Currently, I have a 9V external power supply, but from my understanding this is woefully inadequate, especially since I would like to run multiple servos at once. I have purchased four 6V batteries (four AA's in series) and would like to make my servos run off these.
I'm leaning towards connecting each one individually to its own battery, but I'm not sure quite how that would be wired.
Should I connect the four batteries in parallel? I'm a little worried about melting the servos, as I don't know what happens if I hook up a servo to four batteries in parallel.
pnev:
I have purchased four 6V batteries (four AA's in series)
This is confusing. Do you mean you have bought 16 x AA cells and you have connected them into 4 packs of 4?
I would expect to be able to power all 4 of those small servos from a single pack of 4 xAA alkaline cells or NiMh cells.
If the single pack discharges too quickly then you could add another pack wired in parallel (so the voltage is the same) or you could simply swap the used pack for a new pack.
Something like a 6v 4amp mains adapter might be a more sensible option. You might even find that 2 amps is sufficient if the load on the servos is light.
Whether using batteries or a mains adapter make sure to connect the servo GND to the Arduino GND.
If they are good quality AA cells one of those 6V packs should do the job. Connect battery + to all the servo + wires (usually red) and then the battery - to all the servo grounds (usually black) and also to the Arduino ground. And that's about it.
You may find that even a fresh 4xAA battery pack is inadequate for four servos moving at once, however each pack can usually handle two servos.
You can use more than one battery pack if you connect all the grounds together.
A good rule of thumb is to have a 5 to 6V servo power supply capable of supplying one Ampere per moving servo. A fresh 4xAA pack can supply about 2 Amperes.
jremington:
A fresh 4xAA pack can supply about 2 Amperes.
If you can't get more than 2A out of fresh alkalines or fully charged NiMH AA something is very wrong. But I suppose providing more power than needed doesn't really do any harm.
If you can't get more than 2A out of fresh alkalines
Don't take my word for it. Try it yourself with fresh alkaline batteries and let us know the results. (You can get more out of NiMH AA cells). But do keep the internal resistance of a fresh alkaline AA cell in mind:
Johan_Ha:
If 4 AA batteries in series can provide 2 A, then a single AA battery can provide 2 A.
I have an AA alkaline battery from Asda (i.e. cheap) and it's not brand new but I can't remember what I used it for. It's been sitting on a shelf for a while. It easily produces 4.5 amps.
Do we need a remedial course in Ohm's Law? From the plot in reply #7, the internal voltage drop of a fresh AA alkaline cell providing 4.5A is about 0.25 Ohms* 4.5 A = 1.1 V, leaving about 0.4 V to power your device. And the internal resistance goes up with age and discharge.
Extending this to a fresh 4 x AA alkaline battery pack supplying 4.5A leaves you just 1.6 V to power your servos. That is not going to work!
Do the calculations jremington shows. They explain what you observe by measuring the voltage of your power supply with a multimeter while your servos are operating.
Do we need a remedial course in Ohm's Law? From the plot in reply #7, the internal voltage drop of a fresh AA alkaline cell providing 4.5A is about 0.25 Ohms* 4.5 A = 1.1 V, leaving about 0.4 V to power your device. And the internal resistance goes up with age and discharge.
And the IR goes down with temperature. Alkaline battery starts at around 1.63V (at least the 3 I just measured do). The IR you're quoting is notional and at 0 degrees. It's about half that at 20 degrees, a much more normal temperature to be working at.
So we started talking about 2A. 0.15 Ohm, 2A = 0.3V. 1.63V - 0.3V = 1.33V. 4 of those = 5.32V. No trouble running servos on that.
But I'd still prefer NiMH rechargeables. Lower initial voltage but hold up better under load. These days who uses primary cells for this sort of application?
So we started talking about 2A. 0.15 Ohm, 2A = 0.3V. 1.63V - 0.3V = 1.33V. 4 of those = 5.32V. No trouble running servos on that.
The OP started talking about 4 servos.
For 4 servos moving at once, one should budget 4 Amperes, and not be so optimistic interpreting an idealized chart for a fresh cell.
Agree about the rechargeable cells, which have significantly lower internal resistance, but many beginners use alkaline battery packs and then wonder why their project doesn't work.
I've used a lot of SG90 servos and I've never seen one take more than about 3-400mA. If you have all 4 servos heavily loaded or completely stalled and you try to start all of them at once you might get over 2A for a few milliseconds.
It probably is worth being extremely cautious in a professional mass-production situation but for a one-off hobby project it's often just wasting money. But that's only my opinion as, these days, just a hobbyist.
Obviously, you've never looked properly. Servos and other brushed DC motors briefly draw the stall currentevery time they start up.
For the SG90, the manufacturer states the stall current to be 650 +/- 80 mA, so budget 750 mA to be safe. For larger servos, the stall current is much, much higher.
For this reason, those in the know recommend to budget at least 1 Ampere per servo. A glance through these forums reveals that over 95% of servo problems arise from people who do not appreciate this advice and do not plan accordingly.
Specifications are interesting but I'll settle for actual measurements. Though I suppose if the startup "stall current" lasts for an incredibly short time it's always possible my loggers are (invariably) missing it. It is true I don't generally measure actual stall current for the obvious reason that stalling a servo for more than a second or so will probably damage it.
Anyway since there's no harm, other than to the pocket, in over specifying the power supply requirements I'll shut up now. I'd rather see people overdoing it than trying to run servos on a PP3 battery or the Arduino 5V pin (the cause of most of the problems I've spotted).
