Blue LED at 3.3V

Excuses! Excuses! Excuses! Where does a capacitor belong? (Giving you the benefit of even having a clue :-) ...Jim Thompson

--
| James E.Thompson, CTO                            |    mens     |
| Analog Innovations, Inc.                         |     et      |
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    |
| Phoenix, Arizona  85048    Skype: Contacts Only  |             |
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  |
| E-mail Icon at http://www.analog-innovations.com |    1962     |
             
      The only thing bipartisan in this country is hypocrisy

Stray?

Grant.

--
http://bugs.id.au/

Sure. CMOS ICs have input capacitance. I guess JT doesn't know much about CMOS ICs.

I obviously left out the cap, on purpose, to hit the "one resistor" target. It *will* oscillate without a discrete cap. What else could it possibly do?

John

how

At a ridiculous speed, IF at all.... giving him the doubt that he "left off the bubble"... what a dork fish :-)

Larkin huffs a lot. While he huffs and puffs and blows-hard, I design custom CMOS every day, but I leave the bull shit out of mine.

Has Larkin EVER posted a working circuit... one that didn't need additional bull shit as an added ingredient ?:-) ...Jim Thompson

--
| James E.Thompson, CTO                            |    mens     |
| Analog Innovations, Inc.                         |     et      |
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    |
| Phoenix, Arizona  85048    Skype: Contacts Only  |             |
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  |
| E-mail Icon at http://www.analog-innovations.com |    1962     |
             
      The only thing bipartisan in this country is hypocrisy

I find the spectrum to be smooth, but not even.

I see the spectrum running low in truly and fairly violet wavelengths. (Shorter than 430 nm, though that actually means little as far as practical illumination is concerned since in that part of the spectrum, human vision has low sensitivity.)

Another irregularity that I see is the "blue peak" - usually visible to me as a "brightish hump" in the spectrum, even if I merely use a CD as a diffraction grating to view the spectrum. The reduction in diversity of blue/bluish wavelengths has some negative impact on color rendering properties, probably minor.

The next irregularity that I see is a dip/weakness in the blue-green to bluish side of mid-green. I find that to have a minor negative effect on color rendering. Another thing I find that to have is to make the light's "scotopic/photopic ratio" or "s/p ratio" less than that of daylight or blackbody radiation of the same color. That affects illumination ability in dimmer environments where scotopic vision or mesopic vision (both scotopic and photopic are significantly functioning) is the vision mode that is being used.

After that, comes the irregularity of surplus of yellow and yellow-orange and shortage of mid-green (even if shortage is only on bluish side of mid-green) and shortage of mid-red. (The "red shortage" is even worse in the deep red, but that matters less due to low sensitivity of human vision to deep red.) This irregularity causes dulling of the color of green things, and dulling and darkening of the color of red things.

As the years have gone by, it appears to me that the color rendering index of white LEDs (excluding enhanced or high CRI models) has drifted downwards, from 75-plus of older ones around and before 2000 to around

65 for the highest efficency recent ones as of 2009. One reason is that advancement of the phosphors used for these included narrowing the yellowish emission band of the phosphor to produce less bluish-green and blue-side-mid-green and less deep red and deeper mid-red, and more yellow and greenish-yellow. Another is that the "blue peak" wavelength has been shifted shorter in highest-efficiency white LEDs, from around 467 nm in the late 1990's to closer to 450 nm now. Shifting the "blue peak" towards 450 or upper 440's of nm makes that spectral feature hit the "blue peak" of human vision better, and that means less blue light has to escape the LED in question to make white light, and more of the blue light can be converted to yellowish light that human vision is more sensitive to in photometric terms. However, this shift does deepen the "blue-green valley" and also decreases s/p ratio.

For examples of white LEDs:

Older style spectrum: Nichia NSPW500S, NSPW500CS. Especially NSPW500S, some of which may still be available from some nationally-selling-to- hobbyists electronic surplus outfits. That LED is often noted as a 5600 MCD white LED. These have nominal color temperature of 6500 K, but appear to me more like 6000 K. I determined a Nichia NSPWF50DS to have s/p ratio of 2.3, which is only slightly short of the 2.36 of a 6000K blackbody or the 2.45 of 6500K.

