Electromechanical circuit to alternate between basement sump pumps

[Arsene]

Hello everyone,

I would like to build a circuit, hopefully using only components that work off the regular household 110vac supply (relays, neon lamps, etc.) , which would do the following:

It would receive its power from a regular household AC outlet, and it, in turn, would power two basement sump pumps, both located in the same sump, submerged at the same depth (unlike the typical low voltage backup pump stacked configuration, which has floating switches at different heights). The circuit should alternate between pumps 1 and 2 as demand arises, thus extending the life of both pumps and also allowing for one of the pumps to act as a backup in case the other pump dies. In addition, a few lamps would provide the following feedback information: Two lamps, one assigned to each pump, would indicate which pump is currently active, whereas a second set of lamps would only turn on (and do so permanently) if either of the two pumps dies, thus indicating which pump needs to be repaired or replaced. A buzzer to provide audio feedback in case a pump dies would probably be a nice addition.

Can this be accomplished with a strictly electro-mechanical circuit that does not include a microprocessor or other electronic components?

Thanks in advance for any help

Arsene

 

[user_88]

...
There is a device called a flip-flop ... a low voltage DC .... 5 or maybe 12 V logic gate .... which would allow you to activate two relays, alternating between the two. There would only be one float switch, which would act as an input to the flip-flop. The one complication associated with the flip-flop is that you would have to provide a DC power source ... say +5 volts ... not much power required, but a part of the design that might conceivably fail at some future date. There is a component called a solid state relay ... should be readily available at surplus stores ... which takes a logic signal, the output of the flip-flop, and closes an AC relay ... to turn on an individual pump.
How much current do your pumps require?

The back-up logic would have to be separate from the alternating duty circuit. The back-up circuit would have to detect a no-operation fault with one of the pumps, and send a control signal to the remaining pump ...

What is the most likely failure mode of the pumps ... the electric motor, or something that is flow related such as a restriction of some sort? That is, how would you detect, in a control sense, that the pump is not working properly?

Are you going to require that both pumps be turned on at the same time ... if the occasion arises?

 

[Reloadron]

Actually what you are looking at is pretty common. I have a large scale system I designed here at work with my sketches dated 2005 and it still is working fine. I used a timer to alternate between pumps. The pumps are large 5 HP 480 volt pumps but the concept is the same. Also should a pump fail and the water reaches a set high level the other pump will run. Then there is the final dearly beloved level where an audible alarm sounds before we start threading water. Other than the timer it is all relay logic. I have them set to change over hourly but that becomes a function of how quick the sump needs drained.

A project like this can be as simple or complex as you want to make it and as inexpensive or expensive as you want to make it. The more whistles and bells, the greater the cost. I use contactors to turn the pumps on/off since you have 120 VAC pumps (I assumed) it just becomes a matter of the HP rating of the pump motors how you choose to turn them on / off.

Ron

 

[ADWSystems]

Here is an alternating relay https://www.electro-tech-online.com/custompdfs/2012/01/AlternatingRelays.pdf
and a time delay relay https://www.alliedelec.com/search/productdetail.aspx?SKU=70059690

I haven't put too much thought into this, so this is a little off the cuff. Use the time delay relay as the changeover trigger for alternating relay. You can find both in 110V versions. This would get you the alternating part of your quest. As for detecting a failure, I would need some more info. Such as, do the pumps run all the time?

 

[Arsene]

Thanks for your reply, User_88. You have given me some ideas, as I originally was thinking that both sump pump floating switches would be used, however, it seems a single independent switch to control the flip-flopping betwen the two pumps is the way to go, although I want to avoid low voltage, electronic components inasmuch as I can. To answer some of your quetions, oly one pumpt needs to work at a given time, as each pump has the capacity to handle worse case scenario by itself, however, I want to have a backup pump.

Thanks

 

[Arsene]

Thanks for the info, ADWSystems, specially the alternating relay.

No, the pumps do not run all the time. When it rains, and the water is collected around the house, it is chaneled to a small pit in the basement, where the pump sits. When the pumps floating switch senses the water level rising, it is activated, thus pumping the water coming into the pit (via a pvc pipe), up to ground level and then out to the roadside, away from the house foundation.

