Orlando, FL
Offline
Full Member
Karma: 0
Posts: 128
|
 |
« on: October 01, 2010, 01:06:30 pm » |
I'm trying to take some -48V signals into an Arduino. Using a voltage divider circuit i'm dissipating about 3 watts of power. Anyone know of a better way to feed 48 V into an optoisolator?
|
|
|
|
|
Logged
|
|
|
|
|
North Yorkshire, UK
Offline
Faraday Member
Karma: 104
Posts: 5531
|
 |
« Reply #1 on: October 01, 2010, 02:28:12 pm » |
Using a voltage divider circuit i'm dissipating about 3 watts of power Use much larger value resistors in your voltage divider circuit... But as Richard says, surely a current limiting resistor would suffice? Mowcius
|
|
|
|
|
Logged
|
|
|
|
|
Orlando, FL
Offline
Full Member
Karma: 0
Posts: 128
|
 |
« Reply #2 on: October 01, 2010, 02:45:31 pm » |
a current limiting resistor is causing too much heat  I need at least 40ma though the circuit to kick on the opto. I'll post a schematic in a bit.
|
|
|
|
|
Logged
|
|
|
|
|
0
Offline
Full Member
Karma: 1
Posts: 222
Arduino rocks
|
 |
« Reply #3 on: October 01, 2010, 03:44:49 pm » |
Darlington optocoupler?
|
|
|
|
« Last Edit: October 01, 2010, 03:50:15 pm by lefstin »
|
Logged
|
|
|
|
|
Orlando, FL
Offline
Full Member
Karma: 0
Posts: 128
|
 |
« Reply #4 on: October 07, 2010, 01:06:56 pm » |
here is the circuit. as you can see I have an input voltage of -48VDC and i'm trying to have that send a high signal to the arudino when the 48 V signal is lost. the problem is for a circuit that will be in the ON state for most it's life .7Watts of heat across R1 is just too high. 
|
|
|
|
|
Logged
|
|
|
|
|
Orlando, FL
Offline
Full Member
Karma: 0
Posts: 128
|
 |
« Reply #5 on: October 07, 2010, 02:10:36 pm » |
Because when you have 115 channels of this the total heat on the circuit board will be over 80 Watts dissipated. And i'm not trying to build a heater =) I could use relays but i wanted the solid state reliability.
|
|
|
|
|
Logged
|
|
|
|
|
0
Offline
Faraday Member
Karma: 12
Posts: 2857
ruggedcircuits.com
|
 |
« Reply #6 on: October 07, 2010, 02:57:05 pm » |
Maybe you want to consider an isolator with lower current demands, like a digital isolator: http://search.digikey.com/scripts/DkSearch/dksus.dll?Detail&name=ADUM7440ARQZ-NDThe input current per channel is only 10uA worst case so you can use a voltage divider with large value resistors to minimize power. The unit does need a ~2mA power supply though, but you will still probably come out ahead as it has 4 channels so it's really only 500uA per channel. -- Need a custom shield? Let us design and build one for you.
|
|
|
|
|
Logged
|
|
|
|
|
0
Offline
Full Member
Karma: 1
Posts: 222
Arduino rocks
|
 |
« Reply #7 on: October 07, 2010, 03:19:21 pm » |
I repeat, a Darlington optocoupler ought to do the trick for you.
It will saturate with just a few mA or less of input current. For your application you do not have to worry about output current or switching times.
|
|
|
|
« Last Edit: October 07, 2010, 03:19:46 pm by lefstin »
|
Logged
|
|
|
|
|
Orlando, FL
Offline
Full Member
Karma: 0
Posts: 128
|
 |
« Reply #8 on: October 07, 2010, 03:48:53 pm » |
Here is the circuit i'm using. Each isolator is only pulling 1.6mA now and the power dissipation is only 76.8mW each. 
|
|
|
|
|
Logged
|
|
|
|
|
0
Offline
Newbie
Karma: 0
Posts: 17
|
 |
« Reply #9 on: October 08, 2010, 01:07:42 pm » |
stealthtransam,
What kind of linearity are you getting with the optocoupler? I was thinking of going with the LOC110 which is supposed to be linear but I haven't received it yet. The regular opto couplers suffers from non-linear outputs thus not recommended for ADCs.
The problem I have is being able to measure -48 v systems and +24v systems with one unit. Altho I found a solution for that(relays, voltage inverter), I am still looking for alternate solutions.
|
|
|
|
« Last Edit: October 08, 2010, 01:08:38 pm by royco »
|
Logged
|
Panucatt Devices
|
|
|
|
Orlando, FL
Offline
Full Member
Karma: 0
Posts: 128
|
 |
« Reply #10 on: October 09, 2010, 12:25:35 pm » |
I've run the input voltage from 18-54 on both a positive +24V system and -48V system and still had a stable TTL logic out of it using that 33K resistor. I know it's under driving the current at that point but it still works great.
|
|
|
|
|
Logged
|
|
|
|
|
|