Underwurldes - Diy Auto Top Off Project

[sophos9]

Right, I want an auto top off kit however cannot afford the ones in the shop so I've set about building my own...

The idea is to have coloured LEDS indicating status, so green for powered, amber for topping up and red for manually switched off or a switch out (is it possible?)

Here is the plan, I'm OK with basic electricals - basically, I'm looking to run a 12vDC around the 2 switches into a relay which will activate a mains powered Eheim 300 Compact Pump, I want to leave it switched on all the time so it gets little and often. I also want a switch to deactivate the level switches for water changes so I've incorporated one it

This is the rough plan, I've ordered all the parts and will do a comprehensive "how to"



So for all the sparkies out there, can you help on how to get the LED's to correspond to the requested state?

Many thanks

 

[pengy666]

I wouldnt bother with fancy electrics, all you want it do is top up water not the christmas decor

 

[sophos9]

I wouldnt bother with fancy electrics, all you want it do is top up water not the christmas decor

Dont think they are fancy, if I wanted to go fancy I would got for a PLC controller... Visual controls are good especially when you are me!!

 

[Musho3210]

ATO's are awsome, they are 100% better than manual top off since they do it in short frequent bursts of water topping off and they keep the level constant all the time for skimmers as well as keeping salinity in check. Mine probably comes on 3-4 times a day opposed to me dumping water in once a day, what do you think is more stable? Its not about laziness and activeness, they are a great thing to have.

 

[sophos9]

ATO's are awsome, they are 100% better than manual top off since they do it in short frequent bursts of water topping off and they keep the level constant all the time for skimmers as well as keeping salinity in check. Mine probably comes on 3-4 times a day opposed to me dumping water in once a day, what do you think is more stable? Its not about laziness and activeness, they are a great thing to have.

Hey mate, thats exactly true. I'm looking to decrease the variance of SG swing in manually topping up for a system that narrows it - oh and it takes pain in the ass job out of the way to.... May included an additional pump to mix kalk in the same top up

 

[Musho3210]

Kalk in contact with the main pump can shorten the pumps lifespan due to the calcium content in kalk. You can somehow add a calcium dispenser on which doesnt require the kalk to be in contact to the pump but im not sure how....

 

[SkiFletch]

Well, that's almoast right. What you want is an AC relay with DC control. You wire mains->powerhead->relay (AC side)->mains. Then you wire 12V+->Float switch->Float switch->deactivate switch->relay (DC side)->12V-

LED's become more complicated... Usin this method you would have only one LED on the DC side (since A/C LED's are hard to find). Would tell you when the topoff is working if you set it up anywhere on the DC side.

If you want to incorporate an LED for manual off, you'd need to impose a second relay.

 

[sophos9]

Thanks for the reply ski, I've been doing some digging...

The green led can go across the 12vDC circuit, the amber can go across the coils on the relay. To get the red to work the switch I need is a double pole, double throw switch so connecting across the poles will power the red LED and cut the 12v circuit.

Did not order 12v LEDs so gotta work out what ohm resistors I now need

What you think about safety aspects, if the relay fails short circuit then that will create problems - any ideas of a way around it?

PS, the coil numbers to not represent the actual coil numbers at the minute!!!

 

[Underwurlde]

1K, 1/4W resistors. BTW, your system lacks hysteresis.

 

[sophos9]

1K, 1/4W resistors. BTW, your system lacks hysteresis.

I lack a brain

Are you talking about resistor location for the LED's?

Would a deterministic system be better in this instance (not a challenge, a genuine question to get better understanding!)

Please exapand on why resistors and point the location of them please

Many thanks

 

[Underwurlde]

Your question was:

Did not order 12v LEDs so gotta work out what ohm resistors I now need

My answer was:

1K, 1/4W resistors.

Put another way then, you need 1000ohm (1 KiloOhm, i.e. 1K) reistors with a rating of about a quarter of a watt (1/4W) - (power rating for that current not really important TBH, but if you were asked...) Position of the resistor not an issue, as long as they are placed in series (in line) with the LEDs. Note: LEDs have 'direction'.

