Need to keep a relay coil powered for a short time on loss of coil input voltage. (grid failure)

Thread Starter

jeffy104

Joined Jun 11, 2020
14
I am putting an automatic backup system together to power a small cooling pump for thermal solar panels when the grid goes down. The pump moves glycol in the panels to heat exchangers in the basement and the thermal energy is stored in insulated tanks. When the power fails, the glycol quickly reaches 400 degreesF and is ruined- and lost through pressure relief. I’m using a RIB2401D relay to switch between the grid and an inverter. The 2401D is great for disconnecting from the grid but when it comes to bringing the inverter on line, it’s too fast. I need a delay in the relay that brings the inverter on line and it needs to do it’s switching well after the power goes off. The delay will eliminate the possibility of two ac sources being in the same place at the same time and will give the small computer running the pump a chance to consume any voltage inside itself so that when it powers back on with the inverter, it does a normal boot-up. I have both 120vac and 24vac available for powering coil relays. I thought that a capacitor would be able to do the job but don’t know how to go about setting that up. I don’t have any background in electronics and thought this would be the best place to find some help.
 

crutschow

Joined Mar 14, 2008
25,132
A capacitor won't work to delay AC but it can with DC so you would rectify the AC to get DC and then use that to power a DC coil relay.

Do you care if the relay has a delay for both on and off, or do you want the delay only for when it turns on?
The circuit is simpler if both have a delay.
 

Thread Starter

jeffy104

Joined Jun 11, 2020
14
Thanks guys for the replies. I don’t think it matters if it delays on make but it has to delay on break (I think). I need to have this relay coil energized for a second or so after we have a power failure and the lights go out. I thought that if a capacitor was in the circuit, it would store enough power to keep that coil holding the contacts in the position they were in before the lights went out. I didn’t know Walmart sold those type of things but I think those delay timers only work when the power is on. I keep wondering if there is a delay relay out there that has a spring loaded gear-driven contact set in it. On my old wood boiler there was such a device which controlled the air intake and when the power went out, this device would start whirring and close the intake baffle to put the fire out. It was actually a type of motor. I need a way of powering a relay for second when the power fails or a relay that moves slowly when the coil loses power.
 

andrewmm

Joined Feb 25, 2011
403
Hi

A suggestion,

My panels are on the roof, above the attic space as us Brits call it.

on the other side of the roof, the one that's in shade,
I have a black radiator, that's higher than the solar panels.
in between , I have a thermostatic valve, pure mechanical,
that opens if the top of the panels gets above about 120 degrees C ,
then a natural convection loop is formed, which dumps the excess heat when we have power fail,

Still gets dammed hot, and the expansion tanks get up to a good pressure, but stops the system boiling.
 

crutschow

Joined Mar 14, 2008
25,132
You can rectify the 24Vac with a diode and a capacitor to ground.
This will give about 34Vdc so use a relay with a 36Vdc coil.

The delay time will depend upon the relay coil resistance and the size of the capacitor.
For example, a 270μF cap with a typical 36Vdc relay coil resistance of 3.5k will give a turn-off delay of about 1 second.

LTspice simulation is shown below:
It shows the relay output (red trace) opening about 1 second after the AC stops.
Due to variations in the actual drop-out voltage of the relay, you will likely have to experiment with the value of the capacitor to get the desired delay.

1591971561303.png
 

Thread Starter

jeffy104

Joined Jun 11, 2020
14
Ok this is genius! This will make things work properly. This is great! I’m more accustomed to checking voltage and continuity and basic wiring but I’m excited to get this project together. Can you suggest a place for me to get the items I need to put that together? I’ve never assembled/tested a circuit using these kinds of components but I believe I can get some help with that. A dpdt relay will only make/break 2 conductors which is fine for the disconnect from the grid but on the inverter side I need to make 3- two outputs from the inverter, plus a ground-to-panel connection on what will be the neutral output. My panel is bonded. Those three connect after the delay so I can’t back-feed to the grid. Am I looking for 3ptt with a 36Vdc coil? Please tell me someone makes that in 120V. My disconnect-from-grid relay is the RIB2401D. I really appreciate you creating that circuit and the simulation is amazing. Thanks!
 

