Hello Ivan

I am using the same set up.

My layout is controlled with a CS3+ and I have 9 x 60832 m83 decoders controlling points and uncouplers. On the layout there are 4 sets of points, for each set 2 points are connected to the same terminal on the m83 ( as you describe). Each set will switch fully. I am using the latest version of the point motors - 74491.

I am afraid I haven’t seen your problem. Is yours a very large layout? Perhaps there simply isn’t enough power for the number of accessories? Perhaps they are older point motors and are failing?

Maybe someone else has seen the issue and can comment.

Adrian

or add a capacitive discharge unit (CDU)

I thought a CDU could be used to give extra curent to the solenoid. i.e. using the discharge to throw the switch motor. Once thrown it starts charging up for the next event, not dissipating the 'on' current.

Hi Ivan,
I used to use a CD unit to power switch matrices on my Lionel layout. These switches use heavy duty Tenshodo switch machines. These were all metal with two big solenoids and could take the shock of switching at high current. The points would "slam over" when the button was pushed. I have doubts about the Marklin switch motors standing up to the high current over time.

Regards,
Roger

Yes, that is another way. Jean complains it is not advised, however Jean, what is your exact complain regarding this solution? I mean 2 decoders with same address? Do you think those decoders may be damaged this way? I wouldn't think so. I might be wrong, but I think decoders don't do any digital communications, they just "sit" on the bus and listen the communication, once they find a message with matching address, they do what is ordered in the message. They don't send any digital message back to the network/control station, do they? If so there isn't any risk of malfunction of digital communication. Only risk is that the layout would not have enough power to finish both switches.

I did another try. I put 1st turnout to the m83, 2nd turnout has built-in decoder. I put same address and they switch correctly. What am I risking, do you see any possibility of any damage?

Btw, this layout will be very small - 100cm x 170cm, one parallel track with switching yard, 10 turnouts, 5 uncouplers, maximum of 2 trains running at the same time. I think there should be enough of power with 1 adapter.

Thanks guys,
Ivan

But each m83 acts as its own capacitor discharge unit, so powering the solenoids is not a problem.

The bigger problem is when it comes to recharge the capacitor, but if the m83 is of a type, or has been modified to use seperate power from the track then this will also not be a problem.

It seems to me that the very first question needs to be teased out before jumping to a singular response.

I am not saying that the points about CDU's and external power sources do not have merit, but they address the power source and not how it is delivered in the context of a m83.

What the original question fails to cover is ...
- the type of turnout motor - M-Track, K-Track, C-Track (for want of some form of grouping)
- the mechanism - specifically 3-way points and double slips.
- condition of the mechanism (armature binds, spring tension (wrong spring), midway adjustment for detent)
- m83 settings changed from the default

The same discussion was rampant with the k83 in the mid 80's and observation (obviously M-Track and K-Track only) influenced my rule of thumb
- Yes - 2 turnouts was OK for K-Track
- No - M-Track needed all the power it could get.


To this level, anyone thinking "why should the track type matter?" has obviously not read the manual, because it shows the recommended values for each type of mechanism.

Anyone thinking this through will realize that the settings in question are in fact altering the pulse duration AND the pulse duty cycle such that we are actually referring to an AC pulse string and not a DC one that is being delivered to the motor. I bring this up because I sense that some think a CDU "dumps" its power in one go (and recharges from the normal power input a bit slower - thus reducing the flicker/dimming of your accessories)

So, I am happy the "CDU = Capacitor" part of the m83 offers a reservoir of power, which brings me to the next point.
Power delivered is a function of both current and voltage - in the case of a pulse for a turnout , that is delivered over a finite period of time.

A brief side point : Items like a solenoid , by its design, has a) a physical resistance to over come and b) potentially better characteristics if the power is pulsed.

Analogue users will appreciate the buzzing of turnout motors and uncouplers - in the later case it has been a benefit to shake the couplings loose as well as lifting them up.

k83 - and 6021 generation users will know that, to give a bit of extra grunt to that "sticky point in the corner" they would hold the button down longer.
In this configuration (6021 with keyboard) you could do that - the k83 port was energized when the finger pressed the button and de-energized when released.
This is not the default for the m83. The m83 receives the "button down command and determines to do the work itself using its internal pulse width definitions without caring about the "button up" command

To cover off the k83 and its CDU capabilities, it may be noted that the k83 delivers 22V whereas the m83 delivers 16V.

The "reservoir" was designed around the expected 250-500mS pulse for a turnout motor. Thus if you held the button down you would extend the pulse, but after the reservoir (Dam @ 22v) was used you would be sucking power direct from the track and transformer . This usually mean that the voltage would drop if there were other loads on the layout.


Combining all this and returning to the m83 with its different "brightness" values for M, K, C Track, it can be seen that the values reflect the improvement in efficiency over time with the net result being a reduced amount of power is "deemed by Marklin" to be needed for (single) C-Track mechanisms, compared to M-Track mechanisms.

So IF you are working with C-Track turnouts in parallel, you might consider increasing the power by adjusting the setting to be more like that needed for M-Track.

Some may think, "why not just set it to max?".
The settings in question address both
- The pulse length
- The PWM dutycycle
an m83 does have the PWM value set to 255 by default , so some of the above may be moot to the "out of the box" user
The pulse width , however, defaults to a value of 4 (which means 200mS) - The manual suggests
- C-Track motors = 250mS
- K-Track motors = 500mS
- M-Track motors = 1000ms (1 second)

Also I will comment that it needs to be understood that a value of 255 in the dimmer setting effectively means DC, with NO AC like pulsing and possibly the loss of a desirable vibrating buzz. (A lesser issue maybe, but I have found stubborn points may be worse at 100% than , say, 75%)
For those contending with poorly designed C-Track motors, there is the possibility of too much power causing the armature to bounce back if the pulse is too strong

In separate threads you can find discussions on the C-track cutoff switch issues (plural)

And finally we come to the m83 port Mode setup, which out of the box is probable 18 , (but is worth checking)
This value indicates "Min. switching with end switch"
So , on literal reading, this lends a lie to the above comment about pulse width being finite as it implies that power will be constantly applied until the motor (armature) has completed its travel.

More, to the point we know from observation that this does not work as such
i.e. as per this thread, the point does not switch fully before power is stopped from being supplied.
Whether this is the part of the circuit that "intelligently detects that power is not being drawn" that is not working , or whether there is a finite value is unclear to me. Either way mode 17 might also be worth considering


In summary.
- You can charge up the m83 (internal CDU?! = Capacitor) and/or provide an external power source to mitigate track power draw.
BUT
- this is meaningless if you do not understand that there are other characteristics of the m83 that decide how power is delivered to a given port

Yes Peter, there is a big problem, lack of information. I have only limited Märklin booklets and internet forums. Is there any good resource in English which would go deeper?

Now I'm going to setup really small layout with 1 parallel track plus switching yard in the middle, so 2 locos running at the same time. I thought the adapter which comes with mobile station handles that. If not I'll put extra adapter to the m83 decoders.

Thank you for your answers.

Ivan