Christmas Countdown Clock - DIY Direct Control of Hardware. Read Only. No more posting. - PlanetChristmas! Forums. Read Only. We've moved to http://talk.planetchristmas.com - The Forums of PlanetChristmas have moved to http://talk.PlanetChristmas.com
pshort wrote: The circumstances are different. The discussion here is about a stand-alone controller, not connected to anything else. The earlier discussion was about computer control of lights, which involves connecting lots of controllers and other circuits together. The lack of isolation in that case would have had the possibility of ground loops (with potentially large circulating currents) and also would have resulted in AC potentials on the surfaces and controls of the computer (quite unsafe).
Care must be taken even in this case to prevent safety problems, but it can be done with relatively simple measures that can be easily described at the right time and are not too likely to fail.
--
Phil
BTW, I'm not sure what your emoticon means. If it was meant to be humorous you can ignore this message.
Ok, I can see your point about this being a stand alone controller, but it might be desirable to have it connected in such a way that it could be turned off if for example it was part of a display with a light show.
As for this it is the "shocked" face. I was shocked to see the resistor idea after our previous discussions about the optocoupler method.
Please go ahead and describe the relatively simple methods that could be used to provide isolation. There's always potential for new and improved designs.
TED
Last edited on Thursday January 18th, 2007 07:53 pm by TED
ErnieHorning wrote: TED wrote: ...you guys jumped all over me because it would not be isolated. Now you are saying just use a resistor?
It was a gentle persuasion so that you would see it our way.
In this case you can connect a resistor to the output of a bridge rectifier off the transformer. This will give you the 120 pulses per second. The exact timing doesn't matter, just that you get them. The transformer provides the isolation though its possible to do this without a transformer too.
Ah, but I expect you to see it MY way! I don't like the opto method due to its inaccuracy especially with the particular optocoupler that is used. You may be interested to know that Dan once posted that the LOR controllers use a single resistor to detect zero crossing. I think it is connected in the way that you describe with the bridge rectifier. This would have been with the original LOR boards. I do not know if it is the case with the newest designs.
The measures that I was mentioning were for safety in general, not for isolation. Make sure that it's completely enclosed in plastic without any exposed metal that might be hot, no ground loops, stuff like that.
If it is going to be connected, then you need some sort of isolation from the AC line, such as opto-isolators, transformers, or perhaps capacitive coupling (not so sure about the last one).
On the other topic, I don't think that the lack of precision in that H11AA1 is very important. Nothing in the dimmer designs is calibrated in any way, everything is pretty much relative. Any small variations from one dimmer circuit to the next that could be blamed on the H11AA1 is just not going to be visible to the average user.
Why do you think that a more precise circuit than the H11AA1 is necessary?
I just completed my Christmas Countdown this past year and had it up from Nov 1st to Dec 26th. Some of you helped give me input on it and I thank you for the help.
This was V2.0, V1.0 lasted ~16 hours before the magic smoke came out and destroyed it in 2005. I used strings of the mini lights cut into 10 light segments which run on 24Vac. I made the SSR's to to do the switching and had a zero cross detect for the timing (V1.0 had a watch crystal and temp seemed to affect it too much). The zero cross seems to drift +/- a hair, but it pretty much dead-on over the life of the thing.
You tell it the current day/time and whether you are counting down to Christmas or some other day/time. If you are counting down to some other day/time it displays 0s forever when it is over. If you are counting to Christmas, when it is done it displays zeros for 5 seconds, scrolls "HO HO HO" for 5 seconds, zeroes, "HAPPY HOLIDAYS", zeroes, "FELIZ NAVIDAD", zeroes, "YOULL SHOOT YOUR EYE OUT", zeroes, "FRA-GEE-LAY", and repeats until you reset it or turn it off. The A Christmas Story movie quotes are in there because I have a leg lamp that goes in the front window near by.
