Help with Water Pump

[KeepItSimpleStupid]

But I'm lost-what is that smart little controller for?

It's a way of electronically connecting one of two power sources with near zero voltage drop. e.g. backup and regular power. The device connects the highest source. If the sources are "identical" in voltage within a small window, the sources can merge. e.g. Two 12 V 6A sources become one 12 A 12 A source. Advantages: near zero switching time and low power. The parts used are relatively new, I think.

It's building block.

 

[alec_t]

With the tide controller at four pumps, eventually the two idle pumps are bound to "flick" at the same time the running pair start. (The running pair will run for 30 sec, off 5 sec) The wave controller will individually turn two pumps on and off all day. The filter pumps will run nonstop with speed control. So it's not hard to imagine five pumps starting simultaniously, with two more pumps running.

Quite. How many power supplies will you be using for all those pumps, Joe? If it's only the two filter pumps which are critical and require battery backup then it might be simplest to run those two from their own power supply. The remaining pumps would then be run from a second (and third?) supply. Bear in mind that each pump draws just over 4A at start-up.

 

[KeepItSimpleStupid]

@Alec
I think that sounds like a good plan.

@Everybody
I think emergency lighting generally uses the trickle charge method. With trickle charging, the charging time is extremely long. Lead acid should be fine.

Fast charging uses algorithms dependent on battery chemistry.

EVERY battery chemistry requires it's own sets of rules. Some examples: NiCd has a memory effect. Car battery - not suitable for deep discharge, SLA - is suitable for deep discharge; LiPo, Li-Ion - think laptops and the recent airplane problem. They WILL explode if not cared for properly.

Better battery charges monitor battery temperature.

Proposals:
1_ Commercial charger and/or UPS
2. Trickle or the design presented by ronv , with undervoltage and temperature lockout

EXPECT to replace the batteries every 4 or 5 years even without use.

 

[salty joe]

Quite. How many power supplies will you be using for all those pumps, Joe? If it's only the two filter pumps which are critical and require battery backup then it might be simplest to run those two from their own power supply. The remaining pumps would then be run from a second (and third?) supply. Bear in mind that each pump draws just over 4A at start-up.

Then let’s plan on the filter pumps with their own power supply. I’ll use two more power supplies for the six other pumps. They are 15A power supplies.

EXPECT to replace the batteries every 4 or 5 years even without use.

As I read about SLAs, I did learn that the battery only last so long no matter what. So, I was going to get a low budget small 12V SLA for testing the system and get a big 12V SLA when the tank is ready to go.
Would this controller be used in combination with Ronv’s emergency battery backup, or instead of it? I’ve got the parts and would like to bring Ronv’s creation to life.
Thanks all.

https://www.mini-box.com/Y-PWR-Hot-Swap-Load-Sharing-Controller

 

[ronv]

Does anyone have a schematic of the 12 volt power supply? I think all the early pictures are missing and I would like to make sure we don't get something bad happening.
Joe, how often and for how long have you lost power in the past. I'm guessing you would rather kill the battery than the fish, but we should get close to worst case.

 

[KeepItSimpleStupid]

Would this controller be used in combination with Ronv’s emergency battery backup, or instead of it? I’ve got the parts and would like to bring Ronv’s creation to life.
Thanks all.

**broken link removed**

In addition too. It should give you a little longer battery life.

 

[salty joe]

Does anyone have a schematic of the 12 volt power supply? I think all the early pictures are missing and I would like to make sure we don't get something bad happening.
Joe, how often and for how long have you lost power in the past. I'm guessing you would rather kill the battery than the fish, but we should get close to worst case.

The power supplies I’m using have adjustable voltage and we are running at 20.1V. They did not come with a schematic.
We might have lost power a little over a day when Sandy hit, usually if we lose power, its half day or less. I still want to get a good sized SLA.

In addition too. It should give you longer battery life.

Very nice.

 

[ronv]

Hi Joe,
I was thinking of the 12 volt regulator. I think it was just a 7812 regulator but can't remember for sure. It is probably on the same page as the original timer logic.

 

[salty joe]

Hi Ron,
Yes, it is a 7812.

 

[alec_t]

@ronv
If you're concerned about 'where is the 12V coming from if the mains fails' that shouldn't be a problem as the logic/timing for the pump drive modules and speed control etc run happily from a lower voltage. I tested it at 9V in simulation.

Edit:
Of course, since only the filter pumps will have the backup, the (nominal) 12V for their control could be taken directly from the battery.

