Help trigger one solenoid valve with 2 different power supllies

Thread Starter

Pjota1

Joined Jun 13, 2019
18
Hey, I'm new to this forum. I came to the forum while searching for help to this question.
I have 2 fire alarm panels that should trigger same solenoid valve (24v low current like 125mA) in case of fire (one panel or another another, when in fire condition). Each fire alarm panels have a 24vdc output for that function (extinguishing release). Positive is always present, and each output sends the negative when extinguishing release condition occur.
I think I can do that with relays but I'm trying to make it easier with diodes due to space limitations.
I've polarized positive and negative from the extinguishing output in each panel with diodes 1n4007. When I try the systems, all seem to work... When one of the panels power the solenoid valve i don't measure 24v voltage in the return to the other panel that is not in alarm state.
But when I try to reboot the panel that is in alarm condition, triggering the solenoid valve, I get an earth fault in the other panel. When I reset the 2nd panel, I get earth fault in the first one... Very strange! Can anyone help me with these connections please?
Thanks in advance!
 

Reloadron

Joined Jan 15, 2015
7,501
Positive is always present, and each output sends the negative when extinguishing release condition occur.
When you say Positive is always present it leads me to believe the output of each unit is an NPN open collector which makes ground (Sinks Ground) when the output goes active or an On Alarm State. What I don't understand is how you worded "each output sends the negative when extinguishing release condition occur".

Anyway should the outputs be an NPN open collector type all I would think you need is a single 24 VDC relay with one coil side tied to either 24 VDC supply and the output of each unit going through a single blocking diode like a 1N4002 having Cathode going to output and both diode anodes tied to your relay coil high side. This assumes both grounds are common to each other.

This is purely a guess based on what I suspect you have. Since this is part of an alarm system involving possible safety of life all I am giving you is a suggestion to try.

Ron
 
Last edited:

MaxHeadRoom

Joined Jul 18, 2013
28,618
Normally a single coil can be picked up by two sources using a couple of single diodes, known as 'steering diodes'.
Max.
 
Last edited:

marcf

Joined Dec 29, 2014
288
Do you have a diagram for the ground fault detection circuit for the FACPs?

I know that FACPs have a high resistance connection to earth ground to the 24vdc supply in order to detect this condition, and are not typically connected directly to each other.

The panels that I have worked on typically had a greater than 2Meg resistance from the panel to earth ground.

I have never seen two FACPs connected together in this manner either.
 

ebeowulf17

Joined Aug 12, 2014
3,307
This feels like a long shot, but I think maybe I've found a leakage path through the protection diodes on coil 1 and coil 2.

Imagine one of the 24V supplies is higher voltage than the other - this leakage could cause the lower of the 24V supplies to be pulled up towards the voltage of the higher one. This in turn would pull the local ground of the lower voltage system upwards, resulting in ground path current leakage.

I'll give an example with exaggerated numbers: the two supplies operate at 25V and 23V. Even with two diode forward voltage drops at roughly 0.7V each, this leakage path would pull the "24V" leg of the 23V system up to 23.4V. Since the 23V system's supply still generates 23V, its ground gets pulled up to 0.4V.

In reality these numbers would probably be smaller and closer, and differences would appear as leakage current, not necessarily as measurable voltage differences. Nevertheless, I think the path might be problematic.
390FAC44-535C-4F15-AFAA-3BE5B775E8C0.jpeg
 

crutschow

Joined Mar 14, 2008
34,283
It it's a problem connecting the two grounds, then you may have to isolate with a relay on each output that switches both power and ground to the solenoid.
Is there any chance that both will try to power the solenoid at the same time?

Below is a connection that uses two DPDT relays, with only one battery connected to the solenoid, even if both relays are energized.

upload_2019-6-14_12-30-10.png
 
Last edited:

Thread Starter

Pjota1

Joined Jun 13, 2019
18
This feels like a long shot, but I think maybe I've found a leakage path through the protection diodes on coil 1 and coil 2.

Imagine one of the 24V supplies is higher voltage than the other - this leakage could cause the lower of the 24V supplies to be pulled up towards the voltage of the higher one. This in turn would pull the local ground of the lower voltage system upwards, resulting in ground path current leakage.

I'll give an example with exaggerated numbers: the two supplies operate at 25V and 23V. Even with two diode forward voltage drops at roughly 0.7V each, this leakage path would pull the "24V" leg of the 23V system up to 23.4V. Since the 23V system's supply still generates 23V, its ground gets pulled up to 0.4V.

In reality these numbers would probably be smaller and closer, and differences would appear as leakage current, not necessarily as measurable voltage differences. Nevertheless, I think the path might be problematic.
View attachment 179723
The strange thing is that I can trigger the coils (the independent one and the common) with each panels, not triggering the other independent one without getting earth fault. Earth fault only happens when i reset the panel removing the negative. I thought it was because first time I was not using protection diodes in the coils.
 

Thread Starter

Pjota1

Joined Jun 13, 2019
18
It it's a problem connecting the two grounds, then you may have to isolate with a relay on each output that switches both power and ground to the solenoid.
Is there any chance that both will try to power the solenoid at the same time?

Below is a connection that uses two DPDT relays, with only one battery connected to the solenoid, even if both relays are energized.

