Control system response time and bandwidth?

[KenWittlief]

has anyone done any testing on the new control system to measure the following parameters?:

1. when you move a joystick, how long until the change in value is output to the mapped PWM (victor) when using the radio link?

2. if you change a PWM output in the code, how long until it is output to the PWM cable?

3. how long does it take to send out the full PWM command? ie, how long is the PWM signal period?

4. When an analog or digital input is changed at the input pins, how long until the code sees the new value in the corresponding variable?

I didnt see this info in the specs for the OI and RC - if you know where to find it please point me in the right direction - if you have NOT actually measured it with a scope, please dont guess - if nobody knows for certain I will drag my digital scope to our site tonight and measure them.

[Random Dude]

Originally Posted by KenWittlief

has anyone done any testing on the new control system to measure the following parameters?:

1. when you move a joystick, how long until the change in value is output to the mapped PWM (victor) when using the radio link?

2. if you change a PWM output in the code, how long until it is output to the PWM cable?

3. how long does it take to send out the full PWM command? ie, how long is the PWM signal period?

4. When an analog or digital input is changed at the input pins, how long until the code sees the new value in the corresponding variable?

I didnt see this info in the specs for the OI and RC - if you know where to find it please point me in the right direction - if you have NOT actually measured it with a scope, please dont guess - if nobody knows for certain I will drag my digital scope to our site tonight and measure them.

Well according to the IFI page:

Digital Input Frequency (typical) 50 KHz
Analog Input Access Time 10 μSec

The analog time seems reasonable according to the actual specs for the microcontroller, though I do wonder about the digital what exactly they are measuring with the digital inputs, since you should see the change almost immediately, (since the digital IO pins are connected directly to the I/O ports on the micro)

[KenWittlief]

my primary reason for needing these timing specs is for using closed loop control with the RC - I need to get a handle of how fast sensor inputs make it to the actual input variables, and how long it takes from the time the code sets a PWM variable until the affected motor actually sees the change across its terminals.

Im also curious about the entire system response time, from the joysticks to the motor terminals to get a idea of how quick the robot will respond to driver commands. I measured it last year with the old control system and it was surprizingly slow -if you watched the voltage on the victor outputs with a scope, and bumped the joystick, the delay was very noticable - from the top of my head I think it was at least half a second, maybe longer.

[KenWittlief]

I found the thread from last year that deals with the same sort of specs on the old control system - if you read it keep it clear that everything being talked about is LAST years system, and the new system is very different - so I dont know how much of it is still true.

http://www.chiefdelphi.com/forums/sh...ighlight=delay