Digital Potentiometer ?
MrScuito
Posts: 6
I'm very new to BS2 programming I just got the BS2 homework board today to practice the circuit and code on. Anyhow, this is what I'm trying to do. I'm planning on building a LED fixture for my reef tank in the next couple of months and I would like to include a digital pot to simulate dawn to dusk cycle. Once I can (if I can) get it to work on the HWB I will scale it up and make a proper board.
Goal:
Lights begin to come on 8am slowly becoming brighter till 11am hold there till 1pm then slowly dim till off at 8pm then repeat next day and so on
What I have done:
Built digi pot on page 268 of Whats a Microcontroller ran code on that page including LOOP at the end of the code. but it cycles only once and stops. I've played with the code a bit and tried TOGGLE but still no continues cycle.
Is it possible to use the BS2 to accomplish my goal. Great with hardware never been good with software.
can anybody point me to a sample code that would give me a continueous cycle I can prob monkey with timing to get it to take 12hrs to complete cycle if I can get it too cycle.
Goal:
Lights begin to come on 8am slowly becoming brighter till 11am hold there till 1pm then slowly dim till off at 8pm then repeat next day and so on
What I have done:
Built digi pot on page 268 of Whats a Microcontroller ran code on that page including LOOP at the end of the code. but it cycles only once and stops. I've played with the code a bit and tried TOGGLE but still no continues cycle.
Is it possible to use the BS2 to accomplish my goal. Great with hardware never been good with software.
can anybody point me to a sample code that would give me a continueous cycle I can prob monkey with timing to get it to take 12hrs to complete cycle if I can get it too cycle.
Comments
The code that is shown in the Homework board book is basic test code to show how the digital potentiometer works. What you also need to remember is the digital potentiometer is designed to go so many steps in one direction or another. I can give you some code if no one does it first.
So if you want to control lighting, that device may not directly be able to handle the power - you might have to put a transistor or some other power amplifying device in between. And if you have to use a transistor, or mosfet; it might just be easier to let the Stamp create a Pulse Width Modulated signal to do all the work.
Much depends on how many lights you want to use. The potentiometer might work well with just one; but burn up with many.
I was able to pick up an unopened BS2 homework board kit at a yard sale for a buck, thats why I'm using it for my test bed
I assumed that once the idea was worked out on the HWB and running one 1w led thru the cycle wanted that the idea could then be ramped up on another more permenent board. Ultimately I want to dimm 24 3w HP LEDs
SCORE! Who says a buck doesn't buy much any more?
http://www.parallax.com/Portals/0/Downloads/docs/prod/edu/28123-WAM-v3.0.pdf
on the Parallax site for free.
1) Set the tap setting from the computer keyboard. Make sure to press enter key after you entered a new tap setting value.
2) Show the tap setting in the debug terminal for what you entered.
' -----[ Title ]----------------------------------------------------------- ' What's a Microcontroller - TerminalControlledDigitalPot.bs2 ' Update digital pot's tap based on Debug Terminal user input. ' {$STAMP BS2} ' {$PBASIC 2.5} ' -----[ EEPROM Data ]----------------------------------------------------- ' -----[ I/O Definitions ]------------------------------------------------- UdPin PIN 5 ' Set values of I/O pins ClkPin PIN 6 ' connected to CLK and U/D ' -----[ Constants ]------------------------------------------------------- DelayPulses CON 10 ' Delay to observe LED fade. DelayReader CON 2000 ' -----[ Variables ]------------------------------------------------------- counter VAR Byte ' Counter for FOR...NEXT. oldTapSetting VAR Byte ' Previous tap setting. newTapSetting VAR Byte ' New tap setting. ' -----[ Initialization ]-------------------------------------------------- oldTapSetting = 0 ' Initialize new and old newTapSetting = 0 ' tap settings to zero. LOW UdPin ' Set U/D pin for Down. FOR counter = 0 TO 127 ' Set tap to lowest position. PULSOUT 6,5 PAUSE 1 NEXT PAUSE 1000 ' Wait 1 s before 1st message ' -----[ Main Routine ]---------------------------------------------------- DO GOSUB Get_New_Tap_Setting ' User display and get input. GOSUB Set_Ud_Pin ' Set U/D pin for up/down. GOSUB Pulse_Clk_pin ' Deliver pulses. LOOP ' -----[ Subroutines ]----------------------------------------------------- Get_New_Tap_Setting: ' Display instructions and ' get user input for new DEBUG CLS, "Tap setting is: ", ' tap setting value. DEC newTapSetting, CR, CR DEBUG "Enter new tap", CR, "setting (0 TO 127): " DEBUGIN DEC newTapSetting RETURN Set_Ud_Pin: ' Examine new and old tap values ' to decide value of U/D pin. IF newTapSetting > oldTapSetting THEN ' Notify user if values are HIGH UdPin ' equal. oldTapSetting = oldTapSetting + 1 ' Increment for Pulse_Clk_pin. ELSEIF newTapSetting < oldTapSetting THEN LOW UdPin oldTapSetting = oldTapSetting - 1 ' Decrement for Pulse_Clk_pin. ELSE DEBUG CR, "New and old settings", CR, "are the same, try ", CR, "again...", CR PAUSE DelayReader ' Give reader time to view ENDIF ' Message. RETURN Pulse_Clk_pin: ' Deliver pulses from old to new values. Keep in mind that Set_Ud_Pin ' adjusted the value of oldTapSetting toward newTapSetting by one. ' This keeps the FOR...NEXT loop from executing one too many times. FOR counter = oldTapSetting TO newTapSetting PULSOUT ClkPin, 1 PAUSE DelayPulses NEXT oldTapSetting = newTapSetting ' Keep track of new and old ' tapSetting values. RETURNI also suggest you build the circuit that is shown in the Whats a Microcontroller? book (page 267) and PDF (Page 294) and circuit will be the bases for your design. As you finalize your design, I would suggest you use TIP120 transistor which can handle the LED load that you are planning on.I will be working on final program code for you and hope to have it finished no later than Saturday morning.
