high speed encoder to prm with AD-06

[elevmike]

Robert,

I think your trying to figure in more then what's necessary. Basicly 1 revolution = 1 wind. After every 1/2 wind, move the reel over by 1/2 the cable OD till you get to the other end of the reel, then start stepping back 1/2 cable OD for every 1/2 revolution etc...

Basicly from what I can see the working parameters are Reel Width, Cable OD & Revolutions. Number of Turns Per Layer = (Reel Width / Cable OD). The operator puts in Cable OD & Reel Width, and the code should be able to figure the position of the reel based on how many turns have been placed on the reel.

 

[g.robert]

Hello Mike. At first thought this may produce a stepping motion as opposed to a fluid movement but at such a slow speed it may not matter.

We have a proportional valve (0-10 VDC) that is to control the reel shift from left to right. I was trying to come up with a linear speed reference, take cable OD from HMI and use some math to get a linear output to the valve. It would be the same output regardless of direction. The direction would be controlled via another external valve with 3.5VDC strokes one way and 8.5VDC strokes the opposite. It is proportional as well but I don't need it to be. Just fully stroked one way or the other.

What do you think? Am I missing something?

 

[elevmike]

Robert, I dont think your missing much given the project parameters (use of the valve to move the reel). Personally I think I would have done it with a lead screw driven by a servo, but there may be good reasons for the hydraulics and the program code would be similer anyway.

Check your PM. Mike.

 

[elevmike]

Regarding stepping,

It's all about stepping, if you want a smother motion for maybe a large dia cable you could step every 1/4 or every 1/8 of a turn.

Regarding the horizontal movement/positioning. Can you provide some specifications on this; like what is the range in inches, and the resolution per inch or mm etc?

 

[g.robert]

Hello Mike.

We lost DSL for a couple of hours. I spoke with the mechanical side. I have to stay with the basic concept originally stated.I made a little progress. I set it up to load the middle of the total count value as the starting point of UDC at power up. They cannot turn enough times either way to bottom out the count. This way I get my direction signal working. It is attached for your review. I think all I have to do is work with my CTA 174 (current count) and V 10026 (previous count) to get a reference of current speed. Please let me know what you think.

I believe I'm on track now but not out of the woods just yet. I'll still need to work (first time) with a EZ-220P to set up operator inputs.

I did not get a PM from you. I am going to head out of the office in a few minutes. If I don't hear back from you, I'll be back on it tomorrow.

 

[elevmike]

Robert,

My in PM box was full try again. Meanwhile I'll download your code.

 

[g.robert]

mike you still out there?

I've made some progress. I am stuck on a conversion with scaled output. I know how I do it in AB but can't figure it out with this AD-06.

I have two memory locations. #1 (CTA174) is the current count. #2 (V10026)is a sample taken from cta174 every second. The idea is to subtract where I was from where I am to get current 1 second pulses. I should have a min of 0 for zero movment and max of 643 for max speed. I need to output to v2003 proportionally 0-4095 based on the reference of 0-643. The screen shot shows my last attempt trying to follow the user's manual.

Any ideas? I still need to get this working for counting down as well but if I can get the output working one way, I'll know what to do to reverse it.

 

[elevmike]

Robert,

As I understood it, what your attempting to do is move the reel horizontally along the axel based on where you expect (or want) the current wind to be placed on the reel. Additionally it seems your attempting to do this by speed and time.

If my understaning is correct, I would suggest you back up abit and take a much simpler approach:

The encoder on the reel will give you the rotation. For evey 1/4 rotation, move the reel horizontally by 1/4 the dia of the cable. Since you know the reel width, when you get to the other end of the reel reverse the process. If you want a smoother moving operation try 1/16 proportions.

Each reversal is a new layer.

If the cable is round each additional layer will be + (Cable dia x 0.866). Given the dimensiion from the inside to the outside of the reel, you should be able to determine when the reel is full and cause an automatic stop.

 

[g.robert]

Mike, makes sense to me but I feel I am very close to having a working program using the original logic. I am stuck on scaled analog output using the counter input difference 0-643 (as oppossed to a full scale input 0-4095)desiring the output to be 0-4095 (0-10 VDC). I would then need to add some math to further manipulate the output to desired levels based on speed and cable OD.


Then again, I could be totally wrong with my counter reference scheme. Did you take a look at the attached program? Does it make sense to you? I will talk with mechanical concerning your idea today.

I also thought I was stuck with integer values (no floating point math). How would I use ".866"?

Question is, I am closer to completing what I have than starting from a different logic?

 

[elevmike]

Robert,

I havnent looked you your code yet becasue this PC only has DS4, I have to get to the lab to look at YOUR DS5 code. Additionally, last week I pounded out a code using the stepping scheme I've been talking about, but that's also on the other PC. I will try to look at your code, & post my example tonight.

Later, MPL

 

[g.robert]

Mike, thanks. I ran your stepping idea by mechanical. I await their response as to it's acceptabillity for this application.I would like to know what you think of my work so far as a viable solution as well as look at your potential alternative. Thanks.

 

[elevmike]

Robert,

Within the same code you could alter the proportional stepping variable so as to cause it to be a mechanically smooth or constant motion. I only used 1/14 or 1/16 as an example. You could step by 1/1024 if necessary...

 

[g.robert]

Well I got most of it working including the scaled analog output. They were not open to the stepping motion out of concern of it's effect on the hydraulic system.


Now I am starting with EZtextedit (again first time user). It will not load. I have XP Pro SP2. It says on the web sight (at AD)software works up to win 2000. Anyone know of this software's compatabillity with XP Pro SP2? I searched AD to no avail. Forgot to call them and now it's to late till Monday.