$50 robot with 6V battery pack?

[cyberfish]

Hello! New member here!

I am trying to follow the $50 robot tutorial roughly (using a different microcontroller etc). I am an electrical and computer engineering major (second year), so that shouldn't be too difficult.

Just a quick question -
The $50 robot tutorial suggests using a 6V battery pack (5xNiMH, 5x1.2V) for both the digital stuff and the servos (unregulated).

But according to Fairchild's datasheet, the LM7805 (and 317) has a typical drop of 2V, so how do we get 5V from the 6V pack?

Many thanks!

[Razor Concepts]

You could just use a 4xNIMH pack and use no regulation at all - that is my preferred method.

[cyberfish]

Ah I see, but I ordered my 5xNiMH already . I have a few regulators (317 and 7805) lying around, so I'll give it a try.

But even if you use 4xNiMH, the nominal voltage will be 4.8V, isn't that bad for microcontroller? I thought I've seen microcontrollers (not the one used in the tutorial, though) giving bad results with lower voltage. Or it could be the ADC, I'm not sure. Or maybe a bug in my code .

EDIT: and also, doesn't the servos need 6V?

[SmAsH]

Servos can work with 4.8-6v and sometimes 7.2v. He meant use the 4.8v pack for the servos and a separate 9v battery for the electronics.
See the tutorial, there are two ways to wire the battery(s).

[cyberfish]

Ah I see, that makes sense.

I'm trying to find a way to use single supply, though.

[SmAsH]

Have you considered LDO (low dropout) regulators? You could use a 6v pack with an ldo and be fine.

[cyberfish]

LM7805 gives 4.4V with a 6V input.

Built my own simple regulator with a PN2222 and a 5.1V zener. I can get 5.00V from 6V, but then I don't get current limiting and all the good stuff (I'm guessing my regulation is not that good, either. need to add some kind of feedback).

[cyberfish]

Have you considered LDO (low dropout) regulators? You could use a 6v pack with an ldo and be fine.

That is a good idea! I've never used one, so any suggestions (part numbers)?

[Soeren]

Hi,

Ah I see, but I ordered my 5xNiMH already . I have a few regulators (317 and 7805) lying around, so I'll give it a try.

The drop out value can for eg. LM317 be 3.0V to 3.5V.
You can get Low Drop Out Regulators (LDOs) with a drop out of less than half a volt.

A Nickel based battery cell is considered flat at between 0.9V to 1.1V, depending on the discharge rate. If you draw areound half the capacity (eg. 500mA or more from a 1000mAh battery, 0.9V will be the "flat" value).

With 5 cells, you will thus have a span from ~7.25V freshly charged down to 4.5V flat. If you want to maintain a steady 5V with 5 cells over the discharge life of the battery, you have to use a switch mode converter/regulator of the SEPIC or CUCK topology, as the LM317/7805 will need 8.0V to 8.5V worst case and the LDO will need around 5.5V.

If you power motors/servos from one (unregulated) battery and servos/logic from another battery, at least you avoid the major loss and depending on your amount of servos you can get by with a smaller battery with a chopper stabilized front end for eg. a zener shunt regulator, or you might go with only 2 or 3 cells and a boost converter for the logic part.

[cyberfish]

My LM7805 gives 4.4V with 6V.

I think I will go with either a LDO or a transistor + zener regulator I just built. I don't really want to go with switch mode regulator for my first robot (trying to keep it simple). I just need the output to be 5V for the majority of the discharge time (the plateau).

I built a simple regulator with PN2222 +5.1V zener + 1N4004 (to compensate for the transistor's Vbe) + 100ohm.

Gives me approx. 5V - 1V/I from 6V. I only tested it up to 200mA since I don't have power resistors (and connecting a bunch of 100ohms in parallel is getting a little tedious), and I have no idea about transient response since I don't have a scope. At 200mA it gives ~4.9V at 100mA, and I'm hoping my 5V circuit will need less than that.

I tried adding an op amp with a negative feedback through the transistor to improve regulation, but that doesn't work because my op amp's saturation voltage is too low (Vdd-1.5V).

I will add a LDO to my next digikey order for sure, though.

[Soeren]

Hi,

My LM7805 gives 4.4V with 6V.

As long as you get under 5V out, it's not regulating at all.
Try loading it with up to 1A and take a look at line and load regulation (when you can get access to a 'scope).

I understand your need for simplicity, but you can get a 7805 equivalent switcher ready to use (or use the simple discreete switchers from Roman Blacks hand).

[cyberfish]

Thanks for the suggestion.

SMPS sounds interesting and I read up on it a bit. I think I will stick with LDO for this one, though, since from 6V to 5V, the efficiency would be ~83%, so there's really no compelling reason to use a SMPS. An LDO would be able to regulate down to about 5.5V, at which time the battery would be pretty much dry anyways.

I will certainly keep that in mind for future projects, though.

[cyberfish]

Built my circuit using a Micrel MIC2940A LDO.

Schematics attached. Thought maybe someone will find it useful.

I use a comparator (LM339) for the "low battery indication" LED by comparing a fraction of the input voltage (using a pot), to the voltage drop of the "power on LED".

Any suggestions?

It seems to work fine in my testing. But no scope . I used a LM317T to give it 5.5V, and tried loading it up to 400mA (as much as my power pot will take...), still regulating fine.

Still waiting for my 6V NiMH pack to arrive... taking its time. For the time being I'm using a 9V to run my robot . Runs the battery dry in about 15 minutes (only 2 servos!)... getting expensive.