An extreme newer style spectrum: Nichia NSPWR70CSS-K1, which Nichia claims to achieve 150 lumens out per watt in at 20 mA. I have personally tested these to achieve 140-145 lumens/watt at 20 mA by a method prone to erring on the conservative side. The nominal color temperature is 5000 K. I have determined s/p ratio to be 1.7, while a 5000K blackbody achieves

2.15. I have also determined *some* red object color/brightness rendering properties and a few other color rendering properties that I can determine numerically to be like those of a common fluorescent lamp whose CRI is 62, so I "guesstimate" that this LED's CRI (specifically the Ra8 CRI) is 62-65. I can name two other nominally 5000K LEDs stated by their manufacturer to have CRI of 65, at least IIRC: Citizen / Cecol CL-L233-C13N and CL-654-C1N. One was on and the other was hardly behind the bleeding cutting edge of overall luminous efficacy for LEDs of their power class when they became available, with convenient higher voltage drops (muti-chip), considering competition that was actually available then, even at ~3.2-3.5V voltage drop. The lower efficiency of those two even has its 8 chips isolated from each other (16 contact package) and IIRC rated to be paralleled with each other, allowing choice of 3 options for voltage drop.

--
 - Don Klipstein (don@misty.com)

Might need a cap at the input, but otherwise OK, I think. If you make the resistor big enough, the input capacitance of the inverter should make it work fine. Nice low parts count.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal
ElectroOptical Innovations
55 Orchard Rd
Briarcliff Manor NY 10510
845-480-2058
hobbs at electrooptical dot net
http://electrooptical.net

I think you'd need to tailor the rate to accommodate the source/sink capability of the inverter (not shown, bubble to be imagined :-) ...Jim Thompson

--
| James E.Thompson, CTO                            |    mens     |
| Analog Innovations, Inc.                         |     et      |
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    |
| Phoenix, Arizona  85048    Skype: Contacts Only  |             |
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  |
| E-mail Icon at http://www.analog-innovations.com |    1962     |
             
      The only thing bipartisan in this country is hypocrisy

Urinating contests aren't my cup of tea. ;)

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal
ElectroOptical Innovations
55 Orchard Rd
Briarcliff Manor NY 10510
845-480-2058
hobbs at electrooptical dot net
http://electrooptical.net

A typical blue LED won't actually be off with a Si diode in series and

3.3V, even without considering temperature and tolerance on the 3.3V.

Can be fixed with two Si diodes in series the way I drew it (but even that might be a bit marginal) or use a red LED for the diode.

Best regards, Spehro Pefhany

--
"it's the network..."                          "The Journey is the reward"
speff@interlog.com             Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog  Info for designers:  http://www.speff.com

how

But John "I left out the bubble" Larkin makes it so-o-o-o easy ;-)

I think he's the one going senile. He claims it's me, but I feel (and think) like an 18 year old :-) ...Jim Thompson

--
| James E.Thompson, CTO                            |    mens     |
| Analog Innovations, Inc.                         |     et      |
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    |
| Phoenix, Arizona  85048    Skype: Contacts Only  |             |
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  |
| E-mail Icon at http://www.analog-innovations.com |    1962     |
             
      The only thing bipartisan in this country is hypocrisy

how

Feeling that way is the first symptom. ;)

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal
ElectroOptical Innovations
55 Orchard Rd
Briarcliff Manor NY 10510
845-480-2058
hobbs at electrooptical dot net
http://electrooptical.net

how

much

cap,

Nah! I thinks like an 18 year-old, then I falls asleep ;-) ...Jim Thompson

--
| James E.Thompson, CTO                            |    mens     |
| Analog Innovations, Inc.                         |     et      |
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    |
| Phoenix, Arizona  85048    Skype: Contacts Only  |             |
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  |
| E-mail Icon at http://www.analog-innovations.com |    1962     |
             
      The only thing bipartisan in this country is hypocrisy

How about when I feel like an 18-year-old? When I do, there are usually none willing around! (Besides, I have exchanged vows with someone that I love, and I have fulfilled them for over 20 years without exception of so much as a second, and my main barrier to being married to him is marriage law where I live... unlike a fair portion of New England, along with all of Canada, Belgium, Spain, South Africa and Holland [IIRC].)

