Bandgap Design

RST Engineering (jw) wrote: > I have no idea if Panasonic knows what they are talking about. Try > these: >

Both of those references echo the more detailed Panasonic pdf on charging techniques for NiMH. Note that not a one them gives any detailed information about charge termination voltage as a function of temperature. The 0.1C overnight charge method uses time termination because a voltage based termination is not feasible- that 1.8V/cell is a rough estimate for a last-ditch-save-the-battery termination. Panasonic does describe and in fact recommends the 0.1C method as the most power conservative and best to use in backup battery top-off applications, but not for general battery use chargers. But if you use that then your algorithm is check for battery Vbatt

I don't see how that works for making a fast charge...if the PVavg doesn't support 0.5C, it just can't be done.

Well, since we're all blowing our own large trumpets Jimmy, I'll point out my own horn. Its a BG that works down to a supply voltage of around

0.7V.

The example, as usually is in the SS download.

Kevin Aylward snipped-for-privacy@anasoft.co.uk

SuperSpice, a very affordable Mixed-Mode Windows Simulator with Schematic Capture, Waveform Display, FFT's and Filter Design.

It sounds like you have your answer already. However, "Designing Analog Chips" by Hans Camenzind has a chapter on bandgaps. It's available for download in PDF. Chapter 7. The book is targeted towards IC designers, but has lots of good information for the rest of us as well.

--
Regards,
   Robert Monsen

"Your Highness, I have no need of this hypothesis."
     - Pierre Laplace (1749-1827), to Napoleon,
        on why his works on celestial mechanics make no mention of God.

You are right that you can't raise the average current this way but if for some reason, you need pulses of current it can help. As I posted elsewhere, I remember reading somewhere that pulses of charging current are a good thing.

--
--
kensmith@rahul.net   forging knowledge

Jim...

As you so astutely pointed out on pages 2 and 3 of your treatise on bandgap design, by choosing the multiplier between R1 and R2 to be (epsilon) the math is made terribly simple. And, by choosing R1 appropriately, you can make the current arbitrarily low through the mechanism.

However (and I admit that I've only modeled it in Electronic Workbench for the time being)...

If you use 10K for R1, this makes R2 27K (using 5% values) and R3 1.2K. All very straightforward. The bandgap voltage comes out 1.320 volts (at room temperature).

Doing a temperature sweep from -15 to +40C, the curve is nearly a straight line from 1.332 volts (-15) to 1.317 volts (+40) giving a total delta of 15 millivolts over the range, or about 270 uV per degree C. Doing a monte carlo with 5% parts says that the bandgap will be somewhere between 1.25 and

1.4 volts under worst case conditions. Certainly we can adjust this small deviation out quite easily.

HOWEVER...

If we start diddling (excuse me, heuristically engineering) with the resistor values, just to see what happens, we find that the temperature curve may be made parabolic by an appropriate choice of R2 and R3, without much change in the percentage error of the monte carlo sweep.

For example, leaving R1 at 10K, making R2 56K and R3 2.7K, the bandgap voltage is 1.696 volts, the monte carlo gives us values from 1.6 to 1.8 volts worst-case, but the temperature sweep is a distorted parabola with the end temperatures (high and low) both being roughly equivalent at 1.680 volts, but a peak (inflection point) of the parabola at 1.7 volts at 22°C.

As a matter of fact, I can put the peak of the parabola at any temperature I choose by an appropriate choice for R3. Values of 2K and 3K will put the inflection point at -15 or +40°C respectively.

I can't explain the parabolic function. I've looked your equations over until I could probably recite them from memory and I can't find anything that looks like the equation of a parabola.

What am I missing?

Jim

Now there is a HELL of an idea that hadn't even crossed my mind. Any idea how the voltage varies over temperature?

Jim

The fine detail.

The TC of a Si junction ISN'T simply -2mV/°C

It has exponential terms plus T to the 3/2 power.

So my first order equations only get you close.

"Diddling" is an absolute part of real life.

From my first cut solution you then "diddle", either in the lab, or with a simulator until you get a parabolic (or sometimes "S") curve... as you will note from the various posted examples.

I don't often visit my own website, except to add things. Visiting to refresh my memory as to what's there, I note several corrupt bandgap files. I'll track down the originals and re-load.

Everyone, Please advise whenever you come across a mangled file and I'll fix them. Some of this stuff has been there for more than ten years... looks like the server hard-drive got corrupted during some website-provider restore operations :-(

...Jim Thompson

--
|  James E.Thompson, P.E.                           |    mens     |
|  Analog Innovations, Inc.                         |     et      |
|  Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    |
|  Phoenix, Arizona            Voice:(480)460-2350  |             |
|  E-mail Address at Website     Fax:(480)460-2142  |  Brass Rat  |
|       http://www.analog-innovations.com           |    1962     |
             
I love to cook with wine.      Sometimes I even put it in the food.

~ -2mV/C

James

You mean the gallium (insert a lot of other elements in here) arsenide sulfide anylide series of diodes have the same variation as silicon? I'd really be surprised. I'd be surprised if green and red had the same tempco, not to mention blue and white.

