Just some questions out of curiosity

[Varryl]

I've been playing Kerbal since .18.2 and I've done some interesting things. I would consider myself an intermediate player, but not an advanced one by any means. I started playing around with the part cfg files, launched some very high thrust experiments, and I had some questions:

1. Does anyone know exactly how much force is used when a part is attached to another part, or how much force is required for that part to break off?

2. Does anyone know how to increase the attachment force from one part to another part?

3. Why do rockets that are perfectly simple (pod/probe, fuel, engine) with a high TWR start to veer off one side or another?

4. Does anyone know how to adjust the ASAS force so that you're not constantly wobbling back and forth while still maintain ASAS stability features?

These questions have probably been asked before. I used search and tried to come up with some answers for myself, but the threads seemed to veer off into something tangential every time. Any information or even a link to where the information is would be greatly appreciated.

[Flixxbeatz]

This might help you: http://wiki.kerbalspaceprogram.com/wiki/CFG_File_Documentation

It's not that easy to modify the cfg files though, there's a chance that you'll mess it up.

[Cranium]

For ASAS, I believe you can use Caps Lock to enable fine control, but I rarely use it. If my rocket can't be controlled decently with manual maneuvering, I usually go back to the drawing board, and either add more struts, or redesign it completely.

ASAS is only going to be able to control it if you are able to control it with perfect(?) precision.

[Awaras]

If you want to play around with the ASAS force settings, try this mod:

http://kerbalspaceprogram.com/trimmable-asas/

[Varryl]

Thanks for all your input guys. I thought I knew what I was doing in the cfg files but looks like there is a lot more than I realized. Does anyone know what Ki, Kp and Kd do in layman's terms? I've googled around, I've seen them in parts cfg files, and I know they're for PID controllers, but the math was a little over my head.

[Awaras]

This is the clearest explanation I could find:

Although you'll find many methods and theories on tuning a PID, here's a straight forward approach to get you up and soloing quickly.

1. SET KP. Starting with KP=0, KI=0 and KD=0, increase KP until the output starts overshooting and ringing significantly.

2. SET KD. Increase KD until the overshoot is reduced to an acceptable level.

3. SET KI. Increase KI until the final error is equal to zero.

[Mr Shifty]

Thanks for all your input guys. I thought I knew what I was doing in the cfg files but looks like there is a lot more than I realized. Does anyone know what Ki, Kp and Kd do in layman's terms? I've googled around, I've seen them in parts cfg files, and I know they're for PID controllers, but the math was a little over my head.

A control loop has some measured quantity, a setpoint, and some control parameter. For example, a thermostat for an electric heater measures temperature and could control the amount of power going to the heater coils. It measures the difference between the measured input (temperature) and the setpoint (thermostat setting) to determine the error signal.

Kp - Proportional. Kp represents how responsive the control output is to the magnitude of the error signal. In the thermostat example, if the error signal is large (large difference between current temp and setpoint), you want the power to the heater to be large, and you want it to decrease the closer you get to the setpoint.

Kd - Derivative. Kd represents how responsive the control output is to how fast the error signal is changing. Back to the thermostat: if you just rely on proportional control, you'll probably overshoot the setpoint, and since you have no "cooldown" method, you're probably going to undershoot when the temperature comes back, which will cause you to overshoot again, oscillating back and forth. (This is called ringing.) If you make the system responsive to the rate of change of the error signal (typically inversely proportional), it will reduce the furnace power not just when the error signal is low, but also when it is changing quickly, thus damping down overshoots.

Ki - Integral. Ki represents how responsive the control output is to sort of the total duration of the error signal. For instance, if you've had a large error signal for a long time (say because someone left the door open on a cold day), clearly your output power isn't high enough, so you need to increase it. You can sort of think of Kd as the system response to short-term trends and Ki as the response to long-term trends.

[Varryl]

I really appreciate all the feedback I've gotten on the forum. Good stuff, guys. Cheers.