Controlling Aquarium Air Pump Output

Hi folks,

I have a large aquarium air pump (Hailea AC-2208) that is quiet and reliable. However I've reduced the number of fish tanks that I have and thus no longer need all of its output. Rather than swap to one of my smaller, noisier and less relaible pumps I'd like to continue using the Hailea but reduce its output (reducing its power draw would also be great).

The pump is the vibrator type. A permanent magnet attached to an armature with diaphragm-type 'compressors' at each end moves from pole to pole of an AC electromagnet.

I was wondering if something like this would do the job without damaging the pump;

formatting link

I think this is similar if not the same unit - I can't find the one Clive reviewed;

formatting link

Thanks in advance for any replies.

--
Shaun. 

"Humans will have advanced a long, long way when religious belief has a cozy  
 Click to see the full signature
Reply to
~misfit~
Loading thread data ...

I don't know much at all about compressors, but it sounds like the effect of using a phase angle controller would be to reduce the force the compressor could pump at more than its speed. If it were a "rotary" compressor based on a normal electric motor, the speed may decrease in better proportion.

I don't know enough about compressors and fish to know the implications of this. At a guess - if you wanted to use the pump for fewer fish tanks, you'd want the same pressure but less volume, so slowing the speed of the compressor would be better than reducing its power. A similar device could be used to reduce the speed by turning off for entire cycles of the AC mains instead of a percentage of each cycle. Or perhaps a DC power supply and controller could be used, alternating the polarity at a variable rate (a lower voltage would be used than with an AC supply). I don't know if either of these things exist as cheap "modules" (the latter, DC, option wouldn't likely be cheap in any case).

I haven't watched the video, but I believe I know the principal of the device in the video's description. I don't imagine it could damage the pump, unless perhaps the pump has its own circuitry that might be affected by inductive pulses.

I'd still strongly recommend closely examining any item that cheap and Chinese before letting it anywhere near mains. Actually as a rule I never use any cheap Chinese stuff from ebay with mains, but that's just me.

--
__          __ 
#_ < |\| |< _#
Reply to
Computer Nerd Kev

This youtube video shows a pump similar to mine, just a little bigger;

formatting link

The reason that I'm so concerned about not damaging it is the pump was quite expensive for me and I have the luck of a very unlucky thing. A wee while back I needed a repair kit for it and had the devil's own job finding one - I ended up buying a second-hand pump that had burnt out and using a valve block from that on my pump. I'd like to keep it running as long as possible as I can't afford a new one.

I'm worried that the essentially shorter pulses might cause more stress on the components or maybe start harmonic vibrations? I don't know, just want a more qualified opinion really.

I always dis-assemble and check anything like this that's going to be used on mains voltage. I don't trust my life to a minimum-wage foriegn person assembling something that they likely don't even understand.

Cheers,

--
Shaun. 

"Humans will have advanced a long, long way when religious belief has a cozy  
 Click to see the full signature
Reply to
~misfit~

Next month. I just discovered I've got 3MB internet data left for this month. :(

Fair enough. The pulses won't damage the coil itself, but the back EMF it will produce when suddenly turned off has the potential to damage any electronics used in the pump (or the controller, though it should be designed to handle it). Any protection circuitry used may not have been designed to cope with the high frequency of these inductive pulses due to the rapid power switching.

I don't know about the mechanical side. My fear would more be of some mechanical action not being reset if the coil is not energised enough to complete a normal stroke. I doubt it's that complicated, but I'm just guessing. Of course the manufacturer would be the best to ask.

--
__          __ 
#_ < |\| |< _#
Reply to
Computer Nerd Kev

Bugger! When you've got data this youtube video is of a pump made by the same company as mine, mine's just a smaller model (30l/m as opposed to

60l/m) but construction of the pump is almost identical in all but dimension.
formatting link
The video is of a guy trying to 3D print replacement parts for his as he can't get spares either - and he's in the UK, not at the ends of the earth. The flapper valves tend to 'go' first and my pump, being smaller than his has glued-shut plastic tops of the valve chambers whereas his are metal and held on by screws. For his there are supposed to be replacement flapper valves avaiable whereas for mine the 'fix' is to fit a whole new valve block - which isn't available here in NZ and I couldn't find anyone prepared to ship one to me from Aus, EU, UK..... :( I fixed the last issue by buying a second-hand one that had a burnt out electromagnet and using the valve block from that. However they're not common second-hand so if it happens again......

