Novice inquiry about Picaxe for motion control

C2sr

New Member
All, Please forgive me in advance for asking questions before reading the manuals.

I wish to use the Picaxe as a multi-axis trajectory generator. (Step and Direction with encoder feedback plus I/O.) I am interested to know if each axis can run its own acceleration and velocity, and what the min/max accel and vel values are. What encoder type is compatible? (Single ended and/or differential?) If I am running 5 axis and encoder, with a home and +/- limits each, which Picaxe should I get?

Thanks in advance for your assistance in my exploration of what seems to be a vastly versatile tool which I have put off utilizing for too long.

Chris
 

moxhamj

New Member
Can you describe the application a bit more - is this for a rocket or a motor controller or something else? There are lots of accellerometers available and many give a 0-5V range which is perfect for readadc or readadc10 on a picaxe. Page 9 of the picaxe manual describes how many ADCs each chip has. Eg if you want 5 then a 40X1 or 28X2 would work. But I would start simple and just get one accellerometer and a picaxe 08M to start with as it is cheaper and you can do some experiments. Then move up to the bigger chips once you know it works.
 

C2sr

New Member
Dr. Acula,
Thanks for your interest and your reply. I will be controlling servo and stepper motors in robotic pick and place handler assemblies that are used by IC manufacturers. The controllers for the Servo and Stepper motors only require step and direction input. Position/encoder feedback is a plus for absolute positioning. My task includes reliability testing at the sub assembly level, but I would like to it do without a PC for each sub assembly under test. Using a Picaxe to run multiple axis, (X, Y, Z and Theta.), in a standalone environment, is what I am trying to achieve. Is this too complex for a PIC?
Regards,
C2sr
 

moxhamj

New Member
This would be a perfect application for picaxes. Have a read of the help in the picaxe program - there are 3 manuals and the stepper/servo control examples are in the third manual. This could be done in modules - each module might contol one motor/servo and its feedback sensor. 08 or 08M chips could do this. Then they could talk to each other via a common bus. Or a single bigger chip could do the job. Ultimately the system could interface with a PC but that could come later.
 

C2sr

New Member
Very well. I appreciate your time. I will study the manuals and report back. Thanks again. C²
 

BeanieBots

Moderator
Can you explain in more detail which part you need the PICAXE to do.
It's just that the last pick & place tester I built took 3 months with a dedicated team of six (two just for software) a budget of £350k and the DO/Production resources of fortune 500 manufacturing company!
The multi-axis controller (VME based 8 axis) was £5k alone.
 

C2sr

New Member
Normally yes, I use a multi axis Oregon Micro Systems (OMS) Tragectory Generator in a PC or VME to generate the PWM (Step and Direction) to a servo or stepper driver. I currently use Teknic servo controls for the servo motors and Intellegent motion systems for the stepper motors. Right now I write script in the OMS edior to home and move to the acceleration and velocity specified and cycle accordingly. A typical Acceleration would be in the range of 1,800,000 in/sec² with a Velocity of 400,000 in/sec. Move distances can be any where from 12 to 60". If I have 7 multi axis test fixtures, I must have 7 PC's. This is a cumbersome solution and I wanted to see if the Picaxe running simple routines could handle the tragectory specifications indicated above. (3 months for a PnP tester? That's a pretty good job!)
 

BeanieBots

Moderator
At the end of the day, you are asking $5 device to do a $5k job!
The PICAXE takes around 250uS per instruction so that should give you an idea of the limitation on reading encoders and updating outputs.
It is also limited to 16-bit integer arithmatic so calculating splines on the fly will not be possible without the aid of lookup tables. The memory is limited so smooth lookups would be tricky without external memory which would slow it down even more.
About the best you could do is make it move to a given position at a leisurely pace.
An option that might be possible (if you can get hold of a controller) would be to get the PICAXE to send the SECSII commands. It's far too complex a protocol to fully implement in a PICAXE but the lower level packets might be possible. ie just the send, get the ACK and ignore the reply or just 'look' for an error message.
It really depends on what it is that you want to 'test'.
 
Top