 

[Arsene]

Ron,

Thanks for letting me know the system I am looking to build is actually pretty common. I am brand new to this site, so forgive me if I commit a faux pas, but, would it be too much to ask for you to share your sketch to give me a solid starting point? I could hopefully then tailor it to my particular application to control a couple of common residential basement sump pumps.

Thanks

 

[Reloadron]

Ron,

Thanks for letting me know the system I am looking to build is actually pretty common. I am brand new to this site, so forgive me if I commit a faux pas, but, would it be too much to ask for you to share your sketch to give me a solid starting point? I could hopefully then tailor it to my particular application to control a couple of common residential basement sump pumps.

Thanks

No problem but I could likely draw you a basic schematic rather than have you work from my old drawings and revise them to what you want or need. Ironically I used the same timer back in 95 that was mentioned. It was a NCC (National Controls Corporation) Model: TMM-0999M-461. One of these with the data sheet taken from Allied Electronics where they retail for about $133 USD as can be seen here. However, that is actually a multi-function timer and overkill for what you need. The beauty of using off the shelf is ease in assembly and the timer and a few relays can easily be mounted on some DIN Rail.

Let me ask you a few questions so I can make a drawing for you and maybe a few more suggestions.

1. What is the HP or the current ratings of the pumps? I assume 120 VAC pump motors.

2. How do the pumps currently get turned on and off? I assume float switches? My industrial system used several level switches that you really would not need but I need to know what you have.

3. Give me a few ideas of exactly what you want.

I can have you a decent enough drawing in a day or two.

Ron

 

[Arsene]

Ron,

Thank you so much for your help. To answer your questions in order,

1. What is the HP or the current ratings of the pumps? I assume 120 VAC pump motors.

It will be two (2) Zoeller residential basement sump pumps, each working off 110 vac, should be drawing under 10 amps each.

2. How do the pumps currently get turned on and off? I assume float switches? My industrial system used several level switches that you really would not need but I need to know what you have.

Currently I have only one pump (no backup!). The pumps comes from the factory with their own float switch (you can see one here https://www.zoellerpumps.com/ProductOtherPartInfo.aspx?ProductID=92)

3. Give me a few ideas of exactly what you want.

What I need is to, basically, build a box with one 110vac input (connected to a wall outlet) and two 110 vac outlets, one each for the two pumps. As the water rises in the basement pit, both pumps will try to engage at the same time, as both their switches will actuate, however, I want this box to cut off power to one pump and only provide power to the other pump. When the incoming water has been removed by the one working pump, the pump should shut off, and, when the water level in the pit goes up again, it should be the other pump that goes to work, therefore, during the time water needs to be removed, only one pump works, while the other is inactive, flip flopping between the two pumps.

In addition, if one of the two pumps has a malfunction and dies, the circuit in question should keep power to this one pump all the time until the failed pump is replaced or repaired, at which time there should be a way to reset the circuit to resume its flip-flopping operation.

Perhaps it would make matters easier if I removed the float switches from both pumps, and instead install a single float switch connected directly to the controlling circuit I will be building, so the circuit, which will be feeding power to each pump alternatively, has water level information from this single floating switch to make its decisions. Well, it is just a thought.

Thanks again.

Arsene

 

[Reloadron]

I agree with your last sentence.

Tomorrow, depending on work demands I'll give you a circuit and some ideas and we can go from there. Worst case over the weekend as I am off Friday so a 3 day weekend leaves me time. I will say this, alternating from pump to pump as you describe would not be a simple as doing a timed system. What I mean is during a period of for example an hour any demand for pump would be the same pump. The following hour it would be the other pump and so on. All in all the run time of each pump becomes a wash (no pun intended) as in the long run they will have close to equal run time.

How deep is the sump?

Ron

I have a similar pump in my basement sump. I just have a spare with a high water alarm.

 

[ADWSystems]

Now that I've had some time to dwell on this and the additional info you've provided, try this circuit on for size.

Somewhat of a "modular design". The first module utilizes two sets of contacts to interrupt power to the alternating relay. The second module controls the power to the pumps.

