Author Topic: LM3914 led driver chip  (Read 2811 times)

0 Members and 1 Guest are viewing this topic.

Offline IconicPCBTopic starter

  • Super Contributor
  • ***
  • Posts: 1564
  • Country: au
LM3914 led driver chip
« on: August 01, 2019, 04:03:49 am »
Is there a second source LM3914 chip or at the very least a look alike device out there in the wilds of Borneo.
 

Offline thm_w

  • Super Contributor
  • ***
  • Posts: 7521
  • Country: ca
  • Non-expert
Re: LM3914 led driver chip
« Reply #1 on: August 01, 2019, 10:14:11 pm »
https://www.aliexpress.com/af/LM3914.html read reviews, some are fake or non-working.

If it doesn't have to be that specific pinout I would look for an alternative IC.
Profile -> Modify profile -> Look and Layout ->  Don't show users' signatures
 

Offline T3sl4co1l

  • Super Contributor
  • ***
  • Posts: 22436
  • Country: us
  • Expert, Analog Electronics, PCB Layout, EMC
    • Seven Transistor Labs
Re: LM3914 led driver chip
« Reply #2 on: August 02, 2019, 01:11:17 am »
Wonder if there's any money in programming a PIC with identical functionality.  Hmm.  I mean, it wouldn't be high voltage obviously, so would need a board with interface stuff.  Hmmmm. In a 0.3" wide DIP-18.  Hmmmm.

Big downside is sample rate, a good description of it is a continuous-time 10-bit unary flash ADC.  You'd need to run a regular SAR ADC quite fast to do that (and over three channels because of the divider and input, unless REF can be set up internally in the same way).  Though some abuse can be done because so few bits are needed (<= 4 bits, binary encoding).  Probably a SAR ADC can be clocked twice its rating and still get those bits correct?

Tim
« Last Edit: August 02, 2019, 01:14:48 am by T3sl4co1l »
Seven Transistor Labs, LLC
Electronic design, from concept to prototype.
Bringing a project to life?  Send me a message!
 

Offline IconicPCBTopic starter

  • Super Contributor
  • ***
  • Posts: 1564
  • Country: au
Re: LM3914 led driver chip
« Reply #3 on: August 02, 2019, 04:29:13 am »
A micro controller solution is a viable option along with multi op amp comparator "flash converter".

While they are an option it is a shame that so much analog functionality has been lost along the way.
 

Offline Richard Crowley

  • Super Contributor
  • ***
  • Posts: 4321
  • Country: us
  • KJ7YLK
Re: LM3914 led driver chip
« Reply #4 on: August 02, 2019, 04:44:24 am »
LM3914 is listed as "active" on the TI website.
But the LM3915 and LM3916 have left the building.

I started a similar thread at:
https://www.eevblog.com/forum/microcontrollers/replacement-for-lm3915/msg1087514/#msg1087514
 

Offline IconicPCBTopic starter

  • Super Contributor
  • ***
  • Posts: 1564
  • Country: au
Re: LM3914 led driver chip
« Reply #5 on: August 02, 2019, 07:48:36 am »
TI also list an LED driver chip capable of driving 16 LEDS however it is bereft of any analog capability ( an I2C bus port expander in essence ) controlled by a small micro, a maiden engineer's dream.
 

Online Zero999

  • Super Contributor
  • ***
  • Posts: 20357
  • Country: gb
  • 0999
Re: LM3914 led driver chip
« Reply #6 on: August 02, 2019, 08:36:23 am »
LM3914 is listed as "active" on the TI website.
But the LM3915 and LM3916 have left the building.

I started a similar thread at:
https://www.eevblog.com/forum/microcontrollers/replacement-for-lm3915/msg1087514/#msg1087514
The problem is they're expensive. Look at the prices in Digikey. Of course they can be bought cheaper elsewhere, but are they genuine of fake?
https://www.digikey.co.uk/products/en/integrated-circuits-ics/pmic-display-drivers/729?k=LM3914
 

Offline SiliconWizard

  • Super Contributor
  • ***
  • Posts: 15797
  • Country: fr
Re: LM3914 led driver chip
« Reply #7 on: August 02, 2019, 01:07:28 pm »
Agreed. I would doubt the cheap ones are genuine. At best, they may be old stock, used and salvaged ICs, etc.

Note that at Digikey, the IC is only available as an SMD one. We don't know what the OP was looking for exactly.

The solution would depend on the project. For any kind of new project, I would definitely go for an MCU-based solution. Cheap, much more flexible, can be integrated with many other features in one MCU, no availability problem, lower power consumption, the list goes on... And if said project already includes an MCU, this would even be a no-brainer.

If it's to maintain a legacy design, then it's not the same problem. Doing what T3sl4co1l suggested may be a relatively simple option.
 

Offline westfw

  • Super Contributor
  • ***
  • Posts: 4349
  • Country: us
Re: LM3914 led driver chip
« Reply #8 on: August 03, 2019, 09:11:24 am »
Quote
Big downside is sample rate, a good description of it is a continuous-time 10-bit unary flash ADC.  You'd need to run a regular SAR ADC quite fast to do that (and over three channels because of the divider and input, unless REF can be set up internally in the same way).
Um.  How fast IS a 3914?  (I didn't see anything in the spec sheet.)Aren't they usually used to provide human-consumable visuals for audio-frequency inputs?  For that you'd only need 100Hz or so, which would be pretty trivial with any modern microcontroller