Hello Guest it is March 28, 2024, 04:22:59 PM

Author Topic: The Brains  (Read 11726 times)

0 Members and 1 Guest are viewing this topic.

Offline thosj

*
  •  532 532
    • View Profile
Re: The Brains
« Reply #10 on: July 21, 2010, 10:42:37 AM »
Well, this is indeed strange, but I can't repeat the MDI input thing. I have my brains active, they work, and the MDI works. I did some tool changes, moved X and Y around with MDI, everything worked fine.

The only thing I can think I did was clean out my brain directory of unused brains so I could see what I was doing. I didn't deactivate my SRO/FRO brains or do anything else. Just started the machine, ref'd all, tried MDI.

I'm sure I had the problem last week and never put it together with the brains, so I was anxious to see what I could learn. Alas, nothing to learn to help you.

Now, I guess, try the brains I attached above, without the timer section, and see if they work for you.

Tom
--
Tom
Re: The Brains
« Reply #11 on: July 21, 2010, 07:14:00 PM »
A useful test would be to test if things are different on mach 3.43.6.
That would help Brian know where to look.
I believe that mach 3.43.6 is available from the DL page via the archived releases link near the top of the page.

Note: MSM will not run on 3.43.6 - if you try, it will check the mach version and tell you to upgrade mach before MSM will run.
Therefore to get bck to MSM, you'll need to reinstall mach 3.43.10+

Dave
Author of the MachStdMill Extensions for Mach3
www.CalypsoVentures.com
Re: The Brains
« Reply #12 on: July 23, 2010, 01:14:58 AM »
UH, Guys....?

OK, I got curious and at least one of those brains frankly strikes me as a bad idea...

Perhaps I miss read the brain logic - but it seems to me to be saying that

if feed Ovrd < 100% then feedhold the machine (and turn on a user led 2222)

followed by a rung that says

if user led (2222) is on
(which I think is intended to indicate that this brain's prior rung created the feed hold)
AND (now)
200% < feed ovrd < 600%
AND
the mach feed hold led is on (I.e. the program is paused)
THEN
     **** automatically do a cycle start action *****

If I understand this correctly, I could drag the FRO slider to < 100% and a feed hold happens, when I then drag the slider to above 200% the machine starts running the loaded code...  without the operator ever touching a cycle start button.

My first reaction was "really"?!?  Doesn't anyone else think that generating cycle start events by any entity other than the human operator is more than a bit dangerous?

Or is there more to this context (i.e. other stuff this brain is working with) that would mitigate the danger?
Maybe there are other semantics to LED 2222?
Did I miss something here?

...AND a cycle start event will terminate an open MDI input line.
MDI stands for manual data input - where the D in MDI is "data", and data here really means a block of gcode.
You can't be trying to put a gcode block into the input queue to be executed at the same time that the gcode interpreter is running a loaded program.... which pretty much explains to me why the activation of this brain will close an Open MDI input window...

Dave
« Last Edit: July 23, 2010, 02:32:12 AM by DaveCVI »
Author of the MachStdMill Extensions for Mach3
www.CalypsoVentures.com

Offline thosj

*
  •  532 532
    • View Profile
Re: The Brains
« Reply #13 on: July 23, 2010, 07:24:28 AM »
No, no, no. I'm not sure where you're getting 100%, 200%, 600%. It's 1%, 2%, 6%.

If feed override is LESS THAN 1%, feedhold, set LED.

If feed override is GREATER than 2% AND less than 6% AND the LED is on, do a cycle start.

In practice, you turn down the feed pot (analog pot on PoKeys for me), feed slows, but in Mach without this brain, NEVER stops but creeps slowly. With this brain, when it gets to less than 1%, it feedholds, sets LED. Turn the pot UP, it gets above 2% it cycle starts. The 2% to 6% is there because if it's 2% to 3% (like the OP's feedhold brain) and you turn the pot too quickly, sometimes it misses it for some reason.

And this does work in conjunction with the other brain of MY 2, HOOD_FRO (credit where credit is due, Hood, your name in lights!). That one reads the pot, on MOD: 126 IN, PoKeys pin 45 (I think, it's 43-47), and sets feed override 0-150% on the pot.

This, for me, is essential, having analog pots to adjust Feedrate and Speed, especially feed. I can't live without it!!! Old CNC guy from the days before membrane control panels. I can't grab the mouse and drag no steenkin' screen slider to control feed rate!!

There is one issue with this, for some reason (Mach bug?). If you're in Single Block, doing this gets you OUT of Single Block but the Single Block LED on is still lit on the screen!!! Click Single Block off, back on, it's good, but scary the first time you experience it.

The OP's feedhold brain does something different from mine and therein may lie the MDI line problem. I don't have the problem although I thought I did! Might have something to do with the timer in his. I don't understand enough to know about that.

