Elk M1 control to P212s common wire

[mikefamig]

I've seen several posts about installing a p212s power supply and the need to tie together the common wire between the control and the p212s. This is also evident in the wiring diagram that comes with the power supply.
 
My question is why is this wire necessary? The way I see it each of the two power supplies will work independent of each other with or without this wire. One may have a little different voltage than the other but does this matter as long as each device is supplied sufficient power to work? Does it have to do with monitoring the supplies?
 
Mike.

 

[mikefamig]

I'd also like to talk about what gauge wire is necessary for the common lead in the Elk system in general.
 
In the Elk -  23awg cat 5 wire is used to supply power to devices on the rs485 data bus. The power supply is capable of creating 1 amp at 12 volts (2amps for the ps212) which is more than the 23 gauge wire is capable of conducting without heating up or even melting. Isn't that a problem? Being that the 12v power is connected in parallel to all devices on the data bus isn't there a potential to overload the cat5 wire? Shouldn't there be a home run of both + and - leads to each device on the data bus or a larger gauge wire for the common lead?
 
Mike.

 

[RAL]

I'm not sure if you are an electrical engineer or not, so I'll assume that you're not and will try and not get too technical.
 
When you have devices operating from different power supplies, if there is to be any communication between them, you generally need a common reference point, and the ground (negative) power signal is often used for that purpose.  For example, suppose you wanted to use an output from the M1 to trigger an input on an M1XIN that was on a different power supply (not sure why you would want to do that, but ignore that issue for the moment).  Without a common ground between them, the M1XIN couldn't tell that it was receiving a +12V output from the M1.
 
But often, the only thing that you have connected between the M1 and devices on other power supplies is the RS485 data bus.  And the only communication between them is over the Data-A and Data-B wires, which operate as a differential pair.   With differential pairs, when you put a positive signal ( a "1") on the A-wire, you put a negative signal (a "0") on the B-wire, and vice-versa.   By having two wires with opposite signals, you don't need any outside reference such as ground to transmit data. 
 
That's the pure theory, but it ignores the effects of long wires which are prone to pick up electromagnetic interference.  This is called the "common mode" signal.  The longer the wire and the greater the outside interference, the worse this becomes.   At some point, you can't transmit data any more.
 
To reduce the effect of the common mode signal and maximize the distance you can transmit data, a ground is connected between the two sides. And the RS485 spec recommends doing this.
 
On your question about the current being carried over the Cat5 cable, you are correct that there are some places where you may need to be concerned.
 
Most of the data bus devices, such as keypads and M1XINs don't draw much current,  just 50 to 85 mA or so.  But the M1XOV and M1XOVR can draw 330mA. 
 
[Edit ]
 
If you have everything home runned (with or without a M1DBH), you should be ok, since it is just the two data signals that get daisy chained, not the power wires.  Otherwise, if you run a true daisy chain from one device to another, you may need to use heavier gauge wire depending on what type and how many devices you have on the chain.  Note that using a M1DBH when you have multiple power supplies gets tricky since you have to be careful not to connect the +12V signals together between the supplies.

 

[mikefamig]

RAL
 
Thanks, that does clear things up a little for me and no I am not an electrical engineer. I've worked professionally as a computer programmer/analyst and am a self-taught electronic hobbyist. I've built several pc's and notebook computers, did all of the wiring in my new garage and have done some transistor radio troubleshooting and repair so I have a good understanding of DC circuits and a little knowledge of AC circuitry. I read a lot and ask a lot of questions.....ya gotta love the internet!
 
I forgot that the system has no earth ground as a reference and that is what confused me. I will park the +12v from the M1 control on the dead terminal of the p212s and then connect the +12v source from the p212 to the DBH via the rs485 ethernet cable. Is it correct to say that there is a home run between the DBH and each device connected to it for the +12v power leads with all devices connected in parallel?
 
I have never heard the term "common mode" but am aware that current can be inducted from one wire to another and create interference of data transmission. For this reason I had the builder bury two separate conduits from the house to the garage, one for high voltage lines and one for data. I don't really see though how adding a ground alongside the wire helps. Does it catch the interference and send it to ground like a shielded wire protects the lines inside the shield?
 
Mike.

