![[H]ard|Forum](/styles/hardforum/xenforo/logo_dark.png){kind=logo}
speed control for a laing ddc pump
- Thread starter BioPort
- Start date
You could go with a rheostat but there are a few potential drawbacks/concerns:
Rheostat connected across the 12v volt supply with the pump connected between the wiper and ground. - If you want the rheostat not to control the voltage in such a way that you don't go from 100% speed to completely OFF in 10 degrees turn of the knob, you'll need a fairly small value resistance for the rheostat. But this burns a lot of unnecessary power and is probably not the way to go.
Rheostat connected in series between 12 volts and the pump - This forms a voltage divider and is a better way to go. Again, the trick is in choosing a resistance for the rheostat such that the pump doesn't go from 100% speed to completely OFF in 10 degrees rotation of the knob. On the upside, this doesn't burn excessive power.
LM317 type voltage regulator - You can do this with a LM317, two resistors (or one), and a potentiometer. This would give you more precise control of the voltage over the rotational range of the potentiometer, and mounting a small potentiometer on a front panel will take less space than a rheostat. Downside is that wiring is more complicated, and you'll never get 100% of your 12 volts out of it (max is voltage applied minus 1.2 volts).
Rheostat connected across the 12v volt supply with the pump connected between the wiper and ground. - If you want the rheostat not to control the voltage in such a way that you don't go from 100% speed to completely OFF in 10 degrees turn of the knob, you'll need a fairly small value resistance for the rheostat. But this burns a lot of unnecessary power and is probably not the way to go.
Rheostat connected in series between 12 volts and the pump - This forms a voltage divider and is a better way to go. Again, the trick is in choosing a resistance for the rheostat such that the pump doesn't go from 100% speed to completely OFF in 10 degrees rotation of the knob. On the upside, this doesn't burn excessive power.
LM317 type voltage regulator - You can do this with a LM317, two resistors (or one), and a potentiometer. This would give you more precise control of the voltage over the rotational range of the potentiometer, and mounting a small potentiometer on a front panel will take less space than a rheostat. Downside is that wiring is more complicated, and you'll never get 100% of your 12 volts out of it (max is voltage applied minus 1.2 volts).
I'd go with a LM318 or LM338 instead - these are rated at 3A and 5A respectively.BioPort said:
You'll need a heatsink too - you can only dissipate a couple of watts in a non-heatsinked TO-220 case, and 0.9A with a few volts of dropout voltage will easily go beyond that.
Silent Assasin
[H]ard|Gawd
- Joined
- Aug 1, 2004
- Messages
- 1,120
Can I just ask why you NEED to reduce the speed of the pump???? If you say it's for noise problem, then you got stuck with one of the crap pumps and you should RMA it....
I just finished testing a circuit for controlling this water pump using PWM from my motherboard. I can varry the voltage from 7.4 to 11.1 using Speedfan. This corresponds to tachometer readings of 2900 to 4100. The LM317 has a pretty costly voltage drop, so your max would be about 10.2 volts.
You could modify the circuit to use a pot. for manual control, instead of PWM.
Did you ever build a circuit, or RMA your pump?
You could modify the circuit to use a pot. for manual control, instead of PWM.
Did you ever build a circuit, or RMA your pump?
BioPort said:
I don't think your pump required 11 volts to start. The problem with a rheostat is that it provides constant resistance. But the pump's resistance is not constant. It is much lower when it's trying to start (it charges an internal capacitor I think), which means that the voltage drop across your rheostat was too high initially. My guess is that you could dial the rheostat down after the pump had started to around 6-7 volts.
With the circuit I made, my pump starts fine at below 8 volts.
You can solve the problem by putting a large capacitor in parallel with the resistor. Then, when the power first comes on, the capacitor allows more current through for a short time, until it's fully charged and the resistor takes over.