BTW, your system lacks hysteresis.

What I mean by this is consider when the level of water is exactly at that level such that the float switch just turns on - the pump goes on, the water surface ripples, the float switch goes on/off/on/off/on/off........, your pump goes on/off/on/off/on/off...........

I'd therefore consider a timer to keep the pump on (regardless of float switch setting) for some time after the float switch initially closes. Perhaps these float switches have this built in (hysteresis, not a timer) - I dunno, just making you aware of things you may not have thought of in your design.

Andy

 

[sophos9]

Put another way then, you need 1000ohm (1 KiloOhm, i.e. 1K) reistors with a rating of about a quarter of a watt (1/4W) - (power rating for that current not really important TBH, but if you were asked...) Position of the resistor not an issue, as long as they are placed in series (in line) with the LEDs. Note: LEDs have 'direction'.

Thanks, appreciate the help

What I mean by this is consider when the level of water is exactly at that level such that the float switch just turns on - the pump goes on, the water surface ripples, the float switch goes on/off/on/off/on/off........, your pump goes on/off/on/off/on/off...........

I'd therefore consider a timer to keep the pump on (regardless of float switch setting) for some time after the float switch initially closes. Perhaps these float switches have this built in (hysteresis, not a timer) - I dunno, just making you aware of things you may not have thought of in your design.

Great and please keep them coming, these are the points that really add value! I totally agree with you, I dont know how they will work out but thinking if they are in series then they will both have to be closed to complete the circuit, this *may* eradicate the inconsistency of using just one.

What timer would you use and can you advise on how to wire it into the circuit (555 timer?)

Thanks Andy, appreciate the help

 

[Underwurlde]

I dont know how they will work out but thinking if they are in series then they will both have to be closed to complete the circuit, this *may* eradicate the inconsistency of using just one.

No, the lowest one will be the 'fiddler's elbow'. (Q) Why two float switches?

Yes, you could use an NE555 timer arranged as a monostable (set it to 20 secs?). The trigger input will come from the float switch output, the output from the monostable will replace the output from the float switch (thus keeping the relay on for a period of time after the float switch triggers it) - ensure the 555 timer has enough oomph to drive the relay of course. You'll need a reverse protection diode across the coil of the relay, otherwise the back EMF will blow any sesitive logic (the 555).

Personally I'd go with what you've got because as you see things can get complicated very quickly. Forewarned is forearmed, so you'll know what's going on if your pump / relay suddenly starts chattering away when it comes to refill time. Thinking about things a bit deeper, the ssyte,m as you have it now may have inherent hysteresis because of the time it takes for the pump to start pumping water and, more importantly because of this 'inertia' if you like, the time it takes for the pump to turn off: If the pump takes time to turn off (for the flow to stop) then the tank is filling up with water, thus preventing the float switch from 'hunting'.

I'll shut up now.

Andy

 

[sophos9]

(Q) Why two float switches?

I was thinking by wiring them in series you would have to have both switches activated before the pump would start - that should do away with the issues of a snail sitting on one and keeping the pump activated, one failing closed etc etc

Yes, you could use an NE555 timer arranged as a monostable (set it to 20 secs?). The trigger input will come from the float switch output, the output from the monostable will replace the output from the float switch (thus keeping the relay on for a period of time after the float switch triggers it) - ensure the 555 timer has enough oomph to drive the relay of course. You'll need a reverse protection diode across the coil of the relay, otherwise the back EMF will blow any sesitive logic (the 555).

This sounds like a great idea, the other posts I have read have experienced a chattering issue so I would really want to pursue this, I know I'm asking a lot but is there anyway you would be able to whip up a quick wiring diagram?

I'll shut up now.

Dont do that fella!!

Many thanks

 

[Underwurlde]

I know I'm asking a lot but is there anyway you would be able to whip up a quick wiring diagram?

It'll take me a while, but I'll see what I can do.

Andy

PS, is there anything else you want it do? I was thinging overflow protection. How would you make it failsafe? (an overflow pipe is the obvious one).