Help with Water Pump

[salty joe]

I am glad you guys are safe. We lost power for only a day-some friends one town away are still without power. What a huge storm. I'm gonna donate to Red Cross to help them help the people that got hit hard. The fires and flooding and falling trees killing kids-it's all so sad.

 

[4pyros]

falling trees killing kids-it's all so sad

I did not here that one. That is sad.
How are the fish, how long can thay go without power.
Should we add a battery back up option?
With the low voltage pumps and PWM control you could run at a reduced rate for quite sometime.

 

[()blivion]

The fish themselves should be relatively fine without power for quite a while. It's the living rock (coral) that will have problems. As I understand it coral need consistent and unending water agitation or they will go belly up pretty quick.

In any case, a battery backup would be a good option all around. I recommend lead acid car or motorcycle batteries. Easier to deal with electronically, and you don't have to navigate shipping on Li-X. (which apparently USPS won't ship anymore.) Another option is NiCd tool batteries. But it's harder to find something like that in the volume you would need in your local stores. Lead acid batteries are far more common in that respect. You could use laptop or RC batteries, but I wouldn't. They generally need special attention to charging and discharging that is a PITA to DIY. Layers and layers of safeties that lead acid or NiCd simply doesn't need from the start.

 

[kinarfi]

Thanks guys. Humidity is what worries me..

I may be very late with this suggestion, DOW CORNING® 734 FLOWABLE SEALANT, **broken link removed**, I've used this on circuit boards for a motor cycle that were exposed to the weather and it lasted for 'ever' as far as I know. Basically, I coats the board and components with about a 1/16 " of silicone and the rest flows off.

 

[4pyros]

In any case, a battery backup would be a good option all around. I recommend lead acid car or motorcycle batteries. Easier to deal with electronically, and you don't have to navigate shipping on Li-X. (which apparently USPS won't ship anymore.) Another option is NiCd tool batteries. But it's harder to find something like that in the volume you would need in your local stores. Lead acid batteries are far more common in that respect. You could use laptop or RC batteries, but I wouldn't. They generally need special attention to charging and discharging that is a PITA to DIY. Layers and layers of safeties that lead acid or NiCd simply doesn't need from the start.

I would just use SLAs they are cheap and readly avalible.

 

[kinarfi]

I am at the end of this thread again, which is where I started, and I have an idea that may be out of the box, may be completely impractical, but what what if you use one 115 vac centrifugal pump that can deliver all the water you could possibly need with 4 valves controlled by servo motors. Centrifugal pump do not mind being dead headed and pull the least amount of current when closed off.
Kinarfi

 

[KeepItSimpleStupid]

kinarfi:

They are not really pumps, but better called power heads. They almost could be called "stationary thrusters" or a boat thruster attached to an immoveable object or an enclosed propeller.. They create water movement.

See https://www.marinedepot.com/powerheads_pumps__index-ap.html for some examples.

Joe wants water movement/pump control that's different that what can be easily obtained. The controllers and the pumps are married to each other and you generally cannot interchange within brands.

His "pump" is a simple brushless 2-phase 24 VDC motor of "Chinese?" manufacturer. The "pumps" are unreliable when operated without "protection". The goal here now is to provide a pump driver that will protect the 24 VDC BLDC pump motor against stalls.

The last "tested" design failed to start 2-3 out of 6 pumps sometimes.

 

[KeepItSimpleStupid]

Joe:

I'll be sending you an envelope again, probably tomorrow.

The goodies: Vector push in terminals used to mount components with leads to perf board with 0.042" holes with instructions. See: https://www.newark.com/vector-elect...lar Products&MER=PPSO_N_C_EverywhereElse_None

A complete 3 pin polarized header/connector(pins)/female sockets with 0.100" centers https://www.newark.com/te-connectiv...d-connector-receptacle/dp/85C1047?Ntt=85C1047 plus a female header and extra pins to practice with and destroy.
Not everything is shown in the page.

A sample wire connected to a crimped pin and some tiny ZIP lock bags that you can use to store components.

KISS

 

[salty joe]

4pyros & ()blivion, battery backup would be super nice. Am I correct to assume greater efficiency using a DC (rather than AC ) motor with a battery backup? What's SLA? Google was no help.
I don't have anything in the tank yet. I need to get these pumps going and we seem so close. Waiting on one part to build the Mk13.
Thanks Kinarfi, I’ll check out that flow able silicone.
Thanks KISS-can hardly wait for Chistmas to come early!
Thanks all!

