Second source horror stories

They make great "current limiting resistors" for common cathode LED 7 segment modules in red, yellow, yellow-green...just connect ref to the TL431 anode and stick them between the module cathode and ground, to drive the display segments constant-voltage. when driven from 5 volts it keeps about a 2.5 volt drop across the segments however many are illuminated.

and with the small 7 segment displays the brightness always seems to end up just about right. No change with temperature

That seems backwards. It is regulating voltage which is supposed to be the wrong way to drive LEDs. Each LED has a different IV curve and those curves vary with temperature - all while the curves are rather steep.

I guess it falls under the heading of "good enough".

--

  Rick C. 

  - Get 1,000 miles of free Supercharging 
  - Tesla referral code - https://ts.la/richard11209

Like a 555, the 431 can be a lot of things if you connect it cleverly. One is a current source.

Cheers

Phil

--
Dr Philip C D Hobbs 
Principal Consultant 
ElectroOptical Innovations LLC / Hobbs ElectroOptics 
Optics, Electro-optics, Photonics, Analog Electronics 
Briarcliff Manor NY 10510 

http://electrooptical.net 
http://hobbs-eo.com

it

nd

the wrong way to drive LEDs. Each LED has a different IV curve and those curves vary with temperature - all while the curves are rather steep.

Sure, but that's not what he was doing. His circuit drives multiple segmen t LEDs from a common cathode connection. A constant current source would m ean the constant current would be split up between the LEDs that were on.

1 segment on gets all the current. 7 segments on each one gets 1/7th the c urrent.

His circuit is just a voltage dropper. Drive one segment and it gets whate ver current the LED draws at that voltage. Drive seven segments and each o ne will get the current it draws at that voltage which will remain constant with temperature while the diode IV curve will vary.

I guess he sees even illumination of all segments as the LEDs most likely a re from the same batch and so are pretty close. Hard to imagine the bright ness won't change with temperature.

I think he would do better with resistors in each anode leg. At least it w on't be constant voltage as the diode current varies so the current won't v ary as much.

--

  Rick C. 

  + Get 1,000 miles of free Supercharging 
  + Tesla referral code - https://ts.la/richard11209

We synchronize all the switchers in a box to a PWM output on the micro so 1) there can't be any heterodyning anywhere we care about and 2) the one clock can be moved if it's interfering with a radio (if sub-AM band switchers are used where there is an AM receiver, the micro will tune the switchers away from the receiver).

SS switchers are not going to heterodyne!

I wonder how they do it.

--

John Larkin         Highland Technology, Inc 

lunatic fringe electronics

They can't be used to dodge receivers, either.

You're just too used to designing jitter *out*. ;-)

The idea is that the EMI is just broadband noise. It's really a cheap trick to fool spectrum analyzer quasi-peak detectors and pass FCC and CE tests. Like the Volkswagen cheat thing.

I have considered using one of the switchers as a noise source. I have a 12-bit ADC that I want to test. I figure that if I feed it analog noise, I can histogram each of the bits and catch stuck bits and probably shorted bit lines.

--

John Larkin   Highland Technology, Inc   trk 

jlarkin att highlandtechnology dott com 
http://www.highlandtechnology.com

nsure

some

any

th,

limit

se)

e f s

the

almost any signal should work, and then just verify that you can get all zeros and every single bit word

Yes, it' a cheap trick to get around the rules. Receivers are still desensitized by switcher noise. It's even in-band.

Seems like a winner. With sufficient averaging you should be able to get some pretty decent results. Maybe a test against a known ADC (or DVM) to make sure your source isn't correlated noise.