Basic electronics help

**maxxell** · 05-26-2005, 08:06 PM

Both 9v sources?
Maybe it's my shoddy schematic-making. There's only one 9v battery in there. The 9v at the top is the positive terminal and the 9v at the bottom is the negative terminal.

**mhornnt** · 05-26-2005, 08:19 PM

based on that schematic, adjusting the trimpot should brighten/dim the leds. increasing its resistance is gonna drop the voltage across the LEDs, which will dim them, and the opposite in the case of lowerin its resistance. Only think i could think of that might cause you not seein difference as you adjust it is that its resistance value is so small compared to the photo reisistors that adjusting it just is only creating a negligible change in the voltage across the diodes, small enough that you arent seein a change in brightness.

**migel628** · 05-26-2005, 08:20 PM

sorry, it looked like you had 2 9v sources in the schematic

**mp3hombre** · 05-26-2005, 08:36 PM

First of all, the photocells (photoresistors) should be the same type. Their resistance in dark conditions might be too high, the leds are not getting enough "juice" no matter how much you spin the pot.
To bring the resistance down, try adding a resistor across PR1 - PR2 or across each photocell as in PR3 and PR4. Start with 20Kohms.
Experiment, experiment!! And keep us posted.

**maxxell** · 05-26-2005, 11:06 PM

Why does adding a resistor in those places reduce the resistance to the LEDs?

**mp3hombre** · 05-27-2005, 10:58 PM

it "reduces" the photocell resistance.
Resistors in series (one after another) add resistance to a circuit, resistors in parallel (side to side)divide their resistance. For ex. Two 10k resistors in series=20k. The same resistors in parallel=5k.

**maxxell** · 05-28-2005, 11:00 AM

Originally Posted by mp3hombre

it "reduces" the photocell resistance.
Resistors in series (one after another) add resistance to a circuit, resistors in parallel (side to side)divide their resistance. For ex. Two 10k resistors in series=20k. The same resistors in parallel=5k.

Okay, I think I get what you are saying. So the photoresistors apply less resistance, but since the new resistors are 'up-stream' of the LEDs, wouldn't you add back up the resistance of all the devices up-stream and have the LEDs end up behind the same amount of resistance?

It's very possible I just dont understand this, thanks for taking a minute and trying to help me out.

**maxxell** · 05-28-2005, 11:21 AM

Alright, new day of experimentation.

I think I found out why it was misbehaving before. I think the trimpot I was initially using was really small. A total resistance of <1k ohm to share. I found a trimpot that is as the book recommends, 20k ohm to share, and now it works as I would expect.

Neat, right?

**maxxell** · 05-28-2005, 12:14 PM

Next experiment: (This has become somewhat of a log of my experiments more than any individual question)

I decided to simplify things and just use the two bigger photoresistors. I ran them in parallel as before, and stuck them together so they can be assumed to be receiving the same amount of light. I removed them from the circuit completely, and set the trimpot so that there is equal voltage down each path. Then I put the photoresistors back in their place, taped together. I then tested a variety of lighting conditions and checked the voltage just before each LED under each scenario.

Here are the results:
(Y & G are the different paths, so-named because of the color of the alligator clips I am using for each path)
Finger over PRs - Y: 1.639V G: 1.69V
Ambient light - Y: 1.72V G: 1.738V
Under little lamp - Y: 1.739V G: 1.744V
Right next to 25W bulb - Y: 1.744V G: 1.746

The differences, therefore are:
Finger over PR: .051
Ambient: .018
Under light: .005
Next to bulb: .002

Result: It would seem that there is a non-linear difference in resistance in the PRs. Does this matter for my simple line-following robot?

I guess the ideal situation would be to find another Photoresistor whose resistance profile matches the non-linear difference between my current two. That would perfect the circuit, but is it really necessary?

I could even dial in a new trimpot setting so that they match closer to the middle of the spectrum rather than only in bright light, right? This seems the most sensible approach to the problem.