Power Circuit Of Robot, Required Capacitor Banks??

[Abdullah M.A.]

Hi guys,
I just would like to ask you if anyone know how to calculate the capacitor value
(capacitance,voltage) required to connect with voltage regulator in a robot draw
3A,6v ( 5v go out from voltage regulator 7805 to supply the control circuit, 6v go
to +Vcc of servo motor), the used battery is a lead Acid (6v, 10A).
The reason of connecting capacitor,I've got restarting in microcontroller when the
robot want to start moving.
any help will be really appreciated.

Abdullah

[Colt45]

A 6V battery really isn't high enough voltage for a 7805, 7.5V or 8V is necessary to not have drop-out.

So you'll need a low dropout regulator (LM1117 for example, but there are many).

[Geek]

LT1084?

[MacFromOK]

Or a resistor or couple of diodes for the control circuit... 6V to 5V isn't much of a drop.

[johansen]

3 amps isn't a lot for a 10 ah battery.

1000 uf placed near the linear regulator should be fine.
you also should have at least 100 uf per amp of motor current, placed close to the motor controller.

[CWB]

"The reason of connecting capacitor,I've got restarting in microcontroller when the
robot want to start moving."

voltage sag/drop under load .
the surge/start current drawn by the motors is dropping the supply (battery) voltage .
the LM1117 has a dropout voltage of 1.10-1.30 volts differential depending on load (Vin-Vout) as shown here :
http://www.ti.com/lit/ds/symlink/lm1117-n.pdf

the LT1084 has a bit better differential performance but still runs very close to the edge :
http://pdf1.alldatasheet.com/datasheet-pdf...NER/LT1084.html

the LDO regulators are a bit fussier to work with than their regular brothers (so to say) .
tim mentioned this in another thread a while back ... they want to whistle more so than the regular versions (ie : 7805) .

i would go with a schottky diode (.15V-.45V Vfwd) or a regular silicon rectifier (about .7 Vfwd) ... this is a possible simple solution with readily available parts .
a "stiffening" cap on the "downstream" side of the diode will provide the needed current when the 6 volt line sags under the momentary load of the motors . it will also prevent back-feed of the stiffening capacitor into the 6 volt buss/supply .
you might try about 2000 uF to start with .
the size needed will depend on how much load is on the controller side and the duration of the sag in the 6 volt line .

[Sch3mat1c]

How much current does the controller draw?

If it's fairly high, you may be interested in a wide range switching regulator. SEPIC would be excellent here.

Tim

[Nothing40]

I'd just toss a diode or two in series with the micro's Vcc line..stick a 'big' cap after that,and call it good. KISS.

[Abdullah M.A.]

Hi guys,
I used in the beginning 1N4007 with 2200 uF cap., but it does not work.
seems I needed a fast diode, so I changed with schottcky diode SB360, so it worked fine.
Thank you CWB and all of you guys.

@ Tim
about 300mA.

[johansen]

heh, if the reason the 1n4007 didn't work was because it was slow... man that's some serious spikes you got there!

the schottky diodes would have only dropped .2 volts instead of 1.1, so that's probably the difference.

[CWB]

yep ... it is the Vfwd difference between the two types .
the uPC needs 5 volts and the supply (battery) is 6 volts ... not much difference to play with .
(i took it that abdulla had already done the small cap bypassing/decoupling at the uPC)

i'm glad the simple fix worked (at least this time) .