### Author Topic: Power supply question  (Read 1239 times)

0 Members and 1 Guest are viewing this topic.

#### arixrobotics

• Full Member
• Posts: 119
##### Power supply question
« on: January 22, 2014, 09:16:07 AM »
Hi all, can someone be so kind to explain to me what are the things to consider when choosing a power supply (240V AC to 12V DC)?

I am trying to power some DC motors, and I thought voltage and current is all I need to consider.

I started with a power supply with output rated at 12V 1A. However, some of the motors started to 'hiccup' when connected directly to the power supply output. So I thought the supply current is not enough, so I got a 12V 1.5A power supply. However, it doesn't always solve the problem. On some motors, it gets worse.

So I am guessing that there is more to just voltage and current ratings in this case. Maybe the over current protection system? Maybe latching power supply can handle DC motors better than switching types? How can I get the motors to spin without hiccups??

#### jwatte

• Supreme Robot
• Posts: 1,345
##### Re: Power supply question
« Reply #1 on: January 22, 2014, 11:25:32 AM »
Do you know what the rated voltage/current is for those motors? If the motors want to draw more than 1.5A (together) then clearly the power supply is not sufficient.

You are right that voltage must be proper for the motors, and current must be greater than the sum of the current demands of all loads (motors in this case.) Additionally, you MAY want to look at things like transient step response. Ir just add a big capacitor and call it good :-)

Three other properties that matter for power supplies are:
- Output voltage ripple. Not so important for motors; very important for logic.
- Overcurrent protection mode. If it's "hiccup mode" then the power supply may stop delivering power when a large load is connected, and then come back, repeat. I prefer "constant current limiting" mode that drops the voltage until the current is properly controlled.
- Efficiency. A 75% efficient power supply will generate a LOT more heat than a 95% efficient power supply.

My guess, from the vague data available, is that your power supply is "hiccup mode" protected, and that your motor or motors draw more than 1.5A together, at least temporarily (say, when starting up.)
Reading the specifications for the motors, and measuring things with an oscilloscope and multimeter, would be the next step in trying to define the problem.

#### arixrobotics

• Full Member
• Posts: 119
##### Re: Power supply question
« Reply #2 on: January 23, 2014, 06:01:56 PM »
Thank you very much for the reply jwatte

I think u r right, the motors are drawing high startup current and the power supply are in "hiccup" overcurrent protection mode.

I've also tried using another power supply which is a bit more bulky and heavier (more like laptop power supply, not like the smartphone chargers). This one worked well, although the current rating is lower. So I guess yeah, its the overcurrent protection mode.

But the thing is, with these cheap power supplies, it doesnt mention what protection mode it is in, what output voltage ripple, what efficiency etc... and due to the nature of this project, I could only change the power supply (not the motors), I guess I'll just go shop around for a couple of *promising* power supplies and test them out

thanks again!

#### jwatte

• Supreme Robot
• Posts: 1,345
##### Re: Power supply question
« Reply #3 on: January 23, 2014, 08:37:52 PM »
I agree: Cheap "wall warts" are seldom well specified, and also seldom achieve the rated output current. My suggestion is to not buy cheap power supplies :-)

If you're doing electronics experimentation, then you will likely want a benchtop power supply, and you probably want something that can deliver at least 30V and 10A.
Benchtop power supplies work in "constant current" mode, so they cut the voltage if the current draw is too much. This will let you try various power settings easily by just turning dials.

Another option might be to use a car battery charger as a poor man's power supply. It outputs about 14V and you can typically limit the current to a few amps for "trickle" or "float" charging. They may also need output filtering.

Another option is to use the 1.5A supply and make it not "see" the start-up load. To do that, you might add something like a 1000 uF 25V capacitor, followed by a 400 uH 2A inductor, followed by another 1000 uF 25V capacitor. This will make a "Pi Section" filter which will attenuate transients of about two milliseconds or less. That may or may not be enough to make the power supply happy.

If you don't have the specifications for the motors (nor the ability to measure them, say with a multimeter) then you're going to have to experiment, though.

• Full Member
• Posts: 119