Thread Starter

jeffy104

Joined Jun 11, 2020
14
AHHHH.....I was so excited about your answer that I forgot to think about the power coming back on. O man. I need to think about that. Let’s see- the 3ptt can get energized right away, disconnecting the inverter. I believe the RIB2401D on the grid side needs to be on a delay. Yes, that sounds about right. Right? O boy. I have a sketch that I’ll put up.
 

crutschow

Joined Mar 14, 2008
25,132
Please tell me someone makes that in 120V.
What's "that".
Are you referring to the coil or the contacts rating?

Certainly if you have a lot of contacts to make or break you can use more than one relay, but 4PDT types are available.

All the components can be purchased from an electronics supplier such as DigiKey, Mouser, or Farnell.
 

Hymie

Joined Mar 30, 2018
837
I think when the OP is referring to a 120V coil, they mean one that could be powered directly from the mains failure.

Although DC relay coil ratings above 48V are available, they are not common.

I would suggest you rectify the 120V mains (giving a peak of 170Vdc) and then use dropper resistors to reduce this to 48V. You could then include a capacitor to hold-up the 48V relay coil voltage for a second or so, after the mains power is lost – giving you what you want.
 

Thread Starter

jeffy104

Joined Jun 11, 2020
14
@ Crutshow the “that” is a relay with 120V contact rating with 36Vdc coil. I had a quick look at relays last night and they seemed to be for automotive applications with 11 pins, so there would be a base with that type too. I’ll check w/digikey as they were able to supply me with the in-line 120V fuse holder I need. That was not easy to find with 12AWG.
@Hymie yes the input for the relay coil could be from the panel and built as you say. Unfortunately, I’m not able think as you guys do because I don’t have the education or experience. You guys have forgotten more about circuits than I will ever know but now I have one more problem to solve. I’m having a little trouble picturing it.
Relay #1 disconnects from the mains on mains failure with no delay. Relay #2 connects inverter to load on mains failure with a delay. Relay 1 connects to mains on power restore with a delay. Relay 2 disconnects inverter with no delay. I could be thinking incorrectly about this because I don’t know what the component capabilities are but I need to disconnect and connect at different times to prevent back-feed and equipment damage. Thanks again......making a new sketch to put up shortly.
 

Thread Starter

jeffy104

Joined Jun 11, 2020
14
I drew that with all contacts open. The ac receptacle in the upper right of the drawing is only used when I leave the house and need the backup. Normally the computer/pump is plugged into another receptacle with no special wiring. This keeps me from having to run the pump on the inverter with the charger running as I only want to use this when needed. Rather not be relying on the charger, batteries, and inverter all the time.
 

Thread Starter

jeffy104

Joined Jun 11, 2020
14
IMG_9217.JPG

This picture shows the small computer (green screen) running the pump (just below pressure gauge) on 6/13/20 and the outside temperature is in the 40’sF. System is running close to 150F and keeps the house nice and warm. It’s works so well until the power goes out.
 

crutschow

Joined Mar 14, 2008
25,132
Relay #1 disconnects from the mains on mains failure with no delay. Relay #2 connects inverter to load on mains failure with a delay. Relay 1 connects to mains on power restore with a delay. Relay 2 disconnects inverter with no delay.
So sounds like you need a relay delay on make (relay operating) and none on break.
That's doable but a little more complicated.
Let me give that some thought.
 

crutschow

Joined Mar 14, 2008
25,132
How about using one DPDT relay to switch between the two sources for your pump?
It could have a 120Vac coil powered from the main's voltage with the NO connects also connected to the mains and the NC contacts connected to the inverter output.
The wiper connections go to the motor.
No delay needed.

Or why not just run the pump from the inverter all the time and just let the charger provide the power when the AC mains are on (or does the charger not output sufficient power to continuously power the pump?)
 