--
 - Don Klipstein (don@misty.com)

Using the specs from the 74HC14, in this circuit it oscillates at about

140Khz with a 1 meg resistor:

Version 4 SHEET 1 880 680 WIRE 128 192 112 192 WIRE 224 192 192 192 WIRE 112 256 112 192 WIRE 224 256 224 192 WIRE 224 256 192 256 FLAG 144 96 0 FLAG 144 16 Vcc FLAG 288 336 0 FLAG 288 112 Vcc FLAG 112 320 0 SYMBOL Digital\\schmtinv 128 128 R0 SYMATTR InstName A1 SYMATTR SpiceLine Td = 5n SYMATTR Value2 Vhigh=3.3 Vt=0.85 Vh=0.56 SYMBOL voltage 144 0 R0 WINDOW 123 0 0 Left 0 WINDOW 39 0 0 Left 0 SYMATTR InstName V1 SYMATTR Value 3.3 SYMBOL res 208 240 R90 WINDOW 0 0 56 VBottom 0 WINDOW 3 32 56 VTop 0 SYMATTR InstName R1 SYMATTR Value 1MEG SYMBOL cap 288 176 R90 WINDOW 0 0 32 VBottom 0 WINDOW 3 27 62 VTop 0 SYMATTR InstName C1 SYMATTR Value 10n SYMBOL res 272 176 R0 SYMATTR InstName R3 SYMATTR Value 220 SYMBOL LED 272 272 R0 SYMATTR InstName D1 SYMATTR Value NSCW100 SYMATTR Description Diode SYMATTR Type diode SYMBOL res 272 96 R0 SYMATTR InstName R2 SYMATTR Value 1k SYMBOL cap 96 256 R0 SYMATTR InstName C2 SYMATTR Value 3.5p TEXT -6 54 Left 0 !.tran 0.1

how

Yup, it's perfectly reasonable. 5 pF, 1M, tau is 5 usec. Right in the ballpark.

What else could it possibly do than oscillate?

John

--
If you don't have the time to devote to proving that what you profess
here is true, then why are you here?

Well, you'd expect something of the order of 2*R*C where C is something like five puffs, so 100kHz-ish, rough order of magnitude.

Best regards, Spehro Pefhany

--
"it's the network..."                          "The Journey is the reward"
speff@interlog.com             Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog  Info for designers:  http://www.speff.com

[...]

I didn't notice - it was pretty obvious to me what the idea was. It's a sketch of an idea, not a schematic entered into a cad system. I thought it was very nice, impressively minimalistic. I like the idea of using the input capacitance as a component. And as others have pointed out it in fact it runs at perfectly reasonable frequencies (lower than I would have guessed without calculating it).

[...]

--

John Devereux

Just to fool around with ideas and learn things.

No, that was a scribble, just an idea. I did no math and it had no values.

It could be useful, both for powering LEDs and for generating bias voltages, like a little negative voltage for a gaasfet. Or both at the same time. A schmitt oscillator deesn't need a capacitor just because you're used to seeing one.

Why do ideas offend you so?

I don't expect anything, except whining from you and JT.

The obvious one.

Tout? Did you used to be a newspaper reporter in junior high school or something?

Sloman doesn't do electronics. I had 5 or arguably 6 seriously good ideas yesterday, things that *will* be rigorously implemented and sold. And of course a bunch of silly ideas. If you aren't willing to create silly ideas, and make mistakes doing it, you're not going to ever come up with the seriously good ones. The rigor can come later.

The guys who immediately criticize ideas, instead of riffing on them, are always the guys who have few of their own.

So rail on!

John