The biggest killer of these types of pump (at least the valves and diapragms) is excess back pressure. Lifetime is shortened logrithmically in relation to working pressure beyond their stated maximum. With the smaller ones used in home aquaria the airstones they use get populated by aerobic bacteria quite quickly and within six months or less they can increase backpressure enough to destroy the pumps neoprene parts. I change my airstones every three months and soak the used ones in bleach overnight then rinse and store for the next change cycle. ;)

I've ordered one of the controllers - they take a while to arrive and even if I don't use it for this I dare say I'll find a use for it some time. The pumps themselves are 'dumb' - just a coil attached to the mains from what I can tell. There's not even a switch. I guess then my question is a bit outside of this newsgroups charter but I didn't know where else to ask a question like this and I'd hate to (effectively considering the dearth of parts) kill the pump.

You'll see in the video/s when your data re-sets. It's just an armature that oscillates in a magentic field at mains fequency attached to neoprene diaphagms at each and which pump a small amount of air each stroke through cicrular 'rubber' flapper valves which are held in their centres. I just want to reduce the output of my pump without overstressing it.

Cheers,

--
Shaun. 

"Humans will have advanced a long, long way when religious belief has a cozy  
 Click to see the full signature
Reply to
~misfit~

s I unserstand is those things are resonant systems tuned to the mains frequency, so anything that messes with the stroke frequecy would require alteration of the pump to re-tune it. this pretty-much leaves you with the only option being reducing the stroke length.

probably reducing the input voltage will do what you want. so a series dropper of some sort, or a step-down transformer. you could try a fan speed controller, or to test this out a 100w lamp.

--
  \_(?)_
Reply to
Jasen Betts

formatting link
"YouShop gives you delivery addresses on the ground in the UK (for shopping across Europe), USA and China..." not free of course.

--
  \_(?)_
Reply to
Jasen Betts

A small Variac would work but probably cost as much as a new pump.

What is its maximum power consumption? If it's only (say) 50W you could operate it from an old audio amplifier with a simple sinewave oscillator. Step up the output using a mains transformer in reverse and use the volume control for speed.

Does the armature have any centering spring around which it oscillates? If so, then Jason is right and it is resonant at somewhere near the mains frequency. Changing the frequency will reduce the efficiency.

No - don't do that. A thyristor-based controller will mess with the coils badly. You need a sine-wave power source.

Yes, but the resonance is not very high-Q. The goal is to convert the energy into compressed air, after all. Changing the frequency if the Q is low will just reduce efficiency, but not catastrophically.

Clifford Heath.

Reply to
Clifford Heath

What a load of BS all these answers have been :-? The pump will only operate properly at its rated voltage and FREQUENCY

The easy way is to just let the old delivery hoses dump to space or even into the remaining tanks and adjust the various flows to do the job lol You do have taps on each these to adjust the flow,Don't You?

Too often we get these pie in the sky fixes that require the silliest or unmanageble actions to acheive the simplest of tasks :-?

--
John G Sydney.
Reply to
John G

Thanks, I'm aware of that (expensive) service. I should have been clearer - not only couldn't I find anyone to ship the parts to me but I couldn't actually find a supplier period. That's why the guy in one of the videos I linked was 3D printing parts for his pump - he couldn't get parts either and he's in the UK.

Cheers,

--
Shaun. 

"Humans will have advanced a long, long way when religious belief has a cozy  
 Click to see the full signature
Reply to
~misfit~

The OP doesn't want it to operate properly. He wants it to operate at reduced output, cheaply. Stoopid ideas yes, but the definition of "cheaply" depends on what he has lying around and how he wants to use it.

Any reduction in voltage or change in frequency will reduce efficiency - but not as much as bleeding of excess air will.