There are several ways to achieve what you want to do, but something you will want to consider first: What kind of LED drivers are you going to use, and what are their dimming circuit requirements. (EDIT: I see you posted the kit you plan on buying, but i don't see the specs for what the dimming requirement is exactly, all it says is to use a pot...There's more info needed to know what to do- After reading some posts from some other people about this kit, it appears the kit comes with 22ohm pots for dimming, and several users indicate that these drivers are NOT microcontroller compatible.... I would suggest maybe thinking about using the meanwells...)
Some drivers need a 0-10v analog signal, and others need a 5v Pulse Width Modulation (PWM) signal. Depending on which type of driver you use, you may or may not need the digital pot at all.
What I found as i looked at the specs on the led drivers dimming circuit for the drivers I was going to use is that they appeared that they would need more current than just about any of the digital pots I could find could handle. (needed ~2ma, and it was hard to find a digital pot that can handle this amount of current)
One other thing to note is that LED's will not dim to 0% on any driver that I am aware of. Usually, all you can get is down to ~10%-15% as the minimum brightness before the LEDs cut out. And on my tank, that level of brightness is actually fairly bright even on the low end. I wish there was more control, but my particular build is made up of 24 3watt Cree's driven by three Thomas Research Drivers.(two blue 700ma drivers, and one 1050ma driver for the whites) Thomas Research Drivers need 0-10v analog. (same as Meanwells, and a few others)
So what i settled on initially was to use 3 digital pots + 3 741 op-amps configured with a gain of 2 to be the digital pots buffer on the output, to allow me to get 0-10v out of a 0-5v digital pot signal. This however required me to have 2 different power supplies- a 5v + a +12v (as there is some loss on the op amp from what I could see when prototyping the circuit) + other misc items (resistors, cap's, wires, etc) to make this happen, The chip count really started to increase as i needed to do 3 of these for my build.
A little while later, as I researched my build, I bumped into a little talked about item - the LTC1257.
This little beauty had several things going for it:
#1 it can go up to ~15v.
#2 it can source up to 2ma of current.
#3 it is a 12bit ADC which gives 4095 steps of resolution on the output.
#4 you can daisy chain them together.
The price however, is a bit steep: ~$8.00 each. ouch. I defrayed this somewhat by asking for 2 samples... and buying one from digikey.
However, this was considered reasonable for me because of the other circuit parts reduction + lack of complexity in implementing the circuit design using the basic stamp made for an easy trade off in my mind.
The kit you are looking at doesn't seem too shabby overall, but the drivers that you were indicating you were going to use are of unknown origin, and also have no data sheet.
Which is why I would consider simply altering the LED driver choice, to get something (such as the meanwell D or P drivers) that is more supported out in the communities such as reef central, or nano reef. Meanwell, Buckpucks, Inventronics seem popular. It appears that the Inventronics are very similar to teh Thomas Research drivers in that they have a self supplied dimmer voltage output
Note: the meanwell D drivers need a 0-10v analog signal, and the P drivers are PWM if i remember correctly. The problem with the meanwells is that they need to be adjusted out of the gate for current and also need an external source for the dimmer voltage. Thats why I went with the Thomas Research Drivers when nanotuners was still around. less fuss and no muss. I like to follow the KISS principle even if it costs a couple bucks more overall.
There are several ways to get PWM if needed based on the driver choice that you make- but the Stamp is not really one of the best ways (but it can be done), I would consider using external circuits to provide a constant PWM signal, as the stamp doesn't really have a way to constantly output a nice PWM signal all the time like what you will need if you went with a PWM driver. And use the stamp to control the PWM generator if you go that route.
