#20555 - 11/19/08 06:14 AM
Re: Timer Troubles...
[Re: Nathan Gregory]
|
ADI Tech Support
addict
Registered: 12/22/02
Posts: 608
Loc: Branson West, MO
|
In C-Max, click Controller Utilities, Set Controller Clock to PC Clock. See if that gets your timers working again.
Kevin Barrett
ADI Tech Support
|
|
Top
|
|
|
|
|
#20556 - 11/19/08 08:43 AM
Re: Timer Troubles...
[Re: ADI Tech Support]
|
Nathan Gregory
active contributor
Registered: 02/17/03
Posts: 157
Loc: San Jose CA
|
Hi Kevin,
Need some coffee this morning? Or had too much? 
quote "tried the 'set controller clock to PC clock' function all to no avail. The latter sometimes advances timers by one count, but does not restart them." end quote...
So, no it does not restart them...
Any better suggestions?
Nathan
|
|
Top
|
|
|
|
|
#20557 - 11/19/08 09:07 AM
Re: Timer Troubles...
[Re: Nathan Gregory]
|
ADI Tech Support
addict
Registered: 12/22/02
Posts: 608
Loc: Branson West, MO
|
Sorry, typed before I had any:-)
Check the clock (controller utlities, get controller clock). Does it report a valid time. Try it a couple of minutes later and make sure the clock is correctly updating. The timers in the Ocelot are based off the real time clock in the Ocelot.
If that is all good, look at your code. See the there is anything that is setting a clock to 1 based on a constant event.
For example:
If time of day is > than 0600
Then timer 0 = 1
In this example, as long as the time of day is later than 6 am, the timer would be continually set to 1 and never count. You would change the "is >" to "becomes >" in the If statement.
Another example;
If Variable 3 is = to 1
Then timer 0 = 1.
In this example, as long as variable 3 is setting at a value of 1, the timer would be constantly being reset to 1 and never count up, You would change "is =" to "becomes =" in the If statement.
Kevin Barrett
ADI Tech Support
|
|
Top
|
|
|
|
|
#20559 - 11/19/08 09:53 AM
Re: Timer Troubles...
[Re: Nathan Gregory]
|
ADI Tech Support
addict
Registered: 12/22/02
Posts: 608
Loc: Branson West, MO
|
You said "I tried a power cycle and reload again, checking timer operation with no program loaded. Ditto. No advance."
What is the application version being reported by the Ocelot at the bottom of the Controller Access Screen?
Kevin Barrett
ADI Tech Support
|
|
Top
|
|
|
|
|
#20561 - 11/19/08 10:19 AM
Re: Timer Troubles...
[Re: Nathan Gregory]
|
ADI Tech Support
addict
Registered: 12/22/02
Posts: 608
Loc: Branson West, MO
|
At this point then I think it needs to come to me to be looked at. Something is not right.
Kevin Barrett
ADI Tech Support
|
|
Top
|
|
|
|
|
#20581 - 11/24/08 09:47 PM
Re: Timer Troubles...
[Re: Nathan Gregory]
|
Nathan Gregory
active contributor
Registered: 02/17/03
Posts: 157
Loc: San Jose CA
|
Ok, I've been banging my head on this a couple of days. At times the system would start working, only to stop again. At other times it might work for a couple of hours, then be dead for a full day or more. But in spite of all the intermittency, I decided there was no obvious hardware problem, after extensive testing.
Since my automation system has been effectively dead, I haven't paid a lot of attention to X10, but after a while I gradually noticed that there seemed to be an awful lot of traffic on the X10 side of things, even with the Ocelot shut down. Investigating, I find I have a wireless motion sensor on address H13 that seems to be streaming continuous H13/H-ON messages.
But here's the catch. When I disconnected the TW523 interface from the wall, suddenly the timers started working again.
It may be a bit premature, but it appears that excessive X10 traffic keeps timers from advancing.
IS THIS POSSIBLE?
Now I have to go climb a tree and fix/replace that darn motion sensor and retest everything. But my first hypothesis is that (a) Timers are driven off a maskable Z80 interrupt and (b) while receiving X10 traffic, that interrupt is masked leading to (c) Excessive X10 traffic can effectively halt all the timers.
Crazy, huh....
I would be inclined to count this as a flaw in the CMAX executive. As it stands, if further testing bears out the theory, excessive X10 traffic can in fact halt functions that otherwise have absolutely nothing to do with X10.
I.E. I have several chunks of code that react to a Secu input, and then after a timer operates a bit, closes or opens Secu outputs. e.g. A vehicle sensor at my front gate closes a relay that is seen as a contact closure by a Secu, which after some logic, evaluates to a secu relay closure that opens the gate.
