Best solution for monitoring loss of voltage?

Status
Not open for further replies.
U

ucdoshi

Member
Location
bakersfield
We have around 50 distribution panels (120/240V) spread on 5floors. We need to monitor loss of voltage at individual panels. We are not going to monitor each feeder breakers from each of the panels, but just the main breakers at the DP. The control room is in a different building. What is the best way to monitor loss of voltage at each of these panels such that the operator comes to know something is wrong at a particular room. Using a relay is not an option..
Thanks in advance for the help!
 
ucdoshi said:
We have around 50 distribution panels (120/240V) spread on 5floors. We need to monitor loss of voltage at individual panels. We are not going to monitor each feeder breakers from each of the panels, but just the main breakers at the DP. The control room is in a different building. What is the best way to monitor loss of voltage at each of these panels such that the operator comes to know something is wrong at a particular room. Using a relay is not an option..
Thanks in advance for the help!
Click to expand...

If you cannot use a relay, it is pretty hard to get the information from one place to another using low voltage wiring.
The simple way to detect the voltage at each panel would be to run a set of wires from each panel back to a central point where you measure the voltage. But to be able to do that using reasonably sized wires you would need to put overcurrent protection on each hot line where it originates in the subpanel, and the wires would still have to be insulated to the full system voltage.

If all you need to know is if the feeder breaker serving each suppanel has opened, rather than whether the wires have failed or the main breaker at the subpanel has opened, you can at least just monitor the 50 feeder breakers at what you call the DP. How to get a remote indication from 50 of them is still a major problem.
 
If under normal conditions each phase of each feeder has a load on it you might be able to do this with current monitoring at the distribution panel(s). Of course we do not have enough info to determine that from here.
 
Do not understand why relays are not acceptable since that would be the obvious answer.
I would fit one or more relays to the feed into each panel so as to detect the presence or absence of the usuall line voltage.
Use relays with changover contacts, and run low voltage wires (cat 5 or phone wire)from the relay contacts back to a common control point or room.
Wire such that the presence of normal line volts at the distant panel lights a green lamp, and the absence of normall voltage lights a red lamp.

For really thorough monitoring you would need 3 relays for a 3 wire single phase, 120/240 volt supply. One to monitor for 120 volts between the A phase and neutral, one to monitor for 120 volts between B and neutral, and a third relay to monitor for 240 volts between A + B.
Only one 3 core cable and pair of lamps would be needed per panel since the intention is presumably simply to give a "normal" or a "fault" indication and not say exactly what the fault is.

Cant see any simple way to do it without relays though.
 
Well assuming these are single phase panels, you could just put in a 15A 2 pole breaker in each one and feed a 240V pilot light that you mount in a box into the control room. It satisfies the "no relay" requirement, but is not really the best way to do it.

If you want to do it via a network communications scheme, you could simply get a voltage sensor in each panel that has an Ethernet port or some other network communications capability. There are tons of those out there now. Once it is network ready, you can use wireless communications back to the control room.

Expensive, but that was not one of your criteria.
 
Jraef said:
.

Expensive, but that was not one of your criteria.
Click to expand...

I assumed this also since OP's profile is consultant, whatever that means, I guess a consultant dispenses advice.

And so my advice is to the OP is: Hire a company that specializes in these monitoring installs. Cheaper IMO then trying to figure it out if this not your forte.

Many times DIYing, whether personally or professionally is more expensive in my experience.
 
Jraef said:
Well assuming these are single phase panels, you could just put in a 15A 2 pole breaker in each one and feed a 240V pilot light that you mount in a box into the control room. It satisfies the "no relay" requirement, but is not really the best way to do it.

If you want to do it via a network communications scheme, you could simply get a voltage sensor in each panel that has an Ethernet port or some other network communications capability. There are tons of those out there now. Once it is network ready, you can use wireless communications back to the control room.

Expensive, but that was not one of your criteria.
Click to expand...

Kind of what I said, indicator lights, but that is all they will do is indicate.

I would consider the network connected device to be classified as a relay in a way, maybe not a soleniod operated relay but still some kind of device that accepts input of one voltage and has an output of another voltage, or even different signal type.
 
Thank you all for your reply...My description might have been little misleading. We do not want to use a typical idec relays (coils/contacts) or similar relays. We are ok with voltage sensing relays that have communication capabilities. The DPs are spread on 5 floors and each floor has around 10DPs installed away from each other. The control room is in a different building. There will be too much of wiring if we use a regular relay and a green/red light panel. There is one solution to use BMS software in the control room and install digital voltage sensing meters and connect them to an ethernet switch which sends over information to the software...Recommendations on the voltage sensing meters and the software? Honeywell, Johnson Controls or Siemens??
 
kwired said:
Kind of what I said, indicator lights, but that is all they will do is indicate.

I would consider the network connected device to be classified as a relay in a way, maybe not a soleniod operated relay but still some kind of device that accepts input of one voltage and has an output of another voltage, or even different signal type.
Click to expand...
Meh, I think there is a difference, but you'll get no more argument than that from me, because all relays are a form of transducer, so you could make an argument that saying "No relays" is like saying "No transducers" as well. But we don't know why the silly rule about relays is there to begin with, so if it is just semantics, that may be a workaround.

Semantics works sometimes. I once had a large project where we were using hundreds of Siemens micro PLCs as data loggers because of the native Modbus (rare at the time), but the end user insisted that all PLCs be Mitsubishi, in which Modbus was a royal PITA. So we never used the term PLC on any drawing or documentation, we strictly called then Data Loggers, since that's what we used them for. Everyone was happy.
 