 

[kinarfi]

kinarfi:

They are not really pumps, but better called power heads. They almost could be called "stationary thrusters" or a boat thruster attached to an immoveable object or an enclosed propeller.. They create water movement.

See https://www.marinedepot.com/powerheads_pumps__index-ap.html for some examples.

Joe wants water movement/pump control that's different that what can be easily obtained. The controllers and the pumps are married to each other and you generally cannot interchange within brands.

His "pump" is a simple brushless 2-phase 24 VDC motor of "Chinese?" manufacturer. The "pumps" are unreliable when operated without "protection". The goal here now is to provide a pump driver that will protect the 24 VDC BLDC pump motor against stalls.

The last "tested" design failed to start 2-3 out of 6 pumps sometimes.

I did read most of the thread and toward the end I only read every other page or so, but it sort of appears that we're attempting to fix poor design problems.

With failure rate of these thrusters and their married controls, maybe a single pump with several discharge tubes strategically placed and appropriately nozzled may yet be an option if the discharge tubes aren't too much of an eye sore or are embedded in some of the art work of the tank. And if the coral need varying flow or even reversal, that could be accomplished with remote valving, even a secondary back up pump could be included with automatic switch over in the case of a failure.

what does PDM stand for?

 

[KeepItSimpleStupid]

I did read most of the thread and toward the end I only read every other page or so, but it sort of appears that we're attempting to fix poor design problems.

alec coined a couple of terms and I think PDM is a Pump Drive Module.

We think we are. In fact we thin that the pump uses a Melexis DLDC motor IC and some of us think that it's not designed to operate at 24 V. If you look at the datasheet available on this page: **broken link removed** there is a diagram showing a +12, +24 schematic. From reverse engineering, that's essentially what I found. We don;t know the IC or FET part numbers or whether or not they are N or P, but Melexis has a patented property that the circuit has.

It seems strange that the above page mentions +5,+12 operation and the Max Vcc is 18V, but the datasheet has a pic about +12, +24 VDC operation. We're guessing.

The OEM controller uses a LM2576 switching regulator which does allow a brief surge of 6Amps, but it's a 3A regulator. The OEM controller runs the motor at three different speeds for adjustable times. It does not run the motor for very long at full voltage hence supporting the 18 V limit.

The first PDM that Joe tested was integrated into a sequencer and it had virtually unlimited power available. Something like 10A at 24V and (I think) there was a 6A fuse or so. The internal BLDC controller board that is potted within the motor self-destructed. The motor windings also shorted and so did what we assume were FETS.

So, the question is:
What can we assume and what can we do to protect the pump while running it at nearly full speed?

As the pump runs, it will develop more friction and a unprotected stalled pump isn't good for the pump.

Joe wants two different kind of controllers.
One is a tide controller. It operates two pairs of pumps for a long time (12 hr or so) and while the other pumps are off, they are constantly "flicking" or turning on briefly every once and a while to keep the fish away from the propeller. With two pumps on each end of the aquarium, you simulate a tide.

The other controller is a wave controller and it uses a pair of pumps. When one is on, the other is off. These pumps provide motion for the coral reef which needs motion to survive.

Not included yet, is the ability to create a "feed mode" which turns off the pumps or slows them down for a time such as 10 minutes so the fish can eat.

The PDM currently incorporates an alarm to detect a stall and attempted restarts every 30 seconds or so.

The pumps Joe is using are cheap, otherwise he could be looking at $6000 for 6 pumps and he still would not know if he could get the sequencing he wants.
It seems that all of the pump manufacturer's consider their pump technology proprietary.

Yes, a PLC or micro controller option did come up, but Joe's motto is simple and cheap and DIY and who knows what would be required to do it if the budget was $10,000 USD.

And, I did suggest that a new BLDC driver be designed from scratch. That would require removing epoxy potting non-destructively from inside the OEM pumps and repotting.

Both options are currently not being considered, but the latter is possible if all else fails.

We have not addressed speed control except for changing the power supply voltages for each pump. It's not known if it's needed.

