Home > Hardware > DATAQ Starter Kit Support > Forum > DI-149 analog input protection

DI-149 analog input protection

Jan 18, 2015
DogT wrote
Will this work to protect the analog input circuits, I seem to have had problems with the relays surge in the past. In6373 is a 5V Transzorb across the + and - terminals. There is a relay for each of the 5 channels, only one shown for simplicity.

[URL=http://s251.photobucket.com/user/DogTi/media/other/1n6373_zps2f47070a.jpg.html][IMG]http://i251.photobucket.com/albums/gg287/DogTi/other/1n6373_zps2f47070a.jpg[/IMG][/URL]
Reply
10 Answers
Jan 18, 2015
DogT wrote
I see inbedded pics don't work here's the direct link

http://s251.photobucket.com/user/DogTi/media/other/1n6373_zps2f47070a.jpg.html
Jan 19, 2015
DATAQ Support agent wrote
You imply that inductive back emf is a problem, but your schematic doesn't show a direct connection to the solenoid. Instead you do show simple dry contacts. If the latter is the actual situation, then all you need are pull-ups on the DI-149 analog inputs to determine open or closed state of the switches. Connect the (-) side of all the analog inputs to the (-) of the power supply, and then each of the (+) inputs to the (+) side of the power supply through a 10 K-Ohm resistor. If you're measuring 5 channels, you'll need 5 resistors. Then place your dry contact switch across an input channel. When the switch is open the DI-149 will see the power supply voltage. When it's closed it will see 0 V. In other words, you don't need the diodes if you're measuring dry contacts.
Jan 19, 2015
DogT wrote
Well, what you suggested is exactly what I did with my last DI-149 and I started loosing channels. Some of them wouldn't work properly any more. Yes, I'm using just the dry contacts off the relays to pull up and down the voltage at the + end of the 10K resistors. The wires are not connected to anything else so there shouldn't be any common mode issues, but I've measured it with the Flluke 198 to the computer ground and don't see any voltage more than 1/10s of volts AC or DC. Each relay has a run of about 50-80' of wire to the DI-149 though, and I did detect a rather large spike on relay open and closing with an O-scope, it's hard to measure with a low freq or non storage scope, maybe just induction from the coil back emf or lightning induction, anyhow the Transzorbs won't hurt anything will they? I already have it breadboarded and it seems to work fine on the bench, but I haven't hooked up the relays yet. The transzorbs should reduce any spikes to about 7V or less and actually I'm only using about 3V for the supply. I still have a DI-194RS that's been working for 10 years now with no pull up resistors or surge protection, just running off the relays to short the analog input, but they have a different input circuit and I need 5 channels now.
Jan 21, 2015
DATAQ Support agent wrote
It is not possible to damage especially the DI-149 inputs, which are exceedingly more protected than the DI-194RS, by connecting dry contacts in the manner described. Also, if you saw a large spike on the signal, then it isn't a dry contact. What I mean by a dry contact is that no voltage or current exists on the field. It's simply a switch closure. This assumption is not at all consistent with the presence of a "large spike" on the dry contact. How exactly did you connect the scope to measure this event?
Jan 21, 2015
DATAQ Support agent wrote
"...that no voltage or current exists on the field END."
Jan 21, 2015
DogT wrote
Maybe I got a bad 149, anyway it was replaced and I don't want to keep buying these things. I think the first one started going bad after about a month. I've had it working for a day now with the Transzorbs and it seems fine, I may even put some .01uF caps across the analog inputs too, I seem to be getting some noise on the lines but it really doesn't affect my measurements. There's probably some crosstalk too since all the - are connected together. Better safe than lose channels. I thought maybe when the coil is de-energized, the field collapses and may induce a spike in the dry contacts even if it isn't directly connected. It's been a long time since I put the scope on the lines, I put it right across from - to +. They're very small 12VDC relays I've pressed into service, they work off a 24VAC line with a diode and dropping resistor. But yeah, I never had a problem with the 194 completely unprotected. I think I can measure small time period spikes with the Fluke 189, but I'll have to read up on that and the transzorbs would absorb it anyhow. I could just disconnect an input for a while to see.

You can see the noise here, I though it maybe was because I turned on my KW transmitter because it happened at the same time, but then it happened later when the transmitter was off. Doesn't appear to be common mode noise. Ignore those glitches on the beginning latch of the bottom track, it's a dirty relay.

http://s251.photobucket.com/user/DogTi/media/other/5channels_zpsc9ba288b.jpg.html

Thanks for the help.
Jan 21, 2015
DATAQ Support agent wrote
Strange...almost looks like weak pull-up resistors. What pull-up values are your using? If 10K, have you verified that they are in fact 10 K and not something else? Also, since the noise appears to occur only while the switch is closed, it could be that the switch is not making perfect contact in some situations. That would definitely cause these results.
Jan 21, 2015
DogT wrote
Actually the noise occurs when the relay is open, notice I've got the channels inverted. 0 shows up high.

10K resistors. I'm sure the resistors are fine, I measured them with the DMM before installation. The noise may actually be because I have some clip leads connecting things and it's not all finally hard wired and soldered together yet. At any rate, it's doing what I need it to do and I have some piece of mind with the Transzorbs. The relay wires do go through a bunch of HVAC and electrical conduit to reach the 149 so it wouldn't surprise me that there could be some induced voltages in them. The - connections are only connected together right at the 149, so all the relay wires don't have any common point at the relay end and that's a lot of loose wire hanging out there.

I wired up a relay on the bench with a 12VDC PS next to the scope and measured the dry contacts while switching the relay on and off and could only come up with mV, probably just relay bounce and maybe some small induced voltage. Same thing with the Fluke measuring fast min/max, I couldn't come up with any significant voltage across the contacts. It kind of makes me think my first 149 just went bad. Anyhow, I was sent a new one and I want to take precautions with it.

Thanks
Jan 23, 2015
DATAQ Support agent wrote
Thanks for the feedback...okay to mark this ticket as RESOLVED?
Jan 23, 2015
DogT wrote
yes