pdf

Edited by user 16 April 2015 02:24:24(UTC)  | Reason: Not specified

And most important thing for me is, what's the different with 74460?

BR,
Sony

I have added a draft review - appended to opening post above - with a bit more substance

Hi Peter,

Thanks for the review - very meaty and helpful

regards
Kimball

As a corollary to the 74461/5, I fully expect a rollup into a 6083x form factor device. The question is how soon - Will it be this year or will Maerklin wait another year for feedback/experience.
And will it deploy the same essential capability and configuration methods? or more ?

The Maerklin Shop has been confusing over the last couple of weeks and the status of the 60830 as "Out of production" had appeared for a short time and also flowed through to the likes of LokShop where it shown but with no price and not orderable.

And then in recent days , it has become listed on the Shop again (albeit as orange "not currently available")

This could mean that the proposed replacement wont be ready in time, so they needed another batch of 60830s, or that there is a significant enough price differential.

Time will tell

Image added to contents section.

Thanks from this excellent review!

I read this one after I bought today an 74465 3-way digital decoder. I wanted to have the decoder under the track bed and not use any external decoders. This 74465 was for some years in Märklin catalogs, but not available anyway. Now when I finally got it to my only 3-way turnout in front of the loco shed I was a bit confused when reading the manual. As I can see I was not the only one...

I can't see any use for those programmable features other than the ability to change digital address if needed for some reason (re-locating the turnout numbers vs. their locations). It was very difficult to understand those instructions. My other 16 turnouts are all equipped with 74460 manual address setting with dip switches. All have also turnout lanters which take power from underside of C-track. So all is Märklin and inside of road bed.

Thank god this 74465 decoder also has manual dip switches to set address. In my case address 17 and it then reserved the turnout number 18 in CS2 (it integrated 17 and 18 to a wider logo numbered as 17). I had previously installed many years ago turnout motors and turnout lanterns (lanters taking power from the tracks). So I wonder why in hell should I solder the turnout lantern wires to the 74465 decoder (and burn it!!) to be able to use a cv value to shut the lights off??? What??

But anyway this 74465 works really well and I am happy with it. Now I dont have to manually switch the 3-way turnout when running locos into loco shed. Before my permanent layout I used this 3-way turnout with control box, so those turnout motors have been at use on the temporary layouts before.

I think that the programmable turnout decoders are very seldom needed. I can use 74460 in all of my 16 "normal" turnouts and this 3-way turnout programmable decoder one is an exception as there is not any other options to go under C-track road bed. But when building an permanent layout it is very important to decide the "right addresses" vs. turnout locations at layout to make them easier to use. Then it is easier to start to remember which is which. I have also put an VERY small sticker to turnouts showing their number = address so it is easy to see that if needed.

Cheers,
Janne

The 2013/02 Maerklin Magazin does lip service to the 74461/74465. There may even be some self deprecating humor in the introduction statement ... "A spectacular software update for even more playing fun will certainly arouse more emotion than a new turnout decode. However .... "

The article offers a few points to ponder
- A stated recommendation to use an external power supply if the layout is controlled by a Mobile Station instead of a Central Station
- The external power source must be "DC" with a maximum of 20 Volt (As originally queried - no optimal voltage is mentioned).
- .. As to the DC reference there is no specific reference as to the polarity of the wiring. While seasoned Maerklin modelers will conclude that Brown is negative and Yellow is positive, this is not stipulated in the documentation.
- the 74465 (three-way turnout decoder) has an addition CV #34. The MM article is at pains to point out that this must be set (But you have to refer to the manual to determine exactly what to set it to ..... to "0" for the CS2 (60213/4/5) and "1" for others (including Mobile Stations).
- ... THE DEFAULT IS "1"
- ... The accompanying description reads "The point is that operationally-safe control of the three-way turnout is already supported by these central units and the motors thus don't need to be allowed for". I find this confusing at best - and dubious at worst - why, why, why ?
- ... ... What does "Operational-safe control" mean
- ... ... Why would the motors be treated differently than - say the 74461 which does not have CV #34 documented
- ... ... If there is an issue surely the MS2 can be updated to work in the same manner
- ... ... Why confuse people unnecessarily -
- ........ Maerklin (Frank Mayer) please don't tell us what to do or not to do without proper explanation of the impact/consequences - instead woo us with what benefits are to be gained (a carrot not a stick please)!

Personal note: I work with a modular group where a CS2 (generally) is used when the group assembles its modules together, however a number make use of their modules as part of an extension to their own layout or standalone, where they utilise a MS - this sort of unnecessary complexity is just not appreciated.


Additional Comment on the 74465 - not specifically reviewed above.
- The instruction manual (pg17) shows the connections to the motors "crossing" - so that the cable comming out of the left of the decoder goes to the right motor and the cable from the right connects to the left motor. A little counter-intuative to say the least!

