Docking port and SDK

[n0mad23]

In a recent email exchange, Tblaxland wrote the following:

[FONT=Arial, Helvetica, sans-serif]When you undock a vessel in Orbiter the docking mechanism creates a longitudinal velocity between the two vessels simulating springs in the docking mechanism. Estimated magnitude of error +1.5m/s.[/FONT]

I'm wondering is there a way to disable this via SKD? Or perhaps a way to work around this feature?

 

[computerex]

Well, as far as I can see, orbiter docks to ports together when they are in a certain range, if you disable the spring effect, the vessel would just stay there, and the two ports would probably end up docked again...

 

[n0mad23]

Well, as far as I can see, orbiter docks to ports together when they are in a certain range, if you disable the spring effect, the vessel would just stay there, and the two ports would probably end up docked again...

Sure in the standard application, but with our Ananke tether-sling the centrifugal force eliminates this problem. But you can see that the spring effect adds one that we probably want to eliminate.

The tangential speed of our tether-slings' rotation gives about 1.5 km/s Delta-V boost upon undocking.

 

[tblaxland]

Perhaps when the vessel code comes further along, Payload Manager would be the solution instead. That is, we use an attachment point instead of a docking port.

 

[computerex]

Well as I had expected, the vessels re-dock upon undocking. NegateDock posted as attachment, along with the source. Try it with the Smack! scenario.

 

[n0mad23]

Well as I had expected, the vessels re-dock upon undocking. NegateDock posted as attachment, along with the source. Try it with the Smack! scenario.

I've been thinking about this as I've been taking apart my musical instruments and putting them in boxes today. I suspect you're absolutely right about the redocking occurring, even with centrifugal force pulling the payload away. The docking mechanism in Orbiter is really 'friendly' in this sense, but a detriment to what we're trying to do. I've downloaded your attachment and will give it a whirl when the opportunity arises later tonight. Thanks.

Tblaxland, you've got the solution. Payload Manager and attachments instead of a docking port will solve this.

Still, it was a good excuse to start a thread and discussion on the SDK thread, no?

 

[computerex]

Well the thing with attachments is that they are unrealistic. They use simplistic physics, with the child having no physical effect on the parent. You can easily apply linear thrust of the child to the parent, but that's just a hack. I think a method in the API which will allow one to supply a user defined undocking velocity will do the trick.

Ps. The NegateAttack was just an experiment, no doubt filled with many bugs...

 

[tblaxland]

If have understood Payload Manager properly, it is designed to overcome this unrealistic behaviour of attachments. From Kulch's site:

The influence from the payload ship is also transferred to the carrier. It changes the dynamics of flight (the moments from payload's engines, change its mass, the cross sections area etc.)