Post by thread:[EE]: 1N40xx voltage drop?

While doing some experiments, I have a diode protecting the A/D input, and thus would expect a 0.7V drop across it. But what I am finding is it is varying from .41 to around .56 volts as the voltage ramps from 0 to 5V. I can accept that it might not always be the golden 0.7V but I didn't really expect such a wide variance. So I went looking at the data sheets for an explanation but nothing really gave me a clue. I was almost expecting it to have a curve but nothing really pointed in that direction. Can anyone else offer some insight? _________________________________________________________________ MSN 8 helps eliminate e-mail viruses. Get 2 months FREE*. http://join.msn.com/?page=features/virus -- http://www.piclist.com hint: The list server can filter out subtopics (like ads or off topics) for you. See http://www.piclist.com/#topics

Not sure if this is the exact problem... but diode voltage drops are also functions of temperature. Is your environment pretty stable temperature wise??? Just a thought.. Quoting Micro Eng <spam_OUTmicro_engTakeThisOuTHOTMAIL.COM>: {Quote hidden}> While doing some experiments, I have a diode protecting the A/D input, and > thus would expect a 0.7V drop across it. But what I am finding is it is > varying from .41 to around .56 volts as the voltage ramps from 0 to 5V. I > can accept that it might not always be the golden 0.7V but I didn't really > expect such a wide variance. So I went looking at the data sheets for an > explanation but nothing really gave me a clue. I was almost expecting it to > have a curve but nothing really pointed in that direction. Can anyone else > offer some insight? > > _________________________________________________________________ > MSN 8 helps eliminate e-mail viruses. Get 2 months FREE*. > http://join.msn.com/?page=features/virus > > -- > http://www.piclist.com hint: The list server can filter out subtopics > (like ads or off topics) for you. See http://www.piclist.com/#topics > ---------------------------------------- This mail sent through http://www.mywaterloo.ca -- http://www.piclist.com hint: The list server can filter out subtopics (like ads or off topics) for you. See http://www.piclist.com/#topics

yes...its room temp, and all readings taken within minutes. I havent even approached thermal issues yet, just trying to figure out the diode voltage drop right now. _________________________________________________________________ Add photos to your messages with MSN 8. Get 2 months FREE*. http://join.msn.com/?page=features/featuredemail -- http://www.piclist.com hint: The list server can filter out subtopics (like ads or off topics) for you. See http://www.piclist.com/#topics

Hi, How much current is going through the diode? A diode is somewhat like a nonlinear resistor. I.e., the voltage drop still depends on the current going through it, it's just that for a reasonable range of current (say 100uA to 1 amp) the drop will be very close to 0.65 V at room temperature for most silicon diodes. If you have a diode in an ultra high impedance circuit, though, it is entirely possible for it to have a very small voltage drop. The relationship between voltage and current is given by: I = Is*(exp(v/Vt)-1) Where Is depends on the diode material, geometry, and temperature, and Vt depends only on temperature (it equals k*T/q where k is Boltzman's constant 1.38x10^-23 Joule/Kelvin, q is the charge of an electron 1.6x10^-19 Coulomb, and T is temp in Kelvins) and is about 0.026 at 300 K (27 C). From this equation, you would assume that I has a negative temperature coefficient. However, since Is is also temp dependent, at least for silicon, there is an overall positive temp coefficient (i.e., for a given voltage drop, current goes up with temperature) because Is has a much larger positive temp coeff. If you measure a few voltage, current point pairs, you should be able to find a value of Is that fits your diode well and then use this equation to predict what the voltage drop will be for almost any current. However, I don't understand why you need to do this. What does your circuit look like? How are you using the diode as a protection? Usually when diodes are used as a protective device to a measurement input, they are not in series but rather go between that input and the supply rails, to clip it if it goes outside of some voltage range. Sean At 07:31 AM 3/14/2003 -0700, you wrote: {Quote hidden}>yes...its room temp, and all readings taken within minutes. I havent even >approached thermal issues yet, just trying to figure out the diode voltage >drop right now. > > >_________________________________________________________________ >Add photos to your messages with MSN 8. Get 2 months FREE*. >http://join.msn.com/?page=features/featuredemail > >-- >http://www.piclist.com hint: The list server can filter out subtopics >(like ads or off topics) for you. See http://www.piclist.com/#topics -- http://www.piclist.com hint: The list server can filter out subtopics (like ads or off topics) for you. See http://www.piclist.com/#topics

On Friday 14 March 2003 06:07 am, Micro Eng wrote: > While doing some experiments, I have a diode protecting the A/D > input, and thus would expect a 0.7V drop across it. But what I am > finding is it is varying from .41 to around .56 volts as the > voltage ramps from 0 to 5V. I can accept that it might not always > be the golden 0.7V but I didn't really expect such a wide variance. > So I went looking at the data sheets for an explanation but > nothing really gave me a clue. I was almost expecting it to have a > curve but nothing really pointed in that direction. Can anyone > else offer some insight? The forward voltage varies with current and temperature. This data sheet shows the curve between 10mA and 15A: http://www.diodes.com/datasheets/ds28002.pdf The 1N400x is not a good diode to protect inputs with; it's really big, and it's slow. Consider using a small Schottky diode (I use the BAT54S series-connected dual surface mount diode for input protection). The forward voltage will be less, and it'll be smaller. The BAT54 has a forward voltage of 240mV max at 100uA, and 400mV max at 10mA. -- Ned Konz http://bike-nomad.com GPG key ID: BEEA7EFE -- http://www.piclist.com hint: The list server can filter out subtopics (like ads or off topics) for you. See http://www.piclist.com/#topics