Last edited:

Thread Starter

jeffy104

Joined Jun 11, 2020
14
"The 2401D is great for disconnecting from the grid but when it comes to bringing the inverter on line, it’s too fast. I need a delay in the relay that brings the inverter on line and it needs to do it’s switching well after the power goes off. The delay will eliminate the possibility of two ac sources being in the same place at the same time and will give the small computer running the pump a chance to consume any voltage inside itself so that when it powers back on with the inverter, it does a normal boot-up.” I know I could be wrong about 2 sources of ac being in the same place because of the physical characteristics of a particular relay, but I need to have that delay for two reasons other than that. And I don’t really know what length delay I’m talking about. 1/10 of second, maybe 1/2 or 1 full second. First because of not being able to match the sine wave, which would't be an issue if things were slowed down. Secondly, I need to stop the flow of power, give the computer a chance to run out of power, and then reboot from a complete stop. I really don’t like the idea of my (pricey) solar energy system depending on a charger, batteries, and an inverter- that would be ok for an emergency. So far this year, the power has been off 3 times. The sun was out the last time it went off and it was around 9 am. I learned about the failure while I was away from the farm and quickly headed back to fire up the generator. It took less than 2 minutes for me to get there. I checked the collector sensor temp on the computer that runs the pump (generator running) and it was approaching 200F so I knew I got there without much time to spare. This whole business of being tied to this solar panel thing like a dairy farmer tied to his cows is a real drag, and the power goes off here all the time. I must admit I got a bit lazy after my last post and thought, this is a lot of work thinking about this while knowing at the same time that I don’t have the knowledge to not be second guessing myself all the while. Am I getting a headache? LOL You are the first person to let me know that it’s possible to make this work the way I want it to. I’m going to keep scheming here and when I come up with something, I’ll post it to see what you think. Awful busy here on the farm but this is something that I’d like to get out of the way this year. Thanks again for all your input!
 

crutschow

Joined Mar 14, 2008
25,132
will give the small computer running the pump a chance to consume any voltage inside itself so that when it powers back on with the inverter, it does a normal boot-up.”
Why does it need to power down and boot back up?
First because of not being able to match the sine wave, which would't be an issue if things were slowed down.
I don't see that as a problem.
Why do you?
The delay will eliminate the possibility of two ac sources being in the same place at the same time
A single DPDT relay would do that (below).
The contacts break one connection before the other is made (break-before-make).

1592147865504.png
 
Last edited:

Thread Starter

jeffy104

Joined Jun 11, 2020
14
The computer can’t boot unless it’s powered down completely. With all the money tied up in the system in the basement, (much of it grant money- not mine) I don’t want to mess with the power supply to the computer, unless it’s to shut the power off (the delay) and then power it back up. I just don’t think it’s a good idea to rapidly unplug and plug back in any of my computers. They’re just not made for it. Not going to take a chance of having the computer go into fault mode or some such thing and then have it sit and not be able to run the pump while the fault codes display with nobody there to see it while the panels quickly head for 400+ degrees F. I suppose that a pump with no computer to run it would make things a LOT easier. I wish this system was simpler. The computer handles a lot of jobs related to the system- something on the order of 20 or so settings on it. I remember setting parameters to my system conditions but that was last year and It took some time to do. I see the rapid switch between two ac sources as a problem after reading about it where sensitive things like computers are involved. It was an article about the sine wave of ac and they talked about one sine wave being 180 degrees (or more- or less) out of phase with another as the problem and the possibility (or guarantee of) of a voltage spike. And they talked about certain motors not coasting to a stop before being re-powered as being a problem- I don’t believe that’s a problem in my case. Trouble is, with my limited understanding of things of this nature, I have to err on the side of caution. I’m glad you included the description of “break before make” in your post. I remember discussing that with a tech at the relay company- but we also talked about the elapsed time between break and make and how this amount of time can be a problem in certain applications. The company is “Functional Devices” and the guy I spoke with seemed to be happy to talk about things as long as I kept asking questions. They make the RIB2401D. I’m going to spend some time tonight thinking about this and going over my drawing- and by the way, your schematics are pretty cool. I even printed the first one out to keep with my drawing. I’ll print your latest as well as a reminder. I bookmarked a site where I can drag and drop things to make a circuit but no time to spend learning about it. I admit that this is a lot to take in for me and I appreciate your patience and the fact that you are looking at my problem. I also acknowledge that there is the distinct possibility that I can over complicate this project. One of the things hindering my moving forward on this in an organized fashion is the fact that I have A LOT going on here and it can be difficult to find even a small window of time to work on this. Consequently, I lose my train of thought, forget things- however you want to call it. Interruptions are constant. When I do find the time, 2 hours can go by and it seems like only about 20 minutes. I find the subject fascinating. The old saying......”Too soon too old. Too late too smart” comes to mind. I hope to post again soon. Thanks.......Jeff
 
Top