Optimum impact speed for the claw?

[strask]

When using the advanced grabbing unit ("the claw"), moving too fast or too slow causes either a complete collision or a bounce or a bump instead of actually grabbing. For docking vessels using the claw, I usually try for about 0.8-1.5 m/s and that tends to work out ok.

But I've got a mission to retrieve a random engine floating in LKO, it's a "thud" radial engine, and no matter what speed I approach it bounces off the claw instead of getting grabbed.

Anyone have knowledge of the "correct" speed to use or something else I might be doing wrong?

[George Kerman]

My approach that works almost always: Approach at 0.3 - 0.5 m/s (I know it's a cruise control speed but...). In some cases, you still bounce off it (bad orientation). Have RCS activated. After the bounce, accelerate again towards the object (immediately) either with RCS or with normal thrust if you are correctly positioned. The second grab attempt works 90% of times in my case. If not, warp and unwarp (to cancel rotation) and position yourself in a different orientation so you grab a different area of the object (you do have RCS right?).

This works on Thuds, Labs, Kickbacks and anything in between.

By the way, it goes without saying but you do arm the claw and have the "Control from here" for the claw activated right?

[strask]

Thank you so much for the tips! I approached at 0.3 m/s and was poised over the full-throttle key (my tug has wimpy engines so full throttle would be appropriate) but it grabbed on the first try this time. I had previously only gone as slow as 0.5 so I guess if speed was the issue, 0.5 is a bit too fast for such a small part.

And yes, I have RCS, armed the claw, and both control from here and "aim camera" to get a good view of the grab attempt. :)

[George Kerman]

Great work!
Next time, have lights too! :P (all my attempts happen at night because KSP).

By the way, I hope you have some radiators and a heat shield right?
A probodobodobodobodobodyne HECS would also be cool for retro hold on re-entry.

[strask]

Yeah I'll screenshot my re-entry setup, it involves a 3.75m heat shield, probe core, and reaction wheels -- even fins to keep the heat shield facing into the heat.

[strask]

http://steamcommunity.com/sharedfiles/filedetails/?id=799214229

[Joschi]

Nicely done!

Do you use KER in 1.2.x?
Is there a dev version I am not aware of or did you edit the req. KSP version within the mod?

[George Kerman]

I like your design, it's pretty cool. I'm kinda the cheapest when it comes to missions and this crap:

http://steamcommunity.com/sharedfiles/filedetails/?id=799218658

has recovered even labs and Kickbacks from LKO successfully. If it works and it's cheap it's good for me.

[George Kerman]

Originally posted by »BoH.Crew×IRA™:

Nicely done!

Do you use KER in 1.2.x?
Is there a dev version I am not aware of or did you edit the req. KSP version within the mod?

Last KER works with the last update as well, even though it gives that warning when you load the game. But do make sure you have the last KER version (perhaps use CKAN). You don't really need to edit the version.

[strask]

Originally posted by »BoH.Crew×IRA™:

Nicely done!

Do you use KER in 1.2.x?
Is there a dev version I am not aware of or did you edit the req. KSP version within the mod?

Yes I use KER in 1.2.1, the version of KER I am using is 1.1.2.0 -- The addon version checker issues a warning at startup that KER was built for KSP 1.2; but that doesn't keep it from working.

Originally posted by George van Doorn:

I like your design, it's pretty cool. I'm kinda the cheapest when it comes to missions and this crap:

http://steamcommunity.com/sharedfiles/filedetails/?id=799218658

has recovered even labs and Kickbacks from LKO successfully. If it works and it's cheap it's good for me.

Nice!

[strask]

Yeah George your recovery setup is 1/5 the cost of mine, lol.

[maj.solo]

I think it behaves just like docking ports. Only requirement is angle and that you hit a flat side and that there is no stuff there allready like other parts or a seam. Cause then you be hitting the other surface at a bad angle instead.

The most difficult part I ever had to contract recover was a single free floating drill-o-matic. Just angles everywhere and stuff and also it is mostly smaller then the claw. Hitting it perfectly on the backside dead center was the only way. But at all other times it work very reliable for me.

I reason like this that it is not the claws fault. Everything is my fault. There are rules the computer program follow. I just need to follow them. But the claw does not work as good as docking ports. So I would never use it instead of docking ports.

Usually I do all dockings at 0.1 m/s. If I am getting boored and think it is going to slow I do not thrust more I time accelerate instead. Accelerate, correct, accelerate, correct course, acelerate, correct, etc and always slow down just before impact.

100 meter from target 2 m/s
30 meter from target 1 m/s
10 meter from target 0.3 m/s
5 meter from target 0.2 m/s
3 meter from target 0.1 m/s

and then I keep it like that all the way in and when

0.5 meter from target break to keep the speed low. Don't matter on light ships but is a good routine when docking multi million dollar ships where you use the new auto strut and rigid mount that seem to hold the ship together very well up to a point where there might be some mathematical error and the ship blows apart too easily. To avoid that you dock slow ... very slow. Good to get into the habit of doing it slow.
I almost think the old way with having a flexible ship where you yourself struted parts and in the direction where forces where ... those designs ... even if heavy .. took the docking better. And if the whole thing started to rock after dock you just time accelerate. Only problem with struting manually is the part count. Otherwise I liked the engineering of sitting there thinking it all out and placing the struts.

[George Kerman]

Yeah but you're winning big on artistic impression and on that (overkill) 3.75 heat shield. Usually you don't even need a heat shield or radiators if you do a soft descent (set peri at ~ 45km , apo 70-100 km) for most recovered parts.

[strask]

Originally posted by George van Doorn:

Yeah but you're winning big on artistic impression and on that (overkill) 3.75 heat shield. Usually you don't even need a heat shield or radiators if you do a soft descent (set peri at ~ 45km , apo 70-100 km) for most recovered parts.

Haha thanks, indeed no heatshield needed for many kinds of parts and a soft descent; I was mostly thinking: 1) Occasionally might retrieve a very heat-sensitive part, and 2) don't want to redesign every contract. But seeing how much less expensive yours is I might redesign a little. Or just steal yours. :P

[strask]

Originally posted by maj.solo:

Usually I do all dockings at 0.1 m/s.

Yeah for docking ports, but at least when the claw was first introduced, it wouldn't grab if you were going that slow. I'll have to test some more because they seem to have improved it.

Originally posted by maj.solo:

0.5 meter from target break to keep the speed low. Don't matter on light ships but is a good routine when docking multi million dollar ships where you use the new auto strut and rigid mount that seem to hold the ship together very well up to a point where there might be some mathematical error and the ship blows apart too easily. To avoid that you dock slow ... very slow. Good to get into the habit of doing it slow.
I almost think the old way with having a flexible ship where you yourself struted parts and in the direction where forces where ... those designs ... even if heavy .. took the docking better. And if the whole thing started to rock after dock you just time accelerate. Only problem with struting manually is the part count. Otherwise I liked the engineering of sitting there thinking it all out and placing the struts.

I haven't gotten into the habit of using auto-strut much yet. I like the engineering of the strut placement, like you, and I also think that having space tape all over the place is just more kerbal.