Post by thread:[EE] Driving Remote LEDs

Hi all. I have a bit of a design challenge I'm grappling with. Here's the deal. I have a friend who wants to make some remote indicator lights. In the beginning they will be controlled by switches, but I will be building a PIC based system in the future. We've talked about how to deal with long cable runs in the past, but this application is slightly unique in that the indicators are bi-colour LEDs. These are the ones where they are two diodes wired back to back. At the moment I'd like to avoid having to put in a remote PIC mainly due to the tight delivery schedule he needs. So how best to provide protection when the polarity of the line will be changing? Let's guess that the line length will be between 50 and 400 feet of cat5 cable. Also, I haven't done the calculations yet for voltage drop. If anyone has done this in the past, tips on good voltage to use (to deal with line drop) would be appreciated. In the future I will be moving to a more solid state design and I'm thinking about using an open collector or similar driver with a slightly higher voltage to drive the long line. Thanks! Josh -- A common mistake that people make when trying to design something completely foolproof is to underestimate the ingenuity of complete fools. -Douglas Adams

> So how > best to provide protection when the polarity of the line will be > changing? Let's guess that the line length will be between 50 and 400 > feet of cat5 cable. A 3v9 zener diode + a small series resistor? > Also, I haven't done the calculations yet for voltage drop. don't, use a constant-current drive. Wouter van Ooijen -- ------------------------------------------- Van Ooijen Technische Informatica: http://www.voti.nl consultancy, development, PICmicro products docent Hogeschool van Utrecht: http://www.voti.nl/hvu

Think current source/sink with a good voltage compliance range, then line length is a non-issue. Josh Koffman wrote: {Quote hidden}>Hi all. I have a bit of a design challenge I'm grappling with. Here's >the deal. I have a friend who wants to make some remote indicator >lights. In the beginning they will be controlled by switches, but I >will be building a PIC based system in the future. > >We've talked about how to deal with long cable runs in the past, but >this application is slightly unique in that the indicators are >bi-colour LEDs. These are the ones where they are two diodes wired >back to back. At the moment I'd like to avoid having to put in a >remote PIC mainly due to the tight delivery schedule he needs. So how >best to provide protection when the polarity of the line will be >changing? Let's guess that the line length will be between 50 and 400 >feet of cat5 cable. > >Also, I haven't done the calculations yet for voltage drop. If anyone >has done this in the past, tips on good voltage to use (to deal with >line drop) would be appreciated. In the future I will be moving to a >more solid state design and I'm thinking about using an open collector >or similar driver with a slightly higher voltage to drive the long >line. > >Thanks! > >Josh > > -- * | __O Thomas C. Sefranek spam_OUTWA1RHPTakeThisOuTARRL.net |_-\<,_ Amateur Radio Operator: WA1RHP (*)/ (*) Bicycle mobile on 145.41, 448.625 MHz hamradio.cmcorp.com/inventory/Inventory.html http://www.harvardrepeater.org

At 04:31 AM 5/26/2006 -0400, you wrote: Hi, Josh:- {Quote hidden}>Hi all. I have a bit of a design challenge I'm grappling with. Here's >the deal. I have a friend who wants to make some remote indicator >lights. In the beginning they will be controlled by switches, but I >will be building a PIC based system in the future. > >We've talked about how to deal with long cable runs in the past, but >this application is slightly unique in that the indicators are >bi-colour LEDs. These are the ones where they are two diodes wired >back to back. At the moment I'd like to avoid having to put in a >remote PIC mainly due to the tight delivery schedule he needs. So how >best to provide protection when the polarity of the line will be >changing? Let's guess that the line length will be between 50 and 400 >feet of cat5 cable. A 1000' pair of AWG24 wires has a loop resistance of about 51 ohms at 20°C. If you're going to be using, say, 470R series resistance, the voltage drop for 50-400' is hardly significant as far as visual brightness goes. >Also, I haven't done the calculations yet for voltage drop. If anyone >has done this in the past, tips on good voltage to use (to deal with >line drop) would be appreciated. In the future I will be moving to a >more solid state design and I'm thinking about using an open collector >or similar driver with a slightly higher voltage to drive the long >line. Maybe you could use an RS485 driver such as the SN65HVD20. Or if you feel lucky, just use a couple of PIC pins and split the total resistance into four, with a couple of unipolar TVS's to ground in a T. Or just a single resistor if switches are doing the driving. >Best regards, Spehro Pefhany --"it's the network..." "The Journey is the reward" .....speffKILLspam@spam@interlog.com Info for manufacturers: http://www.trexon.com Embedded software/hardware/analog Info for designers: http://www.speff.com ->>Test equipment, parts OLED displys http://search.ebay.com/_W0QQsassZspeff