Jim

I read in sci.electronics.design that "RST Engineering (jw)" wrote (in ) about 'Bandgaps -- Jim T.', on Mon, 28 Mar 2005:

If the three independent resistor values really represent three degrees of freedom, it should be possible to produce a cubic function, which could give zero delta-V at three temperatures and a very small overall excursion. If not, it might even be worthwhile adding whatever is necessary to give a cubic function.

--
Regards, John Woodgate, OOO - Own Opinions Only.
There are two sides to every question, except
'What is a Moebius strip?'
http://www.jmwa.demon.co.uk Also see http://www.isce.org.uk

Yes, have a look at a LED as reference. Run it at 500uA to 1mA

Rene

--
Ing.Buero R.Tschaggelar - http://www.ibrtses.com
& commercial newsgroups - http://www.talkto.net

But remember to treat it like an engineer, and keep it in the dark.

Regards Ian

I read in sci.electronics.design that "RST Engineering (jw)" wrote (in ) about 'Bandgap Design', on Mon, 28 Mar 2005:

Well, it MIGHT not be true, but it is. It depends on q/kT and the bandgap energy, and it happens that, at least for Ge, Si and GaAs, the values of dVf/dT are all in the region of -2 mV/K at room temperature.

--
Regards, John Woodgate, OOO - Own Opinions Only.
There are two sides to every question, except
'What is a Moebius strip?'
http://www.jmwa.demon.co.uk Also see http://www.isce.org.uk

I took a quick look at some Stanley and Motorola data sheets and the tempco values for various colors are: Motorola IR -1.6 mV/K red -2.2 mV/k

Stanley red -2.0 mV/k green -2.1 mV/K white -2.1 mV/K blue -3.0 mV/K

So, while they are indeed all "in the region of -2 mV/K", there is still considerable variation by color and, apparently, manufacturer.

--
Tim Hubberstey, P.Eng. . . . . . Hardware/Software Consulting Engineer
Marmot Engineering . . . . . . .  VHDL, ASICs, FPGAs, embedded systems
Vancouver, BC, Canada  . . . . . . . . . . . http://www.marmot-eng.com

wrote (in

2005:

sulfide anylide series of diodes have the same variation

the same tempco, not to mention blue and white.

energy, and it happens that, at least for Ge, Si and

temperature.

values for various colors are:

considerable variation by color and, apparently,

The popular -2 mV/K TC for silicon depends not only on kT/q and the band gap, but the electron and hole mobilities and intrinsic carrier density, together more temperature dependent than the thermal voltage (kT/q). This is easily seen by differentiating the ideal diode equation solved for forward voltage w.r.t. T while holding I constant. You get something like Id = Is (exp(e q / kT) - 1) Id/Is + 1 = exp(e q / k T) e = ln(Id/Is + 1) * kT/q d(e)/dT = ln(Id/Is + 1) * k/q Note that the above suggests a positive TC, and is completely wrong because it ignores the termperature dependence of Is, which is where those other variables come in to create the negative TCs we actually observe.

--
--Larry Brasfield
email: donotspam_larry_brasfield@hotmail.com
Above views may belong only to me.

Brasfield also quoted the preceding post where the issue was apparent: Regarding "tempco" of various "diodes", "It depends on q/kT and the bandgap energy, and it happens that, at least for Ge, Si and GaAs, the values of dVf/dT are all in the region of -2 mV/K at room temperature." (from Woodgate) followed by "tempco values" ranging from "-1.6 mV/K" to "-3.0 mV/K" (from Hubberstey).

something?

To me, that appears to be more your objective. Anybody reading my post who can comprehend the issue in the post(s) I quoted would not have any difficulty recognizing the relevance of my post.

Your difficulty seeing it, or pretense to that effect, says much more about your motive than mine.

--
--Larry Brasfield
email: donotspam_larry_brasfield@hotmail.com
Above views may belong only to me.

Is this post-anything-at-all-with-disregard-for-useful-content day for you or something?

"John Woodgate" wrote in message news: snipped-for-privacy@jmwa.demon.co.uk...

wrote (in

but the electron and hole mobilities and intrinsic

figure*.

Surely. My comments about the factors affecting Is went more toward showing that its TC depends on some material properties that vary from one material to another rather than just kT/q which does not. There was a lingering implication in your post preceeding Mr. Hubberstey's that the TC might be expected to be the same for different semiconductor materials. (He expressed skepticism that different semiconductors would have the same TC of Vf. You came back with "the values of dVf/dT are all in the region of -2mV/K at room temperature."

At the time, I did not think you had made a strong enough statement to be wrong. I apogize if my quote left out the context showing that. I merely intended to show the issue on the table addressed by my post. (Somebody had alleged it to be irrelevant.) I was not trying to play the ever-silly Straw Man game.

--
--Larry Brasfield
email: donotspam_larry_brasfield@hotmail.com
Above views may belong only to me.