The project worked and I am very happy with it, but I think there are some things I would do differently if I did it again. But I am happy to talk/share info about it with anyone who has an interest.
When I get a little more free time (I just bought my first house), I am going to make a better website with explanation, code to download for it, parts list, more pics, things to do differently, and video of it running.
~J
____________________ You used up all the glue ON PURPSE! ~A.C.S.
Thanks for the schematic! I wonder if I have to build one....
Please send more info!
Thanks!
____________________ Kevin Thomas
"When you believe a thing, believe in it all the way" - Walt Disney
99,999+ lights - 500+ channels - FM100B http://www.ChristmasInKent.com
Let your light so shine before others...
Can you post pictures of the back of the board? Are the letters made of wood and in a pair. Would it be hard to add an extra day so it read 364 : 23: 59: 59? I see the SSR's that you made it the top part of your board a timer or is it a controller? Does your clock run on a computer?
What I would have done different is the assembly of the countdown. Right now it is plywood that I drilled out and primed and painted (priming the individual holes was a PAIN!). Then I put the individual lights into the holes and hot glued them into place. If I could do it over (and I probably will do the digits over again in a few years) I would make the digits out of welded 1/4" steel for ease of bulb replacement (and lighter storage).
The digits are made in pairs and all are interchangeable.
The whole unit is standalone, no computer needed. The clock is run off of the H11A1 which takes its input from the 120Vac outlet and pulses everytime the sine wave croses zero (60Hz = 120 times a second). I am using the TIMER1 option on the uC to overflow every 120 pulses.
Adding a digit wouldn't be a problem at all, just would require 7 more SSRs and another digit, it would be doable.
~Jason
____________________ You used up all the glue ON PURPSE! ~A.C.S.
pshort wrote: If it is going to be connected, then you need some sort of isolation from the AC line, such as opto-isolators, transformers, or perhaps capacitive coupling (not so sure about the last one).
On the other topic, I don't think that the lack of precision in that H11AA1 is very important. Nothing in the dimmer designs is calibrated in any way, everything is pretty much relative. Any small variations from one dimmer circuit to the next that could be blamed on the H11AA1 is just not going to be visible to the average user.
Why do you think that a more precise circuit than the H11AA1 is necessary?
I think the resistors you would use if connecting the AC to the comparator are 10,000 Mega Ohm. Would that not give a pretty good degree of safety?
I understand that you don't think the lack of precision is very is very important. But even with that same design why not at least choose an opto that is better suited to the purpose--one with a wave output rather than a square wave output? I guess the bottom line for me is that I just don't like it. In response to the the question of why do I think a more precise circuit is necessary I ask why not use the best design available when the cost is about the same? I'm not saying that it's absolutely "necessary". But why not go with the better choice?
It’s like saying that you have a light switch that can guarantee that power is always applied when the voltage is zero. The end result is that my switch and your switch will both turn on the light.
Please note: I am not jumping all over you at this point, just gentle poking.
____________________ No really... Just one more light and I'll be finished....
Disclaimer: If you take any of what I say as non-constructive, it's not how I intended it.
ErnieHorning wrote: It’s like saying that you have a light switch that can guarantee that power is always applied when the voltage is zero. The end result is that my switch and your switch will both turn on the light.
Yeah but when you are talking about dimming the exact time of the light turning on is critical so why not be as accurate as possible?
TED
Last edited on Saturday January 27th, 2007 09:50 am by TED
TED wrote: Yeah but when you are talking about dimming the exact time of the light turning on is critical so why not be as accurate as possible?
For dimming, yes, though it's not nearly as critical as you would think.
In this thread, we just need to count the cycles. We aren't concerned with exactly when they happened, just that they did. You're only displaying a minimum of one second intervals. So whether the number changes at exactly when the AC wave is at zero or a bit later isn't going to matter when a second goes by.
If I asked to count out loud from 1 to 10 in one second intervals. The fact that you didn't say every number on exactly a 1 second boundary doesn't change the fact that 10 seconds still occurred.