Edited by user 25 April 2013 01:47:20(UTC)  | Reason: Not specified

I add a Corollary as it relates to the 74465 and the "Operationally-Safe Control" feature as mentioned in the Marklin Magazin MM2013/02

I Start by offering a definition for this term:
Operationally-Safe Control (In relation to 3 -Way turnouts") ,
.. is a control function that prevents the invalid aspect of both branch-left and branch-right being attempted. The Control is implemented by adding a step to always send a command of "go-straight "to the opposite point before issuing a branch to the desired directions mechanism.


The implementation of the Marklin 3-Way turnout uses 2 separate motors which, when no safeguards in place, could both be set to branch resulting in the tongues of the 2 turnout mechanisms fighting against each other. The spring loaded mechanism in most Marklin turnouts is a saving grace however this is considered undesirable and poor operation technique.

The Central Station implements this "operationally-safe control" when you configure the turnout with the 3-Way icon, thus removing the option for a Branch/Branch situations. This implementation dictates that the addresses of the 2 turnouts be consecutive with the lower address being the "left" turnout. Should the 3-Way icon set-up NOT be used, opting instead for two distinct buttons, then all bets are off and the turnout is no longer guaranteed to operate safely.

Now, returning to the 74465. It, too, has this feature imbedded within it - not mentioned in the marketing documentation or the instruction manual that comes with the item and only emerging as a point of consideration/concern if you study the CVs or happen to read the MM.
Without doubt this is a nice feature when used with the likes of a MS2. Whereas the CS "behaves safely" by always sending a "Straight Command" to the opposite motor before sending a "Branch Command" to the desired one, The decoder only needs one command and from that determines that it must also send a pulse to the other mechanism. Thus you can change from Branch left to Branch Right with just one button.

Now we come to CV #34. This is set on (1) by default and means "Operationally Safe Control" is enabled. By resetting it (0) you return the decoder to a traditional state where each point is controlled distinctly by its own address and the decoder performs no "safeing" of the obverse mechanism. This ability to disable/enable the feature is all very well but what, I have asked myself, is the purpose of it..

After a bit of banter with the Marklin Service department (I do thank them for their patience with the technical translation). I believe the reason is because, while the CS(with 3-Way set-up) and the 74465(with CV #34=1) both perform the same function, their combined use can result in an uncertain state (Here I speculate a timing issue). Personally I do have a bit of an issue with this as the technical design should be able to streamline the respective commands and make this whole need for a set-up choice transparent - However, in a mix and match world, Marklin have ascertained that there is potential for conflict and provided the solution.

Conclusion:
From a review aspect I now make comment on this function and, as is my want, it is not the individual feature that is the focus but how it is used and seen to be used.

- In the first instance, the most disturbing thing is that the turnout will automatically change from branch-left to branch-right with just one button, However the indication on the controller (MS2, 6021-Keyboard/Switchboard etc.) still indicated the left-branch solenoid as red - therefore it is unreliable and I expect operators would get into the habit of pressing two buttons anyway, just to stay in sync.
- The Second issue is one of mobility, If the turnout with a 74465 installed is used in multiple situations (carpet-bahn, modular group, .... ) where its controller may vary from CS to non-CS, I can envision it being a painful process to alter CV#34 each time (assuming one doesn't forget about it altogether)

For my part I would leave CV#34 OFF (non-default) and continue to use a two key approach on non-CS controllers. This does introduce the issue of programming reliability - but thats a different topic.
I do not see this feature adding weight to the 74465 v 6083x business case.

FWIW - update on information to hand from Jan20014 CS2 update

Lantern
I have updated main report section refering to CV36 for turning Lantern Off(0) or On(1)
Exposed in the new CV lists are
- cv119(for 74461) and
- cv125(for 74465)
as a means to vary the lantern brightness (if cv36=1<On>)- this is a PWM function with the range 0=off to 255=100%

Invert?
Also noted is a reference to cv37 with a description of "reverse"(invert?), however I was unable to apply changes to this CV on my test units

Ports
3 cvs per port are found with similar structure to those found in the 60831 port tables.
The CS2 templates do however imply port modes as 18 whereas my test units are at 16 - and would not work with 18.

This last point along with the inability to use cv37 might indicate that there are some version changes in these products since my test ones were produced.
for the record cv7 (version) for both 74461 and 74465 under test = 2

Hi,

Sorry to bump up this old topic but does anyone happen to know where can I get spare cables for the 74461 decoder? I would need the cable that has one end going to the decoder and the three wires (yellow, brown and red) going to the track. Does Märklin sell these as spare parts at all?

All mine are packed away so I can not verify however I would check with a model R/C supplier as the micro 3 pin SIL plugs & sockets they use are very similar and come with fly leads attached and you would just need to put some 2mm spade connectors on the ends. Of course the cable colours may not match.