1) 110VAC Main
2) Pump failure relay #1 contact
3) Pump failure relay #2 contact
4) Alternating relay to energize each of the pump select relays

This part will keep the relay from alternating if a pump has failed.

1) 110VAC
2) Float switch
3) Split to both Pump #1 and #2 contacts

This will use a single float and only run the pump selected by section #1.

The one aspect I've not figured out is how to detect a failure. But given a way and the output is relay contact able to handle 110VAC.

 

[Arsene]

Ron,

Time is not of the essence, as right now, in the middle of the winter, with no rain, and therefore no pump operation required, I have two or three months time to complete this project before the spring monsoons start in the midwest.

The sump is 26 inches deep (below basement floor level), however, the pump(s) sitting at its bottom must contend with a column of water pretty tall, as, from the bottom of the sump to the ground level is 11+ feet.

A timed system would be fine, as long as, if one of the pumps dies, the system is able to determine the other pump needs power 100% of the time until the dead pump is replaced and the timed operation can resume.

However, how would a timed system manage the following two situations?

(a) The currently active pump dies while in the middle of pumping water - or
(b) The fixed time window assigned to the currently active pump ends while the pump is active and there's still standing water in the sump?

Lastly, you mention you do have a spare (backup?) sump pump in your own basement, with a high water alarm. This setup, which I have seen before, usually includes a second, 12 vdc pump sitting higher than the primary ac pump, and driven off a battery (in case power fails), however, I have yet to see one of these smaller dc pumps that can handle the volume of water required during a really bad storm, something the ac models do easilily. And also, the pump sitting higher up never goes into operation until either the primary pump fails or the utility power fails, so, unless you are thoughtful enough to test it every so often, you will never know its real working status until it is called into action, at which time you might be away. With the flip-flopping system I propose, both pumps are tested all the time, and, if one ever failed, not only would you have a backup pump, but also a lamp, or an audible signal, or both, designed right into the system would let you know you are down to just one pump.

Thanks

Arsene

 

[Arsene]

This proposal kind of reminds me of OOP (Object Oriented Programming), where you design objects (or modules) each with their own functionality, which in turn interact with each other.

Interesting idea, ADWSystems, perhaps breaking down the system into smaller components is a way or the way to go.

Thanks

 

[ADWSystems]

However, how would a timed system manage the following two situations?

(a) The currently active pump dies while in the middle of pumping water -or-
(b) The fixed time window assigned to the currently active pump ends while the pump is active and there's still standing water in the sump?

Actually this will provide a two for one. With the timing and alternating relays, the other pump will start running when time expires. The system I proposed originally would actually alternate the pumps while they were or weren't running. It wouldn't be real effective at equalizing the run time. A simple tweak to make them alternate based on run time would be to setup the circuit:

1) 110VAC Main
2) Float switch split to
3a) Timing relay which is in turn connected to the alternating relay input
3b) Alternating relay output #1
3c) Alternating relay output #2

Now the pumps will only alternate when running, and will keep alternate as long as the float switch is good.

Between each output of the alternating relay and the pump, add a contact to shut off the failed pump and another contact to bypass the alternating relay and engage the good pump.

I've seen two options to wire the alternating relay outputs when driving single phase loads. You can run the pump directly or connect to a contactor or relay that then supplies power to the pump. I prefer the second as it makes a more universal solution. It allows the use of the same timing relay even if the motor/pump size or voltage changes. Just change out the contactor rather than re-spec'ing the timing relay.

I hope you can visualize the circuit. If I have time, I will try to draw it out.

P.S. The timing relay I linked to was from Ron, I could not quickly find the part number for the one I use, but the theory is the same.

 

[Reloadron]

Hi Arsene

Before I forget as to a failed pump. My experience outside of mechanical failure is pump motors just like most motors fail with shorted windings and generally open a circuit breaker. On that note all sump pumps should be on a good quality GFCI Breaker. A pump motor can also fail open circuit and this sort of failure can go undetected until one opens the basement door and says holy **** where did all that water come from. This is where I like an independent mounted float switch used to sound an alarm, then too if you are not home, you are pretty much screwed. More on all that later. A shorted pump will open the breaker and an open pump will not.

