In the first place, you should wait until you get your CS updated. It will then contain the connection for the 6017 boosters and you will not need any interface box. I don't have a CS, but as I understand it (you CS guys correct me if I'm wrong), it will work. You must isolate the section connected to the booster from your main section. The section with the booster will not send back mfx data, so any lok placed on that track will not "report in", but a lok already under control will continue to operate.

Charles,
Without putting you off, in my usual bloated way, I would highlight a few considerations. These may or may not be applicable to your particular situation however considering the future of ones layout direction may sae some duplication/doublehandling (wasted effort)

FIRSTLY:
Please refer the "Schematische Darstellung" @ http://maerklin-info.de/...s/Systemarchitektur2.pdf

Here Marklin have "put a stake in the ground" and shown that any legacy 6015/6017 track must be seperate from any "systems" track.
The ski-lifter solution only isolates the centre rail and then only in a single pickup train. This is the 1st hurdle as it provides a potential solution for when different track sections are driven from different controllers and leverages the tradition common ground for the rails.
1) a previous MM/MNews(2006?) article effectiely implied that, in a 6021 environment, there was no need for the ski-jump as long as.
a)) the booster(s) were in-phase digitally - i.e. all off the same 6021
b)) the ski did not "dwell" across the join.
most operators have known/assumed/gotawaywith this for years.
2) however with the CS , the output to the track must ALSO isolate the ground/return rail from the common ground concept.

Thus a CS/6017x solution can (may not be supported) only work if
1) the ski-jump isolates the the power (reds)
2) the transformer that drives the CS is totally sepearte from any transformer that drives a 6015/17 and does NOT have a ground(brown) in common.

The schematic indicates that the 60172 (unreleased / dropped?) will/was meant to be able to operate on a par with the CS "power to the track" .This is in addition to its ability to support bidirectional mfx.
Note: If the 60172 is wired as shown it would not need to relay mfx data back to the CS from a decoder as the track circuit is tied as one.

SECONDLY:
I reemphasise that any legacy use of a power circuit and the "common ground" concept as it relates to Accessory return (16V AC from the same transformer) or s88 contact (using the common rail as return) must be reviewed and preferably elliminated.

THIRDLY:
Not widely discussed is the subject of problem determination, and trying to identify in which circuit a short occured and its impact on the WHOLE layout.
Unitl the CS-upgrade for booster hits the streets I have no idea if the CS will indicate/isolate that section - the technology is available but the desire may not be .
For the 60129 (of similar DiltaControl as booster) implementation I see an IMMEDIATE benefit in that ONLY the relevent section cuts out. I am forever hopeful that someday the CS will be able to detect and report this information, in the meantime we still have the indicator LED on the booster. (and a momentary switch in the feed to allow a selectie reset)

FOURTHLY:
While the rumoured native attach of boosters to the CS is still pending, the information to hand is that it will use the same 5pin flat cable . This has 2 drawbacks. distance and connector type.
While the CS System-Bus does not take a leaf out of the loconet book and use industry standard RJ12/RJ45 connectors it does offer a more structured way of extending the control circuits up to 100m (As above, I am hopeful that the future will allow for feedback/status information)

My personal summary.
- eliminate any design that connects accessorys to track
- plan for independant sensor(s88) return bus (I use contact tracks extensively so this is a problem)
- plan for a 60129 mentality (one booster per 60129), even if the direct attach ends up able to deliver the same functionality.
-

Don't believe every post in this forum. Do you think Clapcott is right or do you think I am? Go on, guess!

Or take a course in electronics...

Hi perz Please dont play mind games with me your timing is very bad at the moment I have just been told that do to a wreaked spine I will never work again and on top of that my mum was taken into hospitial last night and I had a telephone call at three am to say she has only 48 hours to live. PS wiring is not my thing[V][V][V]

Hi Charles,

Sorry, it is not at all my intention to play games with you or confuse you. The drawing I posted is to the best of my knowledge. Most of it coincides with what Clapcott is writing, but there are some more or less important divergencies.

My somewhat sarcastic comment in the second post obviously didn't get the right effect. I could have tried to make a lengthy explanation about why I think my version is the correct one, but you would need to know something about electronics to check the sanity of my arguments anyhow.

So I was hoping to get "majority support". It may seem a stupid idea to "vote" about facts, but in a serious forum like this it often works.

But nobody "voted" and the result was only more confusion. Sorry for that.

Charles..sorry about your situation, but I believe Perz is correct on the ground issue. You realize that as soon as Marklin puts the modification in your CS, a common ground is going out on both the booster bus and the s88 bus directly form the CS.

Charles Sorry about your new here also. May all things turn out Ok.
Nev

Hi Charles,

Very sad to hear about your situation.I wish you strenght.

Kind regards
Guus

Charles, I'm very sorry to hear about your situation.

I lost my Dad last year, and my Mum was diagnosed with cancer earlier this year. She was very ill for a while after a major operation, but is now showing signs of recovery.

My prayers are with you.

Ray

Out of curiosity, and since I am actually considering a CS or ECoS whenever the computer protocol is released, I sent an e-mail to Marklin last night describing my system (seven booster zones) and asked if I could just attach the CS in place of my current controller. They said it would work. No mention of having to rip up my layout to isolate rails. (I actually got a response in under 6 hours)

Luis,
that's because the Delta shares common ground with the OUTPUT of the CS. All other boosters have a connection with the central in different way; booster bus or system bus.

As a matter of fact, on my current layout (in building phase), I feed all my tracks by Delta controls! The IB output is for the Deltas, and for accessories, including one power consumer (Busch realistic sound).


/Lars

Agree with everything you say, Lars, but there's one detail: the connector cable that comes out of the booster bus from, say a 6021 or a IB, also has ground and signal cables, added by the short detection and booster on/off control cables. The format is different, but it actually also has common ground with the 6021/IB.

Luis

Not with a CS, but with a 6021, yes. And it's not good. There are timing differences between the 6021 signal and the Delta controller, which results that, when a slider crosses over the sections, some loks just start running on their own, like if they were possessed by evil [}:)].

I would expect that the same would happen with the CS+Delta.

Luis

Good luck, Charles - all the best for you and your Mom ...

Fully agree, Clapcot. All the time from the first rumours about /systems, the information about the system has been - well, poor. Mainly absent, but also wrong. Sometimes replaced, but never corrected. I tend to prefer the german way of doing it before the american - the germans speek less but do more IMHO, but in this case it has been to much also for me.

The adapter is more than a rumour IMHO. Pictures have been published. But who knows, things could change again, you might be rigth. But the I find it hard to understand how a "Connect 6021" should be useful. In the description it was said to only include turnout control, abviously because the 6036 etc only have 14 speed steps and 80 adresses and 1+4 funcntions. And the sole purpose with it was to get M Digital users to go for the CS, it was not really intended to be useful. The "sniffer" now provided is smarter, and yes, contrary to the "Connect 6021" could take a second CS as input too. But that's a waste of course.

/Lars