Jraef said:
Meh, I think there is a difference, but you'll get no more argument than that from me, because all relays are a form of transducer, so you could make an argument that saying "No relays" is like saying "No transducers" as well. But we don't know why the silly rule about relays is there to begin with, so if it is just semantics, that may be a workaround.
Click to expand...

As Stephen Colbert stated: "It is not just a matter of semantics, it's also about the words we use."
 
Last edited:
Jraef said:
Meh, I think there is a difference, but you'll get no more argument than that from me, because all relays are a form of transducer, so you could make an argument that saying "No relays" is like saying "No transducers" as well. But we don't know why the silly rule about relays is there to begin with, so if it is just semantics, that may be a workaround.

Semantics works sometimes. I once had a large project where we were using hundreds of Siemens micro PLCs as data loggers because of the native Modbus (rare at the time), but the end user insisted that all PLCs be Mitsubishi, in which Modbus was a royal PITA. So we never used the term PLC on any drawing or documentation, we strictly called then Data Loggers, since that's what we used them for. Everyone was happy.
Click to expand...

OP stepped up to the plate, good thing too. Transducer was a good word to use which I could not seem to come up with an appropriate word before. I guess you equated transducer to include relay and my first thought was to include electronic type devices to be included in what is a relay. Either way there has to be some kind of intervening device to do much more than simple direct connection to the line(s) and very basic indication.
 
I don't know why relays are out. That seems like the simplest solution, and relatively inexpensive.

I concur though that your best bet might be to just hire someone competent to do this for you since it seems outside of your area of expertise.
 
Jraef said:
Meh, I think there is a difference, but you'll get no more argument than that from me, because all relays are a form of transducer, so you could make an argument that saying "No relays" is like saying "No transducers" as well. But we don't know why the silly rule about relays is there to begin with, so if it is just semantics, that may be a workaround.

Semantics works sometimes. I once had a large project where we were using hundreds of Siemens micro PLCs as data loggers because of the native Modbus (rare at the time), but the end user insisted that all PLCs be Mitsubishi, in which Modbus was a royal PITA. So we never used the term PLC on any drawing or documentation, we strictly called then Data Loggers, since that's what we used them for. Everyone was happy.
Click to expand...

I have seen small PLCs referred to as RTUs for the same reason.
 
130522-0759 EDT

ucdoshi:

I am working on the development of a binary voltage presence detector. This interfaces to a "1-wire" bus, and in turn that can interface to an RS-485 bus. This is not yet available and I am not trying to sell it to you. The purpose of this comment is to describe the system.

What do I mean by "binary voltage detection"? This is a voltage sensing device that has two states defined by a threshold voltage. An ordinary relay, optical coupler, vacuum tube, or transistor can all be used to perform this input detection.

By using a single bus system with addressable sensors one can have many sensors and a simple wiring scheme.

At the master end of the bus there can a rather complex analysis program with a simple output display.

The 1-wire network is useful up to 100 ft or so. This is not a high noise immunity bus. RS-485 is usable to 4000 ft at 115.2 kbaud, and 8000 ft at 9600 baud. The "1-wire" name is a registered trademark of Dallas Semiconductor now part of MAXIM.

The device I am working on is intended for monitoring residential furnace blower and thermostat signals, and industrial machine conditions like pallet present, blocked and starved conditions, and other station signals. In addition to voltage sensing other sensors like temperature and current can be on the same "1-wire" bus.

For your application look for an addressable bus system, and possibly RF communication between the buildings.

.
 
ptonsparky said:
Relay or not I have used a Sensaphone product that could easily be configured for contact or transducers. Easily connected to a network. Might be a bit time consuming for hundreds of units but what wouldn't.
Click to expand...

The Sensaphone unit itself will go into alarm if there is a power failure to the unit (and there is a good battery installed), but to do so still requires some kind of "relay" or "transducer" within the Sensaphone unit, which is why it has been said the OP's wish can't really be done other than very basic indicators
 
ptonsparky said:
Relay or not I have used a Sensaphone product that could easily be configured for contact or transducers. Easily connected to a network. Might be a bit time consuming for hundreds of units but what wouldn't.
Click to expand...

kwired said:
The Sensaphone unit itself will go into alarm if there is a power failure to the unit (and there is a good battery installed), but to do so still requires some kind of "relay" or "transducer" within the Sensaphone unit, which is why it has been said the OP's wish can't really be done other than very basic indicators
Click to expand...
Well I think he has now said a voltage sensing "relay" is OK, and if he can run phone wires to his panel for the Sensaphone, he can run any other kind of network for interfacing to the sensor as well.

Ucdohi, if there is already a BMS in place, I would just stick with that, no need to be duplicating efforts. But if there is no existing BMS, getting the likes of Johnson Controls, Honeywell or Siemens Builing Technologies involved will likely become an exercise in overkill because they will not want to do just the voltage monitoring, they will want the whole banana. There may be value in that too, but that's not the question you asked.

If you want to keep it simple and use COTS components, I suggest looking at a product line called Carlo Gavazzi, they have a very plain and simple 2wire network system called "Dupline" that is not terribly capable as networks go, but more than adequate for this task and can go long distances using simple unshielded bell wire.

https://www.gavazzionline.com/dupline.htm
 
Status
Not open for further replies.
Top