Decoupler - Upgraded version

[Lowa]

For those who were wondering, yes, I’m still alive! I apologize for disappearing, things have been / still are pretty hectic with 4DBrix. However, I would like to try spending more time again keeping in touch with the community again.

I haven’t had that much time to work on new products but I did manage to finalize the decoupler. We made a video to show the new decoupler in action!

We redesigned it because the initial decoupler had a few shortcomings:

• when the train wheels are in front of the ‘claw’ of the decoupler, it cannot be closed. From a distance, it’s hard to see whether the train is positioned correctly or not. Therefore, we added a 'magnetic sensor' to the decoupler. That sensor can detect the magnets of the LEGO train couplings. As such it’s easy to verify the positioning of the train.

• the initial design had a detachable motor comparable to our track switch motors. That was a straightforward solution but it was 8 studs wide. That meant that, when integrating the decouplers in a rail yard, the motors were touching the neighboring tracks. To avoid that, we integrated the motor into the decoupler giving it a much smaller foot print; that also allowed us to reduce the height.

Some additional remarks:

• For the video, we replaced the PoweredUp controller and motor of the 60198 train by a PF motor and our WiFi train controller. Controlling the PoweredUp trains is work in progress...

• Positioning the train on the decoupler is not that easy. Therefore we implemented a small script in a generic tile that positions the train automatically based on sensor feedback. In the video, we ‘manually’ parked the train with the first car on the decoupler and then activated the auto positioning: it slowly moves the train backwards until we get a +90% reading on the magnetic sensor. That turns out to work very well. It’s also intriguing to watch because it’s the automation system driving the train autonomously...

The hardware is ready and we hope to be able to make a first batch of decouplers the coming weeks and the release it… From the software side we might have to make a few extensions to the train control panel to improve the control of the train at low speeds, but the basics are there.

As usual, let us know what you think!

[lostdriveway]

Very nice!

[ProvenceTristram]

4 hours ago, lostdriveway said:

Very nice!

Yeah, it's great until you find out that 5 men were killed during the R&D process.

[Lowa]

6 hours ago, lostdriveway said:

Very nice!

Thanks!

[knotian]

ProvenceTristam

That's not an unusually large number for a development team. The important thing is to determine whether it was suicide due to the complexity, or murder by management because they could not do it over-nite.

[Roadmonkeytj]

@Lowa

This is brilliant....  Are there other videos of the mechanism? 

[Lowa]

4 hours ago, Roadmonkeytj said:

This is brilliant....  Are there other videos of the mechanism? 

Thanks!  I think the redesign was worth it...  

I don't have any additional movies at this point.  I have to make a close up of the mechanism when we actuate it. The images below show the decoupler in the 'open' and 'closed' state.  It holds the cars in place by blocking the wheels.  We felt was the most universal solution that should work for all standard and custom cars.

 

[Lowa]

I got a few private questions that might interest some of you:

1) Do you need to raise the track to install the decoupler?

No, the base of the decoupler is flush with the base of the tracks.  This is a side view of the decoupler

2) Can you ballast it, i.e. does it have anti studs ?

Like all our other tracks (we remodeled all of them) it has the maximal amount of anti studs possible. Some areas don’t have anti studs because the internal mechanism doesn’t allow it. We preferred that over increasing the height of the decoupler.  The image below shows the anti studs, all anti studs are fully functional.

[sed6]

This is a very neat design. Looks well thought out and constructed. My only concern, what if you tried to decouple maybe only the one last car instead of all of them. Then you have the locomotive pulling on two cars with the third being held stationary, the couplers with the weakest magnetic bond will be the ones to separate, which won't necessarily be the one you're trying to actually decouple. 

[LEGO Train 12 Volts]

Interesting stuff!

[Lowa]

On 12/4/2018 at 9:19 AM, sed6 said:

This is a very neat design. Looks well thought out and constructed. 

Thank you.  We did take our time to get to this design...

On 12/4/2018 at 9:19 AM, sed6 said:

My only concern, what if you tried to decouple maybe only the one last car instead of all of them. Then you have the locomotive pulling on two cars with the third being held stationary, the couplers with the weakest magnetic bond will be the ones to separate, which won't necessarily be the one you're trying to actually decouple. 

We had the same concern when we started developing the decoupler but it doesn't appear to be an big issue, the train does seem the split at the decoupler.  I just tried again with the 60198 that was on the demo layout and it splits at the decoupler.  I even changed the order of the cars and it still worked.  I'll try to make a video ASAP so you can see it.

The only exception we have are two coupling that are clearly weaker than all the other couplings (they are from a 60051 set).  When I use those it doesn't split at the decoupler but at those couplings.  I only have 2 that are weaker and give me issues, I'm not sure how common that is.  I case the strength of your coupling varies significantly you have to arrange them from strong (front) to weak (back) in your train.  If you do that the train will still split at the decoupler...

On 12/4/2018 at 3:20 PM, LEGO Train 12 Volts said:

Interesting stuff!

Thanks!