 

[mikefamig]

I don't really see though how adding a ground alongside the wire helps. Does it catch the interference and send it to ground like a shielded wire protects the lines inside the shield?

 
Mike.

 
Or are you saying that the circuit uses the ground as a reference to help it to detect a signal on the differential pair?

 

[RAL]

I forgot that the system has no earth ground as a reference and that is what confused me. I will park the +12v from the M1 control on the dead terminal of the p212s and then connect the +12v source from the p212 to the DBH via the rs485 ethernet cable. Is it correct to say that there is a home run between the DBH and each device connected to it for the +12v power leads with all devices connected in parallel?

 
I have never heard the term "common mode" but am aware that current can be inducted from one wire to another and create interference of data transmission. For this reason I had the builder bury two separate conduits from the house to the garage, one for high voltage lines and one for data. I don't really see though how adding a ground alongside the wire helps. Does it catch the interference and send it to ground like a shielded wire protects the lines inside the shield?
 
Mike.

 
I'm not sure I follow what you say you will do to keep the +12V from the M1 and the P212S separate.  All you need to do is leave the +12V wire in the data bus cable unconnected on the P212S end.  If you do that by connecting it to the spare terminal, that is ok.  Don't connect the +12V output from the P212S back to the M1 !
 
With home runs to each device, the +12V and NEG (ground) signals are all connected in parallel back at the DBH.  The data signals get connected in a daisy chain fashion through the DBH circuit board.  That's why two wires get connected to the Data-A and Data-B signals.
 
The ground connection between the M1 and the remote power supply doesn't really catch  or remove interference on the RS485 link in terms of providing any kind of shielding.
 
A differential signal like RS485 uses a voltage change in opposite directions to send data bits.  This is usually a small voltage of about 1.5V volts.   
 
For example, on one end, this could be centered around 3.75V, so that the Data-A wire would be at 4.5V while the Data-B wire would be at 3.0V.
 
Ideally, the other end would also be at these same voltage levels.  Without a ground between the two ends, the signal level can drift apart due to common mode voltage on the cable. The further apart things are, the more likely it is that "ground" at one end will not be the same as "ground" at the other end. 
 
Say there is 5V of difference between the two ends, that would make the voltages 8V and 9.5V.  (Common mode voltage gets added to both wires  - that's why it's called common!)  Things should still work ok at those levels.  But if it goes too high, that's where the trouble starts.
 
In theory, the common mode voltage doesn't matter at all. The receiver circuit just has to detect the difference between the two wires.  But semiconductor circuits have limits on their range of operation, and if the common mode voltage gets too large, they stop working.  Connecting a common ground between the two ends helps reduce the common mode voltage by bringing the two ends more into line with each other.

 

[mikefamig]

I'm not sure I follow what you say you will do to keep the +12V from the M1 and the P212S separate.  All you need to do is leave the +12V wire in the data bus cable unconnected on the P212S end.  If you do that by connecting it to the spare terminal, that is ok.  Don't connect the +12V output from the P212S back to the M1 !

>On the p212s the +12v  terminal input from the control is a dummy that is there just to attach the +12v from the M1 to so it doesn't go astray. The cat5 that connects the M1 control to the p212s gets  all four wires attached on both ends but the p212s. Data A, Data B and common are used and +12v just connects to to a dummy terminal on the p212s doing nothing.
 
I plan to take the +12v output on the p212s to power the DBH by attaching it to the cat5 that is attached between the p212s and the DBH.

With home runs to each device, the +12V and NEG (ground) signals are all connected in parallel back at the DBH.  The data signals get connected in a daisy chain fashion through the DBH circuit board.  That's why two wires get connected to the Data-A and Data-B signals.
 
The ground connection between the M1 and the remote power supply doesn't really catch  or remove interference on the RS485 link in terms of providing any kind of shielding.
 
A differential signal like RS485 uses a voltage change in opposite directions to send data bits.  This is usually a small voltage of about 1.5V volts.   
 
For example, on one end, this could be centered around 3.75V, so that the Data-A wire would be at 4.5V while the Data-B wire would be at 3.0V.
 