View attachment 179727
What do you mean by connecting th two grounds? Connecting the negatives from each panel by removing the diodes in the negatives? I didn't try that yet.
Thats not suposed that both panels try to power the solenoid at the same time. If that happens (it can happen) all extinguishing process will be ruined since it will open both directional valves(independent coils) and let the agent flow for two spaces (agent is released by the common coil).
 

marcf

Joined Dec 29, 2014
288
IMHO, I do NOT think this is a good idea.

Can you find an example of two FACPs ever being wired like this?
How would supervision for a fault condition work?

You should run this by your AHJ and at the very least consult your copy NFPA 72 before proceeding. I am sure that there are industry approved methods of accomplishing this and vendors that provide an approved device.

I have not been able to find one reference on the www as to the approved method of doing this.

Like the old saying goes "Just because you can do something....."
 

ebeowulf17

Joined Aug 12, 2014
3,307
What do you mean by connecting th two grounds? Connecting the negatives from each panel by removing the diodes in the negatives? I didn't try that yet.
Thats not suposed that both panels try to power the solenoid at the same time. If that happens (it can happen) all extinguishing process will be ruined since it will open both directional valves(independent coils) and let the agent flow for two spaces (agent is released by the common coil).
If opening both directional valves at once is unacceptable and delivers inadequate fire suppression, then this whole idea seems to be flawed. What if fires are detected in both zones at once? All valves open and neither zone gets sufficient fire suppression? It might seem an unlikely scenario, but I wouldn't want to bet people's lives on that.
 

ebeowulf17

Joined Aug 12, 2014
3,307
The strange thing is that I can trigger the coils (the independent one and the common) with each panels, not triggering the other independent one without getting earth fault. Earth fault only happens when i reset the panel removing the negative. I thought it was because first time I was not using protection diodes in the coils.
Hmmm, without knowing what the system looks for, when, how, and in what sequence, I can't even really guess why the error only registers at that time.
 

Thread Starter

Pjota1

Joined Jun 13, 2019
18
IMHO, I do NOT think this is a good idea.

Can you find an example of two FACPs ever being wired like this?
How would supervision for a fault condition work?

You should run this by your AHJ and at the very least consult your copy NFPA 72 before proceeding. I am sure that there are industry approved methods of accomplishing this and vendors that provide an approved device.

I have not been able to find one reference on the www as to the approved method of doing this.

Like the old saying goes "Just because you can do something....."
Yeah I understand what you're saying, but
If opening both directional valves at once is unacceptable and delivers inadequate fire suppression, then this whole idea seems to be flawed. What if fires are detected in both zones at once? All valves open and neither zone gets sufficient fire suppression? It might seem an unlikely scenario, but I wouldn't want to bet people's lives on that.
If opening both directional valves at once is unacceptable and delivers inadequate fire suppression, then this whole idea seems to be flawed. What if fires are detected in both zones at once? All valves open and neither zone gets sufficient fire suppression? It might seem an unlikely scenario, but I wouldn't want to bet people's lives on that.
The systems are designed and approved to work that way. We can only use this kind of systems when there's no probability of a fire at the same time in 2 areas. And adjacent areas must be have at least walls or doors with at least 60 minutes of fire resistance.
In the projects where can be or its probably to have a fire at the same time, we use independent systems.
 

ebeowulf17

Joined Aug 12, 2014
3,307
Yeah I understand what you're saying, but




The systems are designed and approved to work that way. We can only use this kind of systems when there's no probability of a fire at the same time in 2 areas. And adjacent areas must be have at least walls or doors with at least 60 minutes of fire resistance.
In the projects where can be or its probably to have a fire at the same time, we use independent systems.
Fair enough. Thanks for the clarification. Certainly not my area of expertise (obviously,) so I'll take your word for it.

As for the ground fault warnings, I'm still clueless. The two-relay solution from @crutschow proposed looks pretty bullet-proof to me, but if you're wanting a non-relay solution, or simply want to understand why your first attempt isn't working, I've got nothing.
 

marcf

Joined Dec 29, 2014
288
If there is no possibility that both releasing valves may be activated at once, why not just install two valves in shunt (parallel) and have each valve controlled by the two different FACPs and keep the FACPs isolated from each other?

It would probably also be a good idea to provide a zone on each FACP were made to monitor the status of the other to prevent any possibility of these two valves being operated at once. A 'pre action' scheme may work here.

I am sure that you agree that the original plan looks a lot like parallel wiring.

I was always made aware that you do not ever use parallel circuits in FACP initiation or notification wiring (or most everything else). "T" taps should not ever be allowed.
 
If you want small, you should be able to use OPTOMOS relays. You would have to design a PCB to use them. They are basically LED/MOS switches with typical b-directional outputs and low on resistance.

If these FAP's are far apart, then I would implement contact monitoring which would be more difficult. A cut wire would indicate trouble.

I did work on FAP systems, BUT my system generated toxic gas and hydrogen alarms, that, in turn activated the building alarm at high levels.

The fire alarm also shut down my system. I did not use a monitored contact, but if the entire cable were cut, I'd know about it. If one wire of the cable was cut, then it would not work. The panel had a light that indicated that when off indicated that the FAP was in alarm. When on, it indicated that there was a good chance for integrity of the cable or it was disconnected. We sometimes disconnected the FAP shutdown when our system was in a safe state and there was fire alarm testing happening.

Hydrogen and Hydride got the same treatment. I would have loved to have found a DIN rail monitored contact gizmo.
 
Top