Go Down

Topic: AC pump high speed dimming problem (Read 2330 times) previous topic - next topic

steven1804

May 20, 2013, 11:58 am Last Edit: May 20, 2013, 12:01 pm by steven1804 Reason: 1
Hi guys, I'm working on my final project to build a 8-bit water printer and I use 2 arduinos which are mega and uno, the mega used for controlling how many solenoid valves to be opened and send the data to uno through ttl serial to control the AC pump, they are about 3 meters separated apart.
I already built a 1-phase ac pump dimmer circuit based on moc3021 + cny17-4 zero cross detector
the only thing I want is the water outlet pressure to be adjustable, so far I get the job done by dimming the AC pump, this is my video for my first attempt reducing the pump speed via serial (only displaying the pump fan movement)

http://www.youtube.com/watch?v=HUzespKyI98

my problem now is that water pump didn't get enough response to the responding amount of solenoid valves opened in about 30-50ms interval.
for example if the arduino mega open two solenoid valves to display "X" character in 50ms interval there isn't any problem because the speed is linear, only 2 valves opened in each time interval.
but if I want to display like "C" character, it means different amount of opened valves in each time interval, the pump is only move in really slow speed (sorry pump movement video not provided).. so I used the pump full speed to display a random amount of opened valves instead a dimmed one.
is there an efficient method to control the pump in high speed?? such as variable speed drive?? my method is phase angle control

sorry for my bad english and long explanation XD

Chagrin

That looks like a capacitor-run induction motor (it has two large caps under that cover?) and should not be controlled with phase dimming. I'm by no means an expert when it comes AC motors, but I know that using phase control with non-shaded pole (cheap AC fans) or universal motors (which have brushes -- electric drill, kitchen blender) leads to overheating in the motor coils and death. Take the time to consult an expert on this if available.

Seems like what you really want is a pressure regulating bypass valve, but I don't think those are a common item (or cheap).

steven1804

Hi Chagrin you're right but that's only one capacitor installed across the pump, and the value is 8uF  :)
why I can't dim that type of induction motor? because I've found Freescale app note describing that we can dim this type of induction motor with phase control
just this morning I've tested again and no problem, except for high speed dimming, like I said before, moving slowly and emitting sound noise
I dunno if my pump is going to overheat, I'll check about this later :)
I'd like to control the pump, because finding a special function valve like that is difficult here, but thanks for the suggestion :)

Regards,
Steven

steven1804

oh no bad news here..
I think you are totally right Chagrin
I just noticed this at the time I tested my project again
my pump gets warmer than usual if I used the phase control dimmer for about 5 minutes, luckily my hands didn't get burned when touching the pump  :smiley-eek-blue:
but it is OK..this is an old pump..I won't use phase dimming to this pump again if I graduated later  :)

Best regards,
Steven

Chagrin

If it's hand-burning hot I don't think it'll last long enough to make it through graduation. Your concern is that you'll start burning the insulation off the motor coils; all you need are a few locations on any coil to short out and everything will go ugly quickly.

dc42

Better to swap the pump for one with a brushed DC motor, those are easier to control.
Formal verification of safety-critical software, software development, and electronic design and prototyping. See http://www.eschertech.com. Please do not ask for unpaid help via PM, use the forum.

steven1804

@Chagrin: if it's only for a while it doesn't matter but I've testing it for a long time my pump going overheat when I touched it. I'm gonna find another solution for this, and I already consulted to my lecturer that this option nearly impossible to be implemented
he suggested that I should put a big water tank on top of my current water tank and pump is used to fill the big water tank, and my current water tank get the water supply from that big water tank, therefore the water pressure is going to decreased, but I haven't tested it yet.



@dc42: I dunno if there is any DC pump available here that providing enough pressure to push water for about 2,5 meter upward?? Yeah..DC PWM is more humane than AC phase control

dc42

#7
May 23, 2013, 03:51 pm Last Edit: May 23, 2013, 04:04 pm by dc42 Reason: 1
You haven't said what flow rate you need. Is this any good http://www.ebay.com/itm/DC40-2470-Brushless-Motor-DC-24V-1A-Magnetic-Drive-Water-Pump-172GPH-7-0M-26-4W-/120934931180 ?

[EDIT: I just noticed it has a brushless DC motor, so it may not be easy to control the speed.]

[EDIT 2: also found this one http://www.ebay.com/itm/DC-12V-pump-water-cooler-motors-speed-line-No-Brushless-/150774292664?pt=BI_Pumps&hash=item231ad920b8 which has a speed control input.]
Formal verification of safety-critical software, software development, and electronic design and prototyping. See http://www.eschertech.com. Please do not ask for unpaid help via PM, use the forum.

inthebatcave

One solution might be to double your solenoids. For every solenoid you open one closes, that way you have a constant flow rate. Also unless you are using a positive displacement pump, centrifugal pumps follow a pump curve, so flow and pressure do not change linearly with pump rpm.

JB_AU

A non return valve between the pump & tank is a good idea too
Amateurs built the ark, Professionals built the Titanic

inthebatcave

If you do not need much pressure, you could always adjust pressure by maintaining a certain tank depth with an overflow tube piped to the the catch basin.

Go Up