Ideally, the other end would also be at these same voltage levels.  Without a ground between the two ends, the signal level can drift apart due to common mode voltage on the cable. The further apart things are, the more likely it is that "ground" at one end will not be the same as "ground" at the other end. 
 
Say there is 5V of difference between the two ends, that would make the voltages 8V and 9.5V.  (Common mode voltage gets added to both wires  - that's why it's called common!)  Things should still work ok at those levels.  But if it goes too high, that's where the trouble starts.

You explained that very well and I think that I get it now. The common mode voltage is inducted into data a and b along it's length. I said earlier that I learned what I know about electronics by reading what I can get and asking a lot of questions and I thank you for the lesson.

In theory, the common mode voltage doesn't matter at all. The receiver circuit just has to detect the difference between the two wires.  But semiconductor circuits have limits on their range of operation, and if the common mode voltage gets too large, they stop working.  Connecting a common ground between the two ends helps reduce the common mode voltage by bringing the two ends more into line with each other.

This sounds similar to how a transistor will switch based on the voltage applied to each of it's three legs in reference to ground and how it will go pop when voltages go too high. You're getting a little over my head but I think that I understand the theory.
 
Mike.

 

[RAL]

On the p212s the +12v  terminal input from the control is a dummy that is there just to attach the +12v from the M1 to so it doesn't go astray. The cat5 that connects the M1 control to the p212s gets  all four wires attached on both ends but the p212s. Data A, Data B and common are used and +12v just connects to to a dummy terminal on the p212s doing nothing.
 
I plan to take the +12v output on the p212s to power the DBH by attaching it to the cat5 that is attached between the p212s and the DBH.

 
Are you mounting the DBH near the M1 and the P212S somewhere else?  Or are they all close to each other?
 
If I understand what you said, you want the P212S to provide power to all the keypads and whatever other devices you have connected to the DBH, correct?
 
It's ok to do it that way.  But then you need to make sure you don't connect the wire from the DBH Control connector "+" terminal back to the M1's +VKP terminal.
 
I think you've probably figured that out, and I'm just not clear on how/where you are keeping the M1's power isolated from the P212S. And I want to make sure that something I say doesn't cause you to screw it up.  
 
My only concern right now with what you said is that you want to run the power from the P212S through the Cat5 cable to the DBH Control "+" terminal.  The 23 gauge wire in the cable can carry 0.7 Amps per conductor.  You'll need to use three conductors for +V (and 3 more for ground) to be able to handle the 2A that the P212S can deliver.  My choice would be to use 18 gauge wire for this instead.

 

[mikefamig]

Are you mounting the DBH near the M1 and the P212S somewhere else?  Or are they all close to each other?

I have two DBH in the system. One is adjacent to the control and used to connect devices in my house and it will be powered by the M1 control. The second is mounted in the detached garage connected directly to the control via about 250' cat5e (which will NOT have +12v lead connected). It is the devices on the second DBH in the detached garage that will be powered by the p212s.

If I understand what you said, you want the P212S to provide power to all the keypads and whatever other devices you have connected to the DBH, correct?

Partially correct. The p212s will power all of the devices connected to the DBH in the detached garage.

It's ok to do it that way.  But then you need to make sure you don't connect the wire from the DBH Control connector "+" terminal back to the M1's +VKP terminal.
 
I think you've probably figured that out, and I'm just not clear on how/where you are keeping the M1's power isolated from the P212S. And I want to make sure that something I say doesn't cause you to screw it up.  

I am clear on this

My only concern right now with what you said is that you want to run the power from the P212S through the Cat5 cable to the DBH Control "+" terminal.  The 23 gauge wire in the cable can carry 0.7 Amps per conductor.  You'll need to use three conductors for +V (and 3 more for ground) to be able to handle the 2A that the P212S can deliver.  My choice would be to use 18 gauge wire for this instead.

This is why I asked about wire gauge in the first place (see post #2). The cat5e really isn't even sufficient to carry the 1 amp that the M1 control is capable of supplying. I agree with you about twisting together 3 cat5 leads for the plus and minus power leads on the cable between the ps212 and the DBH but not for connecting each device to the DBH..  The DBH connects the power leads in parallel and these leads will only need to carry the maximum current of any one device assuming that I am correct that the DBH does indeed connect power to all devices in parallel..
 
Mike.