____________________ No really... Just one more light and I'll be finished....
Disclaimer: If you take any of what I say as non-constructive, it's not how I intended it.
I've been batting around a countdown timer project for a while now.
It started out as a school project I was teaching, the teaching contract was up before I completed it. Then I joined this addiction last summer which brought it back to life.
I dont normally post here but read quite a lot so I thought I'd post here to see if anyone has built such a unit or if new perspectives may aid in this projects completion. (I normally haunt another forum)
A little background on what I am doing.
DDD:HH:MM countdown clock with : as seconds.
Once programmed, its independant of computer input, battery backup circuit will be added once its running.
The segments of the 7 segment digit are comprised of 5 led's per segment.
The TPIC6B595 ic is in essence a 595/2003ULN combo chip to handle the higher mA and Volts needed.
The PIC uC has slowly migrated through several different models and is still open for substitution, current the circuit shows a 16F627A as its the pic I am currently using to learn pic programming with.
The circuit appears sound.
The code is my stumbling block, I just dont know enough about PIC coding yet to code this project to life.
PShort was kind enough to offer up a small code I am trying to work with at the moment but I still dont understand all of what it does so troubleshooting the code to make it work on the circuit is a crapshoot for me.
Right now I await my EasyPic5 dev board and a PicKit2 programmer to replace my JDM and WinPic800.
Has this been already built, I'd rather not re-invent the wheel. If not, anyone that might have PIC coding experience and wants to create a test routine to latch through each segment of each digit, it would allow me to push forward and complete this project. The PIC inthe schem is NOT necessarily the pic that will be used. I dont believe the SPI will work well with this particular pic and will try any pic that will get the job done.
I'm currently learning pic Basic, but I understand some of the asm commands as well.
Cheers
Steve
Last edited on Saturday February 23rd, 2008 11:48 pm by Aurbo
Wow .... kinda getting out of hand here . Poor Brian has asked a few times about certain kits , and if they would work , yet , seems no one wants to answer .
I , myself , do not have the know how , otherwise , I would answer .
I , too , am in the SAME postion as , obviously Brian is ... neither of us has much programming background , or designing electronics abilities and would REALLY like to know ,if there is some sort of a kit that folks like us , can simply purchase/build and either adapt , or use as is ?
Last edited on Tuesday February 26th, 2008 02:24 am by TERBObob
I'm planning a countdown clock this year as well. The plan is to use a PICAXE and real time clock to control a bunch of 595s and individual ssrs for each segment. If anyone's interested I can post board layouts and the picaxe code. Will be using the olsen 595 board from the coop buy going on at DIY Christmas and a home etched board for the picaxe and rtc. 595 board will be out at the clock and picaxe board inside. This method requires only a single cat5 run out to the clock (and power of course).
EDIT: While I like the idea of using zero cross from the AC utility to keep time, I'd prefer to stick with the real time clock. I've noticed the frequency vary slightly over the course of the day. Not to mention it's not always 60Hz. But that's just my opinion. It probably wouldn't affect the timing significantly.
Last edited on Tuesday February 26th, 2008 03:42 am by Mike Montgomery
My countdown clock usues its own PC (an OLD one in this case) with a NI DIO-96 card running relays.
Each light segment is 50 mini lights in coro (still being constructed).
Home built VB6 code runs it(already coded and tested inside)
Pictures once the coro numbes are built.
The whole sign will be about 3ft high by 10 feet long.
It will have weeks days hours minutes and seconds until christmas.
it is the "shocked" face. I was shocked to see the resistor idea after our previous discussions about the optocoupler method.
I don't like the opto method due to its inaccuracy especially with the particular optocoupler that is used. You may be interested to know that Dan once posted that the LOR controllers use a single resistor to detect zero crossing. I think it is connected in the way that you describe with the bridge rectifier. This would have been with the original LOR boards. I do not know if it is the case with the newest designs.
