Voltage Divider Question

[exelanoz]

Hello Hello

I have visited www.darreltaylor.com (http://www.darreltaylor.com) to use his voltage divider calculator.

I notice that on his circuit diagram in addition to R1 and R2 he has C1 and D1.

I guess that C1 is some kind of non-polarised cap... and that D1 is some kind of zener diode...

Am I halfway there?

So.....I have used voltage dividers before, but never came across C1 and D1. Can someone ( or Darrel :-)))) please explain to me the function of those additional components and a suggested value to use?

Many thanks

(link http://www.pbpgroup.com/modules/wfsection/article.php?articleid=25 )

[Darrel Taylor]

Nothing too spectacular.

The capacitor can filter out some of the noise if it's an analog input, and/or add debouncing if it's digital on a Schmitt trigger PIN.

The Zener is only good for digital circuits that might have spikes or other voltages exceeding the dividers maximum design voltage. Don't use a Zener with Analog circuits.
<br>

[aberco]

I have also used this great little tool!

I am trying to use this setup but with high resistor value to avoid power losses. I understand that PIC ADCs requires a maximum impedance of about 1kohms for greater accuracy. However this would provide a large current draw.

My setup is as follow:

http://aberco.free.fr/imagesdivers/voltdivid.jpg

I use the resistor divisor to charge the capacitor, which is used to store the required amount of charges needed for a conversion. The drawback of this setup is that it lowers the number of samples per seconds, but for checking battery level it shouldn't be a big issue.

With the following setup I have a 23µA power drain and a time constant of 2.2 mS for the capacitor. Considering that 5 time constants are necessary to have 99% charge, let's say 10mS are needed to fully charge the capacitor. This means that I should space ADC sampling to no more than 100 times/s.

However, using the regular ADCIN gives me a first reading which is accurate, then either 0 or 1 for the next readings. I've also tried to manually configure the ADC but I have a hard time to understand how to configure it. Should I disable the dedicated RC oscillator and perform a manual "one-shot" sampling? The sampling time itself should not be limited as the 10nF capacitor provide the required charge to the ADC.

[aberco]

I don't really know what I did but I get accurate readings now... who knows! I'll try to understand it later on and post it here.

[malc-c]

Anyone know how to reduce voltage but still provide suitable current.

A friend needs to reduce 24v DC to 12v DC, but needs to drive a load of around 250mA. Whilst voltage dividers work well at providing a reference voltage, they appear impractical when you place a load across (ie in parallel with) the second resistor.

Using a 12v regulator is also impractical for him as the heat generated to dissipate 12V means it requires a large heatsink.

[aberco]

What about a switching regulator? they are usually quite efficient and don't require an heatsink. 250mA is not much, and I'm sure that even an LDO (low dropout regulator) could perform well. 7812 has a large voltage drop and tends to heat.

You can look at the catalog from TI, Maxim, ST and others, they most certainly have the device that suits your need (i.e. TL750M12 from TI in TO220 package)

[Ioannis]

Or from RECOM. Just like a 78xx regulator, only switching. No heat sinking required.

Ioannis

[malc-c]

Thanks guys - I'll pass on the suggestions. Can you provide a part number or link for the RECOM switching regulator - Cheers

[Ioannis]

Sure, here it is for 1A. There are others for .5 or 1.5A

http://www.recom-international.com/?id=48&no_cache=1&user_recom_pi2[series]=R-78Bxx-1.0(L)

Ioannis

[malc-c]

Thanks - I'll pass on the info