Voltage Regulator Wiring

[Rob Schaffer]

I am in the process of building a plane with Navigation LED's, and I would like to power them off the main flight battery. Providing I wire a 5V Voltage Regulator in battery connector (parallel with ESC), I should be able to still connect the servo lead from the light controller to the Rx in order to turn the landing lights on and off correct? This is presuming that I remove the Red (positive) wire from the servo lead plugged into channel 5.

Here's a diagram of what I am trying to accomplish. Please correct me if I am wrong with this logic. I am running the recommended servos already off the BEC, so I need to bypass that somehow, still allowing signal to the light controller.

Thanks for helping me out.

:cool:

EDIT: UPDATED PIC WITH REGULATOR INFO 8-21-03

[Fred Bronk]

Moving to Modeling Science.

FB

[clytle374]

I'll vote yes
cory

[Gary Warner]

Monolithic regulators need some capacitors to contain oscillations and switching noise.

If that is a 7605 or a variant, you might add some capacitors. To the input add a .01uf (some say "optional"- depends on noise carried on the wire) and to the output add .01 and 10uf (in parallel). All three caps reference ground. These will stop oscillations and reduce the idle current. You might also need a heat sink on the 7605, depending on current load.

Gary
--

[clytle374]

The capacitors sure won't hurt. I doubt that you would have a problem with out them, since it's a on/off thing and a small glitch won't be an noticed. LEDs don't care about noise either.

cory

[Rob Schaffer]

Last night I was able to find the 7805, which is a 5v DC regulator witn 1 Amp capacity, 7V minimum input. (I'll be running 8 cells anyway) My LED's and board only draw close to 200 mAh with everything active, so I have plenty of room on the Amps.

Just to confirm, the wiring with the Rx looks correct to allow the signal to pass through right? I wasn't sure where I should run the ground for the regulator or Lighting module, but what I have above seems logical.

Capacitors,... hmm, I might want to play it safe. Can you clarify their location?

Thanks for the help guys.

[Mr.RC-CAM]

The caps are required; do not omit them. Install them directly on the 7805 (not remotely).

I suggest a 10uF electrolytic on the input and .1uF mono-ceramic on the output. The data sheet shows where they go (see fig 5): http://rocky.digikey.com/WebLib/Fairchild/Web%20Data/LM7805_MC78xx_MC78xxA.pdf

A heatsink is needed too. You will be dissipating .6 watts, which will make for a very hot part without some external assistance. A typical TO-220 finned alum type will be fine.

With luck, Radio Shack should have everything you need.

RC-CAM

[clytle374]

Where do you get .6 watts, I get .92 plus a touch of internal.

[Mr.RC-CAM]

I get .92 plus a touch of internal.I agree. The VReg will be a hot little fellow with a full 200mA load.

RC-CAM

[Gary Warner]

I agree with the heat sink at 200ma. That pup is going to get hot. A TO220 76XX usually handles about 100ma max without a heat sink.

The value of the caps are from my general experience and you will find that they can vary greatly from one application to another.

So far as your wiring question, you’re right. It looks good with just the addition of the caps. If you are trying to use the minimal amount of caps, the .1uf (I've seen .01 to .47uf) on the output handles the majority of the oscillations. It should be placed physically on or as near to the regulator as possible. Note that these oscillations are internally generated by the regulator - they are not from "motor noise".

The larger of the caps (the 10uf) on the output is commonly used in TV’s (my expertise) to control load bounces that can upset sensitive circuits that may share the regulated line. I don't think this application will "require" the 10uf cap.

The battery pack offers a large enough capacitance to qualify as the input cap and since the wire inductance and current are low, the .33uf cap would offer little in the way of ripple reduction. The .01uf cap on the input that I suggested is to stop (more like reduce) the motor switching noise (RF) on the line from getting into the regulator. If input ripple is of concern, the use of a .33uf – 10uf cap (or round about there) and a 10uh coil will be very effective in stopping input ripple. If used, the coil goes to the battery + lead and into the regulator (in series with the input line) and the cap goes between the coil/regulator connection and ground. Keep in mind though, that this by far over engineering the regulator circuit , considering the task at hand.

Again, the .1uf (.01uf... what ever) on the output is the most critical and likely all you will need.

Good luck!

[clytle374]

I don't think it is safe to say that the battery has capacitance, I might be wrong, but it does make about the best possible voltage stabalizer.

The reason for the caps is to keep the regulator chip from oscilating.

I'm guessing that if the wires are short running to the battery, it would work fine with out caps. Most likely there is a cap in the lighing controll.

Don't forget the wrong inductor can add trouble:)

cory

[Mr.RC-CAM]

Actually, the battery makes a great capacitor. It is the ultimate coulomb collector. But, if the wire leads to the battery are long, then the Vreg's input cap must be used to ensure Vreg stability.

I agree that you can Muntz the Vreg design to use less parts. But I have first hand experience with poorly behaving 78xx parts (they make nasty RF transmitters if you cheat too much). As such, using the two mentioned caps have earned my respect. They are cheap and easy to install.

Rather than trying to find a low ESR .33uF input cap, as shown on the data sheet, it is easier to just use a readily obtained 10uF electrolyic. Their ESR rating is a fine for this application. The output cap should be a common .1uF mono-ceramic type or equiv. Keep the leads short and be sure to add a heatsink (or wear an oven mit).

RC-CAM

[clytle374]

I'll agree that it collects coulombs. Have you ever tried to check a nicads frequency response.
Just curious to see if it stabality is due to real "capacitance"
Does is pass high freq better than low?

cory

[Mr.RC-CAM]

Just curious to see if it stability is due to real "capacitance"Now I see where you are going. The stability is due to the battery's low ESR. The big battery pack is usually characterized as a low impedance current source, which promotes 78xx stability. Long wire leads impact this, so a more localized cap input is needed in some cases.

I promise not to add any more of my boring drival to this post. ;)

RC-CAM

[clytle374]

OK, I was just really curous as the longer I though about it, I kept thinking that therer are alot of parell conductors in a battery. I'm not very farmilar with NiCad construction.
cory

[Rob Schaffer]

Thanks for the info,... I'll be sure to add the heatsink and caps. Maybe this weekend I'll wire things up.

Thanks again :cool:

[Rob Schaffer]

Thanks for all the help guys. I was able to get everything I needed between the two Radio Shack's in my area, and I soldered things up on Saturday morning. All works well and it stays cool for the duration that I tested it. Now to finish building and installing things in the plane.

Thanks! :cool: