Control Relay with I/O

Jeff

I hired a company to design a system for me, and it includes a Projector, an NI700, and a Kramer VP-211DS Autoswitch. Two Computers plug into the autoswitch.

I specifically pointed out that the NI700 didn't have relays, and was told that the NI700's I/Os could be used to control this autoswitch.

The autoswitch has a 3pin phoenix block with input1, input2, and a common pin. connect input1 to common to select input 1, connect input2 to common to select input 2.

I tried hooking this up to the I/O on the NI700, but I must admit its a bit confusing. I initially connected common to the I/O ground and input1 to channel 1, input2 to channel 2, but that just left me with I/O channel 1 perpetually selected, and I couldn't deselect it or do anything with it. I could turn on channel 2, but that just connected both of them and made the autoswitch unhappy.

Am I hooking this up wrong? Was I informed incorrectly that this would work? Do I need something like an Extron IPA T RLY4? If so, how would I hook that up?

Thanks for all your help, I've never used I/Os before, so I'm a bit confused.

J

Hedberg

I have done exactly what you describe using the IO ports and a Kramer 2 input VGA switcher. It may be the same switcher that you are using (same model, anyway), I'm not sure, but it should work.

The IO ports can function either as inputs or outputs(quite obviously). You can set input detection to either high or low (see the NI700 manual). Default is low which means that if the IO port detects a contact closure (0 v) or a voltage less than 1.5 volts, it infers a push and your light goes on. This implies to me that the connection to your IO port 1 input is somehow shorted to your NI700 ground connection. If you have the connection to the Kramer switch wired correctly, this should not be the case. The Kramer switch should present neither a contact closure or a low dc voltage.

Check the connection to the switch. According to the manual for this device (available at the Kramer site), the contact closure connector has three pins -- the middle pin is the common pin, pin 1 is for selecting input 2, and pin 3 is for selecting input 1. Note that there are a couple switches on the bottom of the Kramer switch for selecting a default input -- for when neither pin 1 or pin 3 is connected to pin 2. This means that you can switch this device between inputs 1 and 2 using just one contact/IO port.

If you continue to have problems, check the Kramer switch without connecting it to the NI700. Check for a voltage or short between pins 2 and 3 (which are connected to the NI700 ground and IO 1 respectively). There should be neither. Check that the switch defaults IAW the buttons on the bottom and that when you hard wire between pins 2 and 1 or between pins 2 and 3 that you get the proper switch behavior.

I don't believe that you can check the IO ports with a VOM, but you can connect a simple circuit to test them. One way is to use a 12vdc pc cooling fan (something which aways seems to be lying around almost everywhere). Connect the IO port 12v to one lead and the IO pin of interest to the other lead. Fan should not run until you turn the IO port on. Note that the IO port has to be set as 'LOW' in order to function as an output.

annuello

Gday Jeff,

You can control relays from the NI-700 ports, but I would not use the Extron IPAT RLY4. Looking at the Extron system, they use a common Gnd between their relays. To drive relays from the NI-700 you need to use a common 12v sysem. Any possible protection diodes in the Extron product would prove damaging to the NI-700 port. Instead, make up your own relay box with protection diodes.

There is an example of using protection diodes on this nice site.
http://www.kpsec.freeuk.com/components/relay.htm
http://www.kpsec.freeuk.com/components/diode.htm

Hook the cathode (K) to the +12V on the NI-700 IO port. Hook the anode (A) to the IO port number that you want (E.g. 1, 2, etc.) Connect the relay coil over the 12V and IO port number. Don't hook any part of the relay circuit to the Gnd.

When I initially tried this I used little DIP reel relays. They ended up breaking down in the soldering process. You can drive up to 200mA per port, but just double check the NI-700 spec file for your own satisfaction. I use the following relay:
http://www.jaycar.com.au/ Search for part number SY4065. Use the coil resistance & rated voltage to figure out your current. In my case it is V=IR, I=V/R=12/160=0.075A

I've done this on 7 systems (2 relays per system), and it's working fine. One relay box was wired up wrong, whic blew up the IO port on the NI-700, so that master will now be used for RMS-type usage.

Hope this helps.
Roger McLean
Swinburne University.

GregoryArndt

I do this all the time in theme parks. I don't always use AMX in the projects, but I/O is I/O. They all work basically the same.

When in Output mode, I/O on the NI700, as with all Ni controllers, acts as a shared collector switch (similar to open collector, only the collectors are connected internally). They each act as a switch to sig. gnd. but cannot produce a dry contact closure. These are 'sinking' outputs, not 'sourcing'. It is best, as already mentioned, to use relays in between the I/O and the controlled equipment. This safeguards againsts back voltages, etc.

I have attached a schematic to show how I, and others, have wired relays to I/O in the past, present and future jobs. Figure 1 shows how the I/O schematically works. Figure 2 is a suggested way to connect the I/O to your switcher. The diodes are not necessary if you get relays with the diodes built in (not always suggested, as client replacement may not be exact).

If you really want to be safe, and protect the power supply from frying everything in an electrical storm, you can put an opto-isolator in between the I/O and the relay. Just in case. Think safety. ;-)

Hope this helps,

Jeff

OK, So I got a multimeter and tried a few things.

With the phoenix block plugged into the "Remote" port on the VGA switch, but no wires are hooked in, pin 2 is NOT shorted to pin 1 or pin 3.

If Pin 2 is hooked up to the ground on the I/O port, it somehow creates a contact closure between pin 2 and pin 3 (Input 2).

This in turn, when I hook the entier thing up correctly, means that the AMX is sensing that contact closure, and activating that port.

So what is it about the I/O ground that is creating this contact closure?

J

Jeff

I also forgot to mention, I"ve tested this on 3 different VGA switches and 3 different masters, and the same problem is constantly repeating itself

J

Jeff

Further bizzarre circumstances:

if I hook up JUST pin 1 on the switch to I/O pin 1 on the AMX, and I turn on that I/O, there is no contact closure between VGA Switch pin 2 and pin 3. Thats to be expected. Whats not to be expected is the fact that turning on that I/O, instead of switching to input 1, causes the VGA switch to ACT as if 1, 2, and 3 are all shorted together.

I want to again repeat that working with relays, this all works perfectly . . . . . I realize that some of you made suggestions for how to make it work by adding relays, but I'd rather try to find out if this is possible without adding anything else. As you can see, I work for the government, which means I wont get any new gear for weeks, even if i CAN convince someone its necessary, which will be hard.

J

DHawthorne

The IO ports are just collector outputs. If the impedance is low enough between the IO port and ground, it's going to trip the thing into sensing an input. If you turn it "on" via programming, it just turns the transistor driving it on, making the collector go high. The coil of a relay is plenty high enough in impedance not to trigger an input state, but if it's the kind with a diode shunt across the coil, I could see it possibly causing issues if it was hooked up reverse polarity.

Hedberg

Like I said, we've installed Kramer 2X1 VGA switches and triggered them from the IO ports. The remote pins on the Kramer are not relay inputs and I can't understand how they would trigger the IO port. Quite obviously, something is going on that didn't happen to me, and I'm glad for that. I surely can't explain how connecting pin 2 on the Kramer to the IO signal ground could cause any sort of a trigger -- the Kramer should be powered with a separate, isolated supply and I don't see how there could be any effect at all with a one wire connection.

Perhaps use of pin one on the Kramer remote could be abandoned and it could be triggered with an IO port just using pins two and three. This thing is controllable with just one relay/port.

Otherwise, this:<http://www.web-tronics.com/duhiswrebok1.html&gt; would be a clean, inexpensive solution.

Sergio

IO with shared collector or no?

GregoryArndt wrote:

When in Output mode, I/O on the NI700, as with all Ni controllers, acts as a shared collector switch (similar to open collector, only the collectors are connected internally). They each act as a switch to sig. gnd. but cannot produce a dry contact closure. ,

I may be wrong, but all IO are open collectors on picture attached , not shared collector. Is it correct?