No X10 involved in the loop at all. Yet, with the TW523 interface seeing this H13 traffic, the gate won't open, because the timers don't run. When I unplugged the wall interface, now the gate operates as I designed it.
If further study confirms this diagnosis, I am now faced with the task of adapting all my logic to reduce or eliminate dependence on timers, as it appears that anything that depends on a timer can fail unexpectedly should something go awry with an X10 device.
Thoughts anyone?
Thanks,
Nathan
Addendum: After writing the above, I climbed the tree and retrieved the motion sensor. I noticed immediately that it had a LED hard on. I removed it's batteries, and reinstalled the TW523 and all my Ocelot programming is back to normal, timers are running, events are being responded to and so on. This seems pretty conclusive to me.
Odd as it seems, excessive X10 traffic from a failed transmitter can indeed stop Ocelot timers from advancing, and thereby cause logic that has nothing whatever to do with X10 to fail.
Edited by Nathan Gregory (11/24/08 10:31 PM)
|
|
Top
|
|
|
|
|
#21685 - 03/06/10 09:57 AM
Re: Timer Troubles...
[Re: Nathan Gregory]
|
RRA
newbie
Registered: 03/05/10
Posts: 8
|
I have almost the same trouble except that timers begin to work after setting controller clock to PC clock. As I noticed, timers along with the system real time clock usually stops after reconnecting power to the controller, the system clock stops just at the moment, when I connect the power plug. Any ideas?
|
|
Top
|
|
|
|
|
#21688 - 03/09/10 07:06 AM
Re: Timer Troubles...
[Re: Guy Lavoie]
|
Brave Sir Robbin
newbie
Registered: 01/09/03
Posts: 24
Loc: Las Vegas, Nevada
|
This is an interesting thread. I use timers a lot and noticed that one timer would never work right. It's a timer that gets triggered with my car monitor, which does send an X-10 signal every couple of seconds.
I could never figure out why this one timer would not work correctly. I'll have to take another look at my CMax code.
|
|
Top
|
|
|
|
|
#21689 - 03/10/10 10:23 AM
Re: Timer Troubles...
[Re: Brave Sir Robbin]
|
RRA
newbie
Registered: 03/05/10
Posts: 8
|
after a few day experiments 2 different ocelots were tested with the same problem. I tried to use different executives, from v3.06 through v3.18 with their native c-max software. No result. The ocelot RTC stops exactly at the moment of power reconnection. It seems like the RTC chip Epson 72421 not switching out of standby mode. Another thing I noticed, that more than a couple of minutes must pass before reconnecting power. Doing this quikly not causes the RTC to stop. The battery is new, RTC stops exactly at the time of power reconnecting even when the system have been unplugged for more than a day. Is this really not the bug but the feature? Two different ocelots with different date of manufacturing? If so, I will be forced to refuse using Adicon controllers at all!
|
|
Top
|
|
|
|
|
#21690 - 03/10/10 11:17 AM
Re: Timer Troubles...
[Re: RRA]
|
Barry Parr
journeyman
Registered: 02/15/04
Posts: 98
Loc: NZ
|
Hi. What is the 'No Load' Measured voltage of your power supply?
|
|
Top
|
|
|
|
|
#21691 - 03/10/10 11:33 AM
Re: Timer Troubles...
[Re: Barry Parr]
|
RRA
newbie
Registered: 03/05/10
Posts: 8
|
12.62 volts 2 A switching power supply. I've try 3 different ones already along with AC and just 12V battery
|
|
Top
|
|
|
|
|
#21692 - 03/10/10 11:34 AM
Re: Timer Troubles...
[Re: RRA]
|
RRA
newbie
Registered: 03/05/10
Posts: 8
|
12.62 - without load... 12.55 under the load
|
|
Top
|
|
|
|
|
#21693 - 03/10/10 12:17 PM
Re: Timer Troubles...
[Re: RRA]
|
Barry Parr
journeyman
Registered: 02/15/04
Posts: 98
Loc: NZ
|
OK. That screws My spike thought.Sorry I don't have any other ideas as I have never encountered this problem.
|
|
Top
|
|
|
|
Moderator: Dan Smith, Monte G, ADI Tech Support, Guy Lavoie
|
2747 Members
19 Forums
3837 Topics
22713 Posts
Max Online: 67 @ 08/16/11 03:08 PM
|
|
|
|
|
|
1
|
2
|
3
|
4
|
|
5
|
6
|
7
|
8
|
9
|
10
|
11
|
|
12
|
13
|
14
|
15
|
16
|
17
|
18
|
|
19
|
20
|
21
|
22
|
23
|
24
|
25
|
|
26
|
27
|
28
|
29
|
|
|
|
|
|