Reply to
Clifford Heath

They're hardly tuned per se. There's an armature made up of permanent magnet (or two) in a light frame suspended between the electromagnet poles attached at each end in the middle of a rubber diaphragm. As the AC alternates and the elctromagent poles change the whole shebang moves somewhere between 5 and 10mm each way. One-way rubber flapper valves let the air in and out of the two diapraghm chambers.

As I understand it, going by what Clive says at the end of his teardown the 'speed controller' reduces the duration of the AC pulses. So it's still 50Hz but instead of a sine wave the positive and negative pulses duration can be modified to, for instance half of a sine wave with the remaining half flat line - zero volts.

So for instance at 'half power' instead of 10ms of positive and 10ms of negative it would be zero volts for 5ms, positive pulse for 5ms, zero volts for 5ms, negative pulse for 5ms. At least that's how I understand it, I could be wrong, this isn't my forte.

My hope is that this would result in the armature only getting half of the motive force so only moving half as much (with the controller set at half power). However it might instead pulse each way then return to centre instead? Also I'm not sure what effect if any this type of 'peak duration reduction' would have on the longevity of the electromagnet?

I don't have any of the above.

Actually I may be able to dig out a 60 or 75 watt incandescant light bulb from the back of the cupboard - I only use LED lamps in the house now, swapped out the CFLs a while back. What would the effect of wiring an incandescant bulb in parallel to the pump be? I mentioned I'm new to electrical theory yes?

Cheers,

--
Shaun. 

"Humans will have advanced a long, long way when religious belief has a cozy  
 Click to see the full signature
Reply to
~misfit~

Yeah. I'd love to own a variac but alas.....

It's 25W but that setup sounds compicated. I don't have a sinewave oscilator either.

Please see my reply to Jason. No spring as such, the centering after pulses is due to 'rubber' diaphragm suspension. Well I guess they *are* actually acting as springs in this device but I doubt they'd have a similar sort of resonant frequency as say a steel spring.

Seriously? Damn.

That's a bit beyond me... Or not if by 'Q' you mean resonant frequancy of the armature assembly?

Cheers,

--
Shaun. 

"Humans will have advanced a long, long way when religious belief has a cozy  
 Click to see the full signature
Reply to
~misfit~

Yup, not much resonance involved then. The "push" energy goes into one cylinder and the "pull" energy the other; very little into any energy storage (spring and mass).

Q is a ratio of the amount of energy circulating (stored in the moving spring & mass in this case) divided by the amount of energy extracted in each cycle. So if it's putting out 25 watts at 50 cycles per second, each cycle puts out half a joule. If the spring/mass oscillation stores (say) 10 joules, the Q is 20 (=10/0.5).

But your pump sounds like it has a Q of less than 1. Not resonant. Probably operates exactly at mains frequency, and will probably work at a lower frequency, producing less air volume. Just don't feed it DC or the magnets will saturate and probably melt wires. You really need an AC current source, enough to produce the correct stroke at your desired power output.

Reply to
Clifford Heath

No not really, it will just make it less efficient but not greatly so before it just doesn't work.

What it needs is an air bleed to bleed off the excess air you don't want I should think.

Reply to
Clocky

No, ideally, frequency reduction, at the right current.

A variac will work, but will reduce efficiency, because it works by reducing the stroke length. A certain volume of air in each stroke is wasted opening and closing valves, and that doesn't become less as the stroke length reduces, so you lose some efficiency. Probably not enough to matter much though.

If you used a fixed mains transformer with, say, a 30-0-30V winding, you could wire the 60v in antiphase from the active to step down to

180v from 240. That might be the cheapest way to reduce output, even though it's not easily adjustable.

It might. I'm not an expert on dimmers, but most don't like inductive loads. There are leading-edge and trailing-edge dimmers. The trailing edge dimmer turns off part-way through the half-cycle. With an inductive load, that will produce massive voltage spikes that will blow up the dimmer and possibly other things in the vicinity. The leading-edge dimmers turn on part-way through the half-cycle, which won't produce a voltage spike.

Neither type is guaranteed to produce balanced AC - you might get more +ve than -ve or vice versa. That's bad for an iron-cored inductor because it can easily cause core saturation and high currents that will melt your windings.

At this point, I'm at the limit of my knowledge. I would trust Phil A's opinion here. Phil? Tell me what I got wrong please...