One slight addendum to what I just wrote: diodes also have some internal pure series resistance that can, in some current ranges, dominate over the diode voltage drop formula, so you might need to take that into account, too (although since you are seeing very small voltage drops, it is probably negligible in your case). Sean At 07:31 AM 3/14/2003 -0700, you wrote: {Quote hidden}>yes...its room temp, and all readings taken within minutes. I havent even >approached thermal issues yet, just trying to figure out the diode voltage >drop right now. > > >_________________________________________________________________ >Add photos to your messages with MSN 8. Get 2 months FREE*. >http://join.msn.com/?page=features/featuredemail > >-- >http://www.piclist.com hint: The list server can filter out subtopics >(like ads or off topics) for you. See http://www.piclist.com/#topics -- http://www.piclist.com hint: The list server can filter out subtopics (like ads or off topics) for you. See http://www.piclist.com/#topics

The "rule of thumb" that a diode dropd a constant 0.7 volts is really inaccurate. A diode really has an exponential voltage drop with current. At picoamps this might be tenths of a volt, and at a few amps the drop might be 2-3 volts (just before it goes kablooey). It just happens to knee over at about 0.65 volts and stay in that region for a lot of practical currents. Got Spice? See if you can inject current into a diode and look at it's response. -- Lawrence Lile Micro Eng <.....micro_engKILLspam@spam@HOTMAIL.COM> Sent by: pic microcontroller discussion list <PICLISTKILLspamMITVMA.MIT.EDU> 03/14/2003 08:07 AM Please respond to pic microcontroller discussion list To: .....PICLISTKILLspam.....MITVMA.MIT.EDU cc: Subject: [EE]: 1N40xx voltage drop? While doing some experiments, I have a diode protecting the A/D input, and thus would expect a 0.7V drop across it. But what I am finding is it is varying from .41 to around .56 volts as the voltage ramps from 0 to 5V. I can accept that it might not always be the golden 0.7V but I didn't really expect such a wide variance. So I went looking at the data sheets for an explanation but nothing really gave me a clue. I was almost expecting it to have a curve but nothing really pointed in that direction. Can anyone else offer some insight? _________________________________________________________________ MSN 8 helps eliminate e-mail viruses. Get 2 months FREE*. http://join.msn.com/?page=features/virus -- http://www.piclist.com hint: The list server can filter out subtopics (like ads or off topics) for you. See http://www.piclist.com/#topics -- http://www.piclist.com hint: The list server can filter out subtopics (like ads or off topics) for you. See http://www.piclist.com/#topics

Diode voltage drop is a non-linear function of current. Bob Ammerman RAm Systems ----- Original Message ----- From: "Micro Eng" <EraseMEmicro_engspam_OUTTakeThisOuTHOTMAIL.COM> To: <PICLISTspam_OUTMITVMA.MIT.EDU> Sent: Friday, March 14, 2003 9:31 AM Subject: Re: [EE]: 1N40xx voltage drop? {Quote hidden}> yes...its room temp, and all readings taken within minutes. I havent even > approached thermal issues yet, just trying to figure out the diode voltage > drop right now. > > > _________________________________________________________________ > Add photos to your messages with MSN 8. Get 2 months FREE*. > http://join.msn.com/?page=features/featuredemail > > -- > http://www.piclist.com hint: The list server can filter out subtopics > (like ads or off topics) for you. See http://www.piclist.com/#topics > -- http://www.piclist.com hint: The list server can filter out subtopics (like ads or off topics) for you. See http://www.piclist.com/#topics

Excellent information, pretty much exactly what I was assuming, and looking for data to back that assumption up. To answer the question why is the diode in circuit...I believe to protect from reverse polarity installations. I'm going to sit down again, this afternoon, modify the method of controlling the input voltage (right now its a pretty loose control) and do exactly that....take a number of measurements again and see if I can predict what the voltage drop will be given any input voltage. The circuit resistance should be fairly constant, such that the current draw also will remain as such. I think part of my problem was the input voltage may have been drifting somewhat so it skewed my numbers in some of the data points. Otherwise, it would have made sense that the actual drop was less than 0.6V and at least consistant thru each datapoint (but some were way off which led me to believe something else might have been going on). Thanks to everyone that answered the question. _________________________________________________________________ -- http://www.piclist.com hint: The list server can filter out subtopics (like ads or off topics) for you. See http://www.piclist.com/#topics

On Friday 14 March 2003 07:43 am, Micro Eng wrote: > Excellent information, pretty much exactly what I was assuming, and > looking for data to back that assumption up. > > To answer the question why is the diode in circuit...I believe to > protect from reverse polarity installations. Instead of a diode in series, why not have a resistor in series, connected to diodes to V+ and ground. This will protect against reverse polarity while not introducing a variable voltage drop. -- Ned Konz http://bike-nomad.com GPG key ID: BEEA7EFE -- http://www.piclist.com hint: The list server can filter out subtopics (like ads or off topics) for you. See http://www.piclist.com/#topics

> > To answer the question why is the diode in circuit...I believe to > > protect from reverse polarity installations. > > Consider placing a REVERSE biased diode from input to ground, > plus a series fuse if the input has any power capability. Approx no > action on normal voltage, clamps to ground on reverse voltage > > RM And definitely a Schottky. A 1N40xx will not protect the A/D pin from negative excursions, you need a diode with a smaller V drop than the PIC's internal diode -- http://www.piclist.com hint: The list server can filter out subtopics (like ads or off topics) for you. See http://www.piclist.com/#topics