Hood wrote these in one form or another, I modded it slightly for MY PoKeys pin and separated SRO out to a separate brain, they were combined in this brain originally.
« Last Edit: July 23, 2010, 07:33:47 AM by thosj »
--
Tom
Re: The Brains
« Reply #14 on: July 23, 2010, 07:48:06 AM »
Hi Dave
I had better explain the brains and their use and my setup as I would hate to think that others were thinking I am doing anything reckless - although we from Australia may be known for it. The 2 brains work together and quite well in the previous versions of Mach. I have a small vertical mill - converted to cnc as a lot of others would have on this site. It is attached to a computer with screen a safe distance from the coolant I can get to travel a good distance without trying too hard. I wished to have some control of my machine when standing close to it so I made a console with various buttons for reset, cycle start, feed hold etc. and a MPG, E Stop and some pots for feedrate and spindle override. Photo Attached. When turning the pot for feedrate override the Po Keys 55 gives me a output placed in a user DRO which my (Hoods) brain reads and adjusts the feedrate within Mach. The Feedrate Override Brain also automatically resets the feedrate to 100% if you get the pot close to 100% (98% to 102%) so you are not chasing 100% constantly with your pot and this stops the Feedrate Override LED flashing at you for no good reason. When turning the feedrate right down to 1% the feedhold brain automatically feedholds Mach. If you turn your pot for feedrate override up again to 2% and above then the cycle will start - not automatically but by turning up the feedrate pot manually - this has worked well for me in the past. This was the point of the whole post - none of this should have anything to do with the MDI entry line but with these brains enabled I can not type anything into the MDI line - before this beta version of Mach which we must use to try out your screen set it all worked fine. I had first thought it may have been something in your screenset but as I did further testing I found it to be the Mach upgrade which I did at the same time. I have found a work around in such that if using the MDI I disable the Brain and only use it for running a program. My mill is not industrial strength and often find recommended feedrates too fast for it and with my setup I find it easy to adjust the feedrate on the fly. I like your screen set and it's new features - like the MDI on multiple pages and the probing inclusions. I thought that I would put these comments up to see if anyone else was having the problem and for some feedback. I'm sure we will find out why this is happening to me and it probally will just go away at the next Mach Upgrade!
Kind Regards
Steve
Re: The Brains
« Reply #15 on: July 23, 2010, 10:37:11 AM »
OK, I'll buy that the numbers are 1%, 2% and 6%.
I guess I was tired and when I read the brain immediate compare operation to the value of 1.0, I converted 1.0 to 100% (thinking that 1% would be in the 0.01) - but I forgot the units on the DRO are %.

So we have a brain that, as long as the FRO is between 2% and 6%, it will cause a cycle start event. If the FRO is set in that range, the cycle start events will be generated at whatever the run period of a brain is (probably the mach 1/10 sec background cycle - but I'm not sure, only guessing re the actual brain run loop run period).

Does the "I can't type into the MDI line before it closes" happen only when the FRO is between 2% and 6%?
I'd guess that if the FRO is reset to 100%, the "I can't do MDI" symptom does not occur?

Since a cycle start will close the MDI Line, this means that the MDI line is getting cleared every  1/10 of a second.
That would seem to explain why one can't get type text into the MDI line - right?

Dave
Author of the MachStdMill Extensions for Mach3
www.CalypsoVentures.com

Offline thosj

*
  •  532 532
    • View Profile
Re: The Brains
« Reply #16 on: July 23, 2010, 10:50:01 AM »
So we have a brain that, as long as the FRO is between 2% and 6%, it will cause a cycle start event. If the FRO is set in that range, the cycle start events will be generated at whatever the run period of a brain is (probably the mach 1/10 sec background cycle - but I'm not sure, only guessing re the actual brain run loop run period).

I think the brain issues a cycle start WHEN FRO gets back to 2-6%, not AS LONG AS it's between 2 and 6, IF I understand the logic. It should only issue a cycle start ONCE, but I could be wrong, probably AM wrong!!

Steve, the OP in this thread, does not have 2% to 6% in his brain, he has 2% to 3%. I found that to miss Mach's 1/10 sec too often, so changes to 2-6.

And, at any cost, I DON"T HAVE THE MDI PROBLEM on Mach 043.010, either with 1024 screens or MSM, so Steve will have to speak to that. I don't understand the logic in his FRO brain, his feedhold brain is essentially the same as mine other than his is 2-3%, Hood's original, and mine is 2-6% after discussing it with Hood.

I'll get out of here as I'm NOT having the issue!!!

Edit: I just went on the machine, turned down the pot 'til it feed holded (held?), turned it up to 4%,  and I could type in the MDI box OK, so that's not the problem, I don't think.

« Last Edit: July 23, 2010, 11:01:53 AM by thosj »
--
Tom
Re: The Brains
« Reply #17 on: July 23, 2010, 12:29:22 PM »
Hello Guys Dave contacted me and told me to look into this.. That darn Hood found a way to make it not work LOL.. I would like to know if any of you know what function in the brain is causing the error? I need to look into this and I am also looking into the errors that Ozzie has had with Probing. My goal was to get this out today but I don't see that happening if there is a brain problem :(

Thanks
Brian
Fixing problems one post at a time ;)

www.newfangledsolutions.com
www.machsupport.com
Re: The Brains
« Reply #18 on: July 23, 2010, 01:09:52 PM »
Well after testing for 30 min I can't make it fail with the brain that I got from this post.. I am ony using the screen and the slider on the screen but it should give me the same results that you are all seeing.. please tell me if I am doing something wrong

Thanks
Brian
Fixing problems one post at a time ;)

www.newfangledsolutions.com
www.machsupport.com
Re: The Brains
« Reply #19 on: July 23, 2010, 06:00:23 PM »
Hi Brian
To reproduce what I was seeing please just load my consoleoverrides brain from earlier in the post and enable it. You don't need any hardware connected as I get the same effect on my house computer. With the brain running just click into the MDI line and try to type something. Mine deactivates the MDI before I can type more the a few letters.

I have just rewritten the brain without the timer module and it now works fine.

I promise I was only doing what Hood said to do and I was not meaning to cause any trouble :)

Kind Regards
Steve