The ideal driver is a RS422 driver. These are protected from ESD and can deliver a lot of current from a long way. I have personally used them at 1000m. They are bidirectional so will work well for the dual-led. LEDs are current devices. As long as the cables are protected from transients, the scheme will work reliably. --Bob Josh Koffman wrote: {Quote hidden}> Hi all. I have a bit of a design challenge I'm grappling with. Here's > the deal. I have a friend who wants to make some remote indicator > lights. In the beginning they will be controlled by switches, but I > will be building a PIC based system in the future. > > We've talked about how to deal with long cable runs in the past, but > this application is slightly unique in that the indicators are > bi-colour LEDs. These are the ones where they are two diodes wired > back to back. At the moment I'd like to avoid having to put in a > remote PIC mainly due to the tight delivery schedule he needs. So how > best to provide protection when the polarity of the line will be > changing? Let's guess that the line length will be between 50 and 400 > feet of cat5 cable. > > Also, I haven't done the calculations yet for voltage drop. If anyone > has done this in the past, tips on good voltage to use (to deal with > line drop) would be appreciated. In the future I will be moving to a > more solid state design and I'm thinking about using an open collector > or similar driver with a slightly higher voltage to drive the long > line. > > Thanks! > > Josh >

On 5/26/06, Wouter van Ooijen <wouterKILLspamvoti.nl> wrote: > > So how > > best to provide protection when the polarity of the line will be > > changing? Let's guess that the line length will be between 50 and 400 > > feet of cat5 cable. > > A 3v9 zener diode + a small series resistor? So the zener would drop down my higher voltage to 3.9V, and the resistor would act as a current limiter, both on the remote end. That way I get to use a higher voltage on the sending unit side, correct? Josh -- A common mistake that people make when trying to design something completely foolproof is to underestimate the ingenuity of complete fools. -Douglas Adams

On 5/26/06, Tom Sefranek <.....tcsKILLspam.....cmcorp.com> wrote: > Think current source/sink with a good voltage compliance range, > then line length is a non-issue. right...as a non engineer, I don't fully understand this. Can you perhaps give me a couple more terms to make my Googling a bit more helpful? And does this go against what the others have suggested, or is there a difference between a current source/sink and a constant current source. Thanks! Josh -- A common mistake that people make when trying to design something completely foolproof is to underestimate the ingenuity of complete fools. -Douglas Adams

On 5/26/06, Spehro Pefhany <EraseMEspeffspam_OUTTakeThisOuTinterlog.com> wrote: > A 1000' pair of AWG24 wires has a loop resistance of about 51 ohms at 20°C. > If you're going to be using, say, 470R series resistance, the voltage > drop for 50-400' is hardly significant as far as visual brightness goes. So in theory for my simple version I can basically just ignore the cable. Everything is going to be very simple in that it will just be switches and LEDs - there shouldn't be much to get harmed by picking up interference. Right? Once I move to the more complex version with actual electronics driving the long lines, I will have to revisit that protection again. The RS485 driver idea is kind of cool...in theory it would be able to control the LEDs directly. I'll have to see what the output of the drivers is to see if it has enough oomph to light a bright LED. The advantages are many - I can drive the driver easily from a PIC and if there is some sort of horrible lightning strike or whatever, the chips are cheap and can be sacrificial (assuming I use 75176's...I have over 200 just sitting around). Thanks! Josh -- A common mistake that people make when trying to design something completely foolproof is to underestimate the ingenuity of complete fools. -Douglas Adams

>> A 3v9 zener diode + a small series resistor? > > So the zener would drop down my higher voltage to 3.9V, and the > resistor would act as a current limiter, both on the remote end. That > way I get to use a higher voltage on the sending unit side, correct? No, at least I don't think you think the way I think. Put the small resistor in series with LED, zener diode in parallel with this combo. Zener shorts a reverse voltage, and limits (by drawing currunt, hence energy) the 'normal polarity' voltage. The resistor limits the (peak) current when the zener is limiting the voltage. In normal use the zener is not active, and the resistor adds a small voltage drop (assuming a constant current driver). Wouter van Ooijen -- ------------------------------------------- Van Ooijen Technische Informatica: http://www.voti.nl consultancy, development, PICmicro products docent Hogeschool van Utrecht: http://www.voti.nl/hvu

> > Think current source/sink with a good voltage compliance range, > > then line length is a non-issue. > > And does this go against what the others have suggested, or > is there a difference between a current source/sink and a constant > current source. That's the same thing, and the advice is good. If you power you LEDs from eg 12V you switch an NPN transistor from the 0/5V output of a PIC (you use a PIC?) and you put a resistor between emitter and ground, let's say 330 ohm, which determines the current. The LED is between collector and +12V. No other resistors. Wouter van Ooijen -- ------------------------------------------- Van Ooijen Technische Informatica: http://www.voti.nl consultancy, development, PICmicro products docent Hogeschool van Utrecht: http://www.voti.nl/hvu