 

[ronv]

Yea, I guess it will be ok as long as the filter pumps have their own 20 volt supply. The speed control logic may try to run, but I don't guess that will hurt anything and the current is low.

 

[KeepItSimpleStupid]

@ronv

Explain? Confused.

1. All logic runs off a single 12 V supply which may dip much much lower during backup. I would suspect a 12 V backup battery might be useful down to 8 volts. This, I think, would mean that a 7812 would loose regulation and regulate to say 5V. I doubt the pumps would run?

2. Joe liked the 11.2 V level as being OK. Doesn't that mean that backup time would be horrible?

Doesn't it make more sense to:

1. Create a buck-boost or buy one off of ebay or aliexpress to take 8 to 24 in and make 20 V. This **broken link removed** puppy goes to 4 A for just the backup pumps.

Even this **broken link removed** one is quite cheap and quite respectable.

I've ordered stuff from the website. Aliexpress is just a HUGE collection of vendors offering stuff from China. You pay only one invoice, but each vendor ships separately. When you confirm receipt or have received the items in satisfactory condition, funds are released to the vendor. Feedback is used like ebay. Shipping times are long: 4-6 weeks. Most vendors seem to ship by China Post and the items have to be signed for and delivered by the USPS.

2. You wouldn't have to replace the 7812,, but we could replace it with a simple switcher, but I doubt it matters.

3. The largest 12 V drain on backup would probably be the alarms. Would it make any sense at all to make the alarms sound for an hour or shorter if on backup?

4. Reminder, that CMOS doesn't generally like it's inputs being driven when the power supply is zero. I'm not sure that happens anywhere.

@Joe
Just a generalization on backup

1. You need a battery/charger system. It's purpose is nothing more than to not overcharge the battery. Longer batter life come with sophisticated charging algorithms. Complexity is usually required when you want fast recharge times. battery temperature should play a role. I believe there is probably a specified undervoltage lockout value. This battery will drop in voltage UNLESS connected to a boost converter. I used the term buck/boost because I think you would really want to operate these gizmos from any voltage from 8-14 and 18-22 or really (8-24).

2. You need a power supply run off the AC mains which you have.

3. You need a means of switching the power from mains to battery. A way to do this is to energize a relay during AC and when that relay drops out the other source is selected. Issues with relays are: 1) There is an operate time and a release time of around 15-20 ms. The funky mini-box device makes the switching time nearly zero and there is a lot less wasted power because you don't have to energize the coils when on line power. There are ways (magnetic latching relay) that eliminates the needing power, but control is fussier.

4. I think we came to the conclusion that 24 VDC backup power is more complex to achieve.

 

[ronv]

Here was my thinking based on the circuit in post 1513.

The 20 volts would be supplied thru the large diode from the battery to +20. This gives us a diode isolated 12 volts for the 20 volt supply. The 12 volts would be supplied by the small diode from the battery to the 7812 output, thus taking the 7812 out of the circuit. The rest is just there to recharge the battery and keep it trickled. I think joe has a pretty good sized battery in mind. I think the voltage on a lead acid at 75% discharge is still 12.06 volts so even with a diode drop you still have 11.4 volts or so. I'm pretty sure the pumps run down to 10.8 if I remember the OEM supply numbers right.

 

[salty joe]

I think joe has a pretty good sized battery in mind. I think the voltage on a lead acid at 75% discharge is still 12.06 volts so even with a diode drop you still have 11.4 volts or so. I'm pretty sure the pumps run down to 10.8 if I remember the OEM supply numbers right.

I found a 70AH battery that was cheaper than a smaller battery from the same brand/same battery line . Must be a clunker in terms of sales, but it looked good to me because physical size is not an issue.
The pumps will run clear down to 3.5V. I took that reading from the +&- terminal points of the pump.
Thanks.

 

[ronv]

Seems like we should be good. It will take a long time to compleatly recharge a battery that big but since it is only for backup that shouldn't be a big problem.

 

[KeepItSimpleStupid]

Isn't Joe's measured 3.5 V an average voltage? The 3.5 would still contain peaks of 20V or so right, like PWM? You still get the torque with a 20 V pulse.

 

[salty joe]

If it makes any difference, that 3.5V reading was taken with the speed control in the low setting.

 

[ronv]

Didn't the OEM controller go as low as 10.8 volts?

 

[alec_t]

Didn't the OEM controller go as low as 10.8 volts?

That figure was for a speed less than max, but not for minimum speed, so doesn't apply to this situation.

 

[ronv]

I guess I still don't see the problem. Seems like the current limit also keeps the voltage below 12 volts and it starts ok. What am I missing?