' ========================================================================= ' ' File......timer_controlled_led_dimmer.bs2 ' Purpose... ' Author.... ' E-mail.... ' Started... ' Updated... ' ' {$STAMP BS2} ' {$PBASIC 2.5} ' ' ========================================================================= ' -----[ Program Description ]--------------------------------------------- 'Lights begin to come on 8am slowly becoming brighter till 11am hold there 'till 1pm then slowly dim till off at 8pm then repeat next day and so on ' -----[ Revision History ]------------------------------------------------ ' -----[ I/O Definitions ]------------------------------------------------- UdPin PIN 5 ' Set values of I/O pins ClkPin PIN 6 ' connected to CLK and U/D ' -----[ Constants ]------------------------------------------------------- IsOn CON 1 ' for active-high in/out IsOff CON 0 DelayPulses CON 10 ' Delay to observe LED fade. DelayReader CON 2000 ' Pause delay setting ' -----[ Variables ]------------------------------------------------------- hours VAR Byte ' Stores hours minutes VAR Byte ' Stores minutes seconds VAR Byte ' Stores seconds counter VAR Byte ' Counter for FOR...NEXT. oldTapSetting VAR Byte ' Previous tap setting. newTapSetting VAR Byte ' New tap setting. ' -----[ Initialization ]-------------------------------------------------- Reset: oldTapSetting = 0 ' Initialize new and old newTapSetting = 0 ' tap settings to zero. LOW UdPin ' Set U/D pin for Down. FOR counter = 0 TO 127 ' Set tap to lowest position. PULSOUT 6,5 PAUSE 1 NEXT ' -----[ Program Code ]---------------------------------------------------- Main: DO ' Calculate hours, minutes, seconds IF seconds = 60 THEN seconds = 0: minutes = minutes + 1 IF minutes = 60 THEN minutes = 0: hours = hours + 1 IF hours = 24 THEN hours = 0 IF hours = 7 AND seconds = 59 THEN newTapSetting = 1 'starting tap setting IF hours = 8 AND seconds = 0 THEN GOSUB Set_Dimming_Direction IF hours = 9 AND seconds = 0 THEN GOSUB Set_Dimming_Direction IF hours = 10 AND seconds = 0 THEN GOSUB Set_Dimming_Direction IF hours = 11 AND seconds = 0 THEN GOSUB Set_Dimming_Direction: IF hours = 11 AND seconds = 1 THEN oldTapSetting = 99 IF hours = 13 AND seconds = 0 THEN GOSUB Set_Dimming_Direction IF hours = 14 AND seconds = 0 THEN GOSUB Set_Dimming_Direction IF hours = 15 AND seconds = 0 THEN GOSUB Set_Dimming_Direction IF hours = 16 AND seconds = 0 THEN GOSUB Set_Dimming_Direction IF hours = 17 AND seconds = 0 THEN GOSUB Set_Dimming_Direction IF hours = 18 AND seconds = 0 THEN GOSUB Set_Dimming_Direction IF hours = 19 AND seconds = 0 THEN GOSUB Set_Dimming_Direction IF hours = 20 AND seconds = 0 THEN GOSUB Set_Dimming_Direction PAUSE 991 ' Pause + overhead ~ 1 second seconds = seconds + 1 ' Increment second counter LOOP ' -----[ Subroutines ]----------------------------------------------------- Set_Dimming_Direction: ' Examine new and old tap values ' to decide value of U/D pin. IF newTapSetting > oldTapSetting THEN ' Notify user if values are HIGH UdPin ' equal. oldTapSetting = oldTapSetting + 25 ' Increment for Pulse_Clk_pin. ELSEIF newTapSetting < oldTapSetting THEN LOW UdPin oldTapSetting = oldTapSetting - 15 ' Decrement for Pulse_Clk_pin. ENDIF Pulse_Clk_pin: ' Deliver pulses from old to new values. Keep in mind that Set_Ud_Pin ' adjusted the value of oldTapSetting toward newTapSetting by one. ' This keeps the FOR...NEXT loop from executing one too many times. FOR counter = oldTapSetting TO newTapSetting PULSOUT ClkPin, 1 PAUSE DelayPulses NEXT oldTapSetting = newTapSetting ' Keep track of new and old ' tapSetting values. RETURN ' -----[ User Data ]-------------------------------------------------------newTapSetting = 0
but not
newTapSetting = 1
and code:
IF hours = 7 AND seconds = 59 THEN newTapSetting = 1
is already present so, I'm not sure where to make the change. Sorry I'm only on day 3 of my self teaching lesson in basic. Thanks again for taking the time to help me out.