![]() | ![]() | ![]() |
| | |||||||
| General Electronics Chat This forum is for general chat about electronics, eg: Dont know what a part does? Dont know how to read a circuit? Want to get an opinion? |
| | LinkBack | Thread Tools | Display Modes |
| | (permalink) |
| Hi all, I want to make a stable voltage reference, should I use LED or Zener diode? If I need a very low noise voltage output? What one is better if comparing with is the noise level of LED and Zener diode? Thx! 8899
__________________ 8899 | |
| |
| | (permalink) |
| They both are noisy devices for a good voltage reference. Another aspect is the drift of the voltage with temperature. I use a LM336 that outputs 2.5 V. That value can be amplified with an op amp and you can make the voltage you need. Whichever device you'll choose, add a RC filter between the output of that device and the non-inverting input of the op amp. This will reduce the noise bandwidth. Last edited by eng1; 8th July 2006 at 11:17 AM. | |
| |
| | (permalink) |
| As suggested, both a zener and an LED are fairly poor voltage references, but you don't even mention how good it needs to be or what it's for?. Personally I use the TL431, which is a variable precision reference chip, which you can adjust from 2.5V upwards - check my PIC analogue tutorial for an example. | |
| |
| | (permalink) |
| I've never considered using an LED as a voltage reference. I'm assuming you use the 2V drop across the LED as the reference. I wouldn't have thought that was very accurate? Brian | |
| |
| | (permalink) |
| A low voltage zener diode makes a lousy voltage regulator. An LED might be slightly better, which you can say is poor.
__________________ Uncle $crooge | |
| |
| | (permalink) | |
| Quote:
| ||
| |
| | (permalink) |
| I've used them as quick zener replacements for generating 1.8V for some low power CMOS logic. Beats the heck out of trying to get a reference divider for an adjustable regulator setup when in a hurry. | |
| |
| | (permalink) | |
| Quote:
The reason I said that I wouldn't have thought it would be very accurate is due to the difference in forward voltage required for conduction on different devices - even devices with the same part number. I presume the difference is not great enough to cause any issues, if it's common to use them as references. You do have an increased current consumption though - a zener typically consumes 10mA where as an LED would probably consume closer to 20mA. Mind you, voltage regulation using this kind of approach is not efficient anyway (most of the power is wasted in heat generated by the current limiting resistor for the zener) so 10mA here or there probably isn't much of a concern. Brian | ||
| |
| | (permalink) |
| Thinking about it, you can get low current LEDs and you certainly wouldn't bother to use a high brightness one as a reference so it probably isn't less efficient to use an LED rather than a Zener afterall. Brian | |
| |
| | (permalink) | |
| Quote:
| ||
| |
| | (permalink) |
| I'm with Nigel on the TL431 chip. It's precision enough for most apps.
__________________ Don't make me reach through this monitor to slap you a good one! | |
| |
| | (permalink) |
| A led is a LOT more stable than a low voltage zener. And there is no reason to use the full operating current. I have used an ordinary dim green LED as a voltage reference in a successful commercial design running at a not very well specified current ( the sleep current of a CMOS real time clock), of somewhere between 100nA and 1uA, with no selection of components. | |
| |