Some BLDC motor IC's have a tachometer and or a fault output which "COULD" be utilized in a total redesign of the internal BLDC controller. We really don't want to go there.

 

[salty joe]

Very nice synopsis KISS.

 

[kinarfi]

May I suggest that Salty Joe is a tinkerer and most likely a perfectionist, but is more into aquariums than electronics and has come to here for his help from others who are also tinkerers and perfectionist as some of us might go to an aquarium forum for help if we were building an aquarium.
Salty Joe, I spent a year and over a grand building a power steering unit for my OHV that worked nicely, but I ended up using a store bought unit, for the time being, because of a flaw in my steering wheel torque sensing design, seems it was sensitive to RF signals and wanted to turn right if anyone keyed a radio near by, even though my unit has more power.
My question to you is could your design, wants and plans be accomplished using the discharge from a tube? and if using tubes, would you still need to incorporate the "flickers"/
Do you have any photos of the set up you have to date?
Kinarfi

 

[ronv]

I think it could be accomplished with one big pump but it would just transfer the complexity to the valves. He needs to turn it on and off at a frequency that matches his aquarium so as to produce waves ~~. If I understand corretly the flick is just to keep the little guys from falling asleep next to a pump and getting sucked in.
All the stall protection stuff is there to keep the pumps from burning out if they eat a fish or a rock and stall.
Separate pumps and retries are there to try and keep the coral alive in case of a stall because they hate stress.

 

[kinarfi]

First, I don't want to sound like I'm advocating anything, but just asking questions, before advocating, the valving I would advocate would be butterfly or ball valve connected so it could be opened fully or partially and it would open and close slowly and could be driven with a servo or linear motor or even some other method with micro switch limits, stalling would not be likely, and critter intake would not be a problem because of the screened intake and what ever makes it through the screen would probably just flow through the system. The wave action could be controlled by the opening and closing of the valves at what ever frequency is desired and if desired, it could have flow reversals.
Second, I'm only viewing in my head what could be done, with out the problems of actually trying to do it, but is seems possible and like it may be simpler than making those thrusters work like Joe wants them too.
That being said, I shall butt out and stand in the shadows and watch for a while as I feel the current design is in good hands.
Kinarfi

 

[()blivion]

My question to you is could your design, wants and plans be accomplished using the discharge from a tube?

There is nothing wrong with the idea. But this project is already long since going the direction it's going. And there is not sufficient reason to change it now I don't think. Doing so would be a slap in the face to the hard work everyone has already done. Otherwise I would have pushed hard to make this a uC project long back when parts hadn't yet been bought.

 

[KeepItSimpleStupid]

Pumps, backup pumps, valves and possibly a PLC is an interesting concept, but even had having Fluid Mechanics, I'm not sure I'd know where to start.

An interesting concept, though.

 

[salty joe]

Well put ()blivion. What does uC stand for. Google insisted it means University of Ca.
I agree with you ronv, that valves and solonoids and their control would be even more complex-certainly more costly. I was hoping so bad that the last part would have been in today's mail for the Mk13. (the latest pump driver module)
Kinarfi, how about if I post a picture ofthe Mk13 as soon as the first one gets done?

 

[KeepItSimpleStupid]

uC or μC

Ok, a difficult acronym. 1E-3 Coulombs or a microcoulomb. Hey, look Wikipedia gets it right: https://en.wikipedia.org/wiki/Microcontroller

No, that's not it either. uC stands for microcontroller. Read about it here: https://en.wikipedia.org/wiki/Microcontroller

I threw in the acronym PLC, so I might as well define that: https://en.wikipedia.org/wiki/Programmable_logic_controller

@Kinarfi: I would probably describe Joe as a "stubborn do it yourselfer". It's not meant to be negative, but he learns better first hand. It's his thread and he's in charge here. His perfectionism lies with wood and to give you an appreciation of his talents, the aquarium is built into a wall and has skylight illumination. The tank is built from scratch. The pic is in this thread somewhere.

 

[kinarfi]

Thanks for the info! Like I said, I'm not advocating anything and I can see a lot of work has gone on here and would love to be able to offer some advice and help, but it's mostly above me, back to the shadow and standing in the corner watching. I did read that Joe made the aqurium himself and that's what I'd like to see a photo of. Good luck and good work, thumbs up
Kinarfi