IMPORTANT. Once fitted it is important to check that there are no collisions or parts catching or getting over tensioned during the normal motion of the print head. Manually drive the print head to the extremes of both the X and Z axes checking for any problems.

In my setup the arm is positioned on the far right of the top rail and only supports the PTFE tubes. My hotend cable is supported using https://makerworld.com/en/models/465086#profileId-373859 

Edit. I have had a couple of reports that the dovetails are too tight on the clamp locking piece. There is clearance built into the model  but the fit is deliberately tight. It SHOULD slide in with just finger pressure. If your print doesn't then a little sanding will be needed. It's tight so that it doesn't wobble/vibrate out in use. Check the fit before trying to attach to the printer,

Also the original base is the one to use in most standard setups. If you find that the arm swings down too low and doesn't come back up in use then try the version with the bump stop. The AMSLite position and orientation as well as PTFE tube length and twist/curve all seem to have an effect.

I got fed up of the PTFE guide tubes slumping forward and down so set out to find a guide for them. All of the ones I found didn't really do it for me so I fired up Fusion to make my own. The arm swings up and down to allow for free movement of the tubes.

I did see the dovetail method of holding on to the top bar but I can't remember where. If it was your model please let me know so I can acknowledge you.

Requires one M3x20 machine screw for the pivot point. Slightly shorter should work fine instead. It threads into a 2.5mm hole. I ran an M3 tap through first. Do not fully tighten the screw, it should be done up enough to stop any side to side movement of the arm but still allow it to swing up and down.

As so many people were splitting the ring on the arm to simplify installation I modelled an arm with the ring split. 

Ozzie (@user_3814793868) added a bump stop on the base to stop the arm from dropping too low in his setup. This seemed like a good idea so after asking him if he minds I have created another base that has a bump stop that only allows the arm to swing down to 15 degrees above horizontal. This is about the lowest that the arm gets to on my setup and looks similar to his version. I have my AMS a bit further away from the printer that normal with longer PTFE tubes and he has his AMS mounted vertically so neither are “normal” setups. The original base seems to work well in most cases. 

The model is parametric so if you would like a bump stop at a different angle please send me a message and I will see what I can do. Also arm length changes can be easily done. 

To keep things simple I have now made 4 profiles available.

The first is my original base without the bump stop and original arm without the split in the ring.
The second is my original base without bump stop and the arm with the split in the ring.
The third is the base with the bump stop and the original arm without the split in the ring.
The fourth is the base with the bump stop and the arm with the split in the ring.