Wastage depends on the quality of the rubber. Good silicon rubber wouldn't waste much - but this type is resonant (the rubber is the spring).

Clifford Heath

Reply to
Clifford Heath

Thanks for the reply and explaination.

So ideally a variac then - which I'd assumed and which I don't have.

Do you know if the device that I've ordered will hurt my pump? This is a cut'n'paste of another post of how I understand it works:

------------------------ As I understand it, going by what Clive says at the end of his teardown the 'speed controller' reduces the duration of the AC pulses. So it's still 50Hz but instead of a sine wave the positive and negative pulses duration can be modified to, for instance half of a sine wave with the remaining half flat line - zero volts.

So for instance at 'half power' instead of 10ms of positive and 10ms of negative it would be zero volts for 5ms, (full voltage) positive pulse for

5ms, zero volts for 5ms, negative pulse for 5ms. At least that's how I understand it, I could be wrong, this isn't my forte.

My hope is that this would result in the armature only getting half of the motive force so only moving half as much (with the controller set at half power). However it might instead pulse each way then return to centre instead? Also I'm not sure what effect if any this type of 'peak duration reduction' would have on the longevity of the electromagnet?

-----------------------------

TIA for any reply. I've ordered one of these devices, it only cost me $13 and if I can't use it for this I dare say I'll find a use for it some time.

I have smaller airpumps. However most domestic models work by having the permanent magnet on the end of an arm (which is usually pivoting on an energy-wasting rubber bush) then the diaphragm/s attached to side of the arm. These can be mass produced cheaply as tolerances don't need to be as tight and 'rubber' diaphragm materials don't need to be as good quality as they're not suspending the armature assembly. However their efficiency compared with the better ones is much lower (when measured in litres per second per Watt). More importantly much of the 'wasted' energy goes into vibration and noise. I like my quiet.

The 'suspended armature' type such as the one I'm currently using are far more efficient and quiet and I'd hate to go back to a noisy pimp just because I've decommisioned some tanks.

(Bleeding off air as suggested elsewhere isn't really an option as that's noisy, wasteful of power and is very hard to balance against the constantly changing backpressure of a airstones running in biologically active systems. I know as I've tried other ways in the past. The only way I got it to work reliably was to bleed off the excess air through an open pipe in a container of water at a greater depth than the outlet in the fishtanks - a primitive but sensitive pressure-release system. [You don't want to physically restrict outlets with simple valves as too much 'hard' restriction distorts and damages diaphragms and quickly kills the fragile rubber valves.] Its downsides are noise, having to have a much taller container than needed for water to contain splash, managing algae and bacterial growth and having to contsantly adjust depth to compensate for evaporation.)

I guess I could design some sort of fancy feature coloured LED-lit bubble-wall or semi-closeded similarly-lit clear acrylic cylinder, perhaps with dyed water (or even oil!) to make use of excess air. However that wouldn't fit in with my current junk-chic decor. ;)

Cheers,

--
Shaun. 

"Humans will have advanced a long, long way when religious belief has a cozy  
 Click to see the full signature
Reply to
~misfit~

Also I found a few places that listed the items I wanted but either said 'out of stock' or, on further questioning or ordering I was then told they weren't currently available. When asked when they would be nobody could say.

Cheers,

--
Shaun. 

"Humans will have advanced a long, long way when religious belief has a cozy  
 Click to see the full signature
Reply to
~misfit~

That's what I was thinking. mebbe use the air to drive a small turbine pump for circulation or power LED lighting.

Reply to
Rheilly Phoull

That'll drop the input voltage to the pump. You look at the light bulb as a resistor and use the voltage developed across it (Ohms law) to subtract from the voltage otherwise applied to the pump. But for heaven's sake wire it in _series_, not parallel.

The controller you've bought is also a form of voltage control, but as it works by switching the input waveform rather than droping the voltage over a resistance, other considerations (inductive spikes, completion of compressor stroke) arise.

As has been noted, frequency control rather than voltage control may be the better option. Does anyone know of a cheap, purchasable device that would achieve this?

--
__          __ 
#_ < |\| |< _#
Reply to
Computer Nerd Kev

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.