Projects like this can get pretty complex and before you know it loaded with electronics. For example (and I won't propose this) let's say a pump under normal operation draws about 7 amps after start. I could monitor the current to each pump using for example a Window Comparator. If the current when the pump is supposed to be running exceeds maybe 8 amps it could be a early sign of bearing failure or if the current is less than 6 amps or no current the pump failed open. High water alarm? How about if the water exceeds a high water level beyond audible I send a signal to one of my always on computers (they run on UPS during power failure till the emergency generator kicks in) and it sends me a text and email to my cell phone telling me I am in trouble. Then I call one of my neighbors or whatever if I am out of town. While I am at it I can wire other things into the computer to call me, flood, fire, smoke, whatever. The list goes on and on. Actually "phoning home" as I call it is pretty simple but other things get complex.

Yeah, I keep a spare pump on a bench in the basement. It is just a new in box pump. I rely on the audible alarm and the whole house has backup generator power so I don't worry about power outages. I have seen the really cool battery backup pumps. Pretty nice. The possibilities become endless so one decides when enough is good enough.

(a) The currently active pump dies while in the middle of pumping water - or
(b) The fixed time window assigned to the currently active pump ends while the pump is active and there's still standing water in the sump?

In the case of B that pump would shut down and the other come on till the sump was empty. A gets a little more complex but generally a high level switch would look to turn on both pumps. I'll cover that more in detail with a drawing and as time goes by we evaluate what is best for you as well as practical.

I also like and am following the suggestions of ADWSystems. Sometmes it is a matter of combining the best of all suggestions.

Ron

 

[ADWSystems]

I whipped up a quick schematic to show what I was thinking.

Please excuse some of the symbology. There aren't a lot of industrial controls symbols in Protel. You'll notice a pushbutton switch representing the normally open/low level in the sump. The DPDT relay symbol used to represent the alternating relay isn't quite right either. The coil of the symbol used for the alternating relay is supposed to represent the alternate input of the device.

The pump failure relays indicate a failed state when denergized. When the pump is good the relay is on. When the pump fails, the relay deenergizes and the NC contacts make and bypasses the alternating relay to start the other pump. It is up to you to determine how to detect the failure and drive the relays. Followed by the pump run relays/contactors and finally the pumps.

Small schematic, but I expect you'll need to shop around for the "most effective" parts (ie., the cheapest place to buy them).

 

[Reloadron]

Attached is a basic version of what I would start with. As I mentioned, the possibilities become endless as to added features.

Ron

 

[ADWSystems]

Attached is a basic version of what I would start with. As I mentioned, the possibilities become endless as to added features.

Ron

Too funny. You did the same thing I did originally. With the timing relay connected directly to the mains it will alternate even though the float is low. I don't think that will provide the most balanced run time. But that's also relative to houw close in run time you want them to be. Some of the poeple I work for get to be real sticklers about it and they want it within 49-51%. I do like you method for alternating the run time by using RY1. That removes the (expensive) alternating relay while still allowing the control relay (RY1) to be selected based on the motor requirements, and keeping the load on the timing relay contacts small (extending MTBF).

If TR1 was after the float and in parallel with the RY1 contacts to neutral, the pumps would only alternate while the float was high (ie., pumps requested to run).

 

[carmusic]

i would just use a programmable relay like those ones **broken link removed**
they are cheaper than specialized relays and very easy to set-up and modify. you just connect your inputs to it and your outputs than make a little program. Timers, counter relay are out of date and most of the time are replaced by those type of PLC in many industries.

 

[Reloadron]

Good link there carmusic, sort of a PLC called a PLR. Yeah, I agree it could be a viable solution. Just a matter of the programming. When this does that then do this. You would still need the relays but would use low voltage coils and switches. As to bang for the buck those PLRs are pretty nice.

Overall I figure it this way, it's like I tell my boss when he begins adding whistles and bells to my designs. Mike, you can have anything you want, write the check. The OP can have anything he wants, just a matter of looking at needs / nice to have and what the project should cost.

I may have to look deeper at those PLR systems, they do seem pretty cool for a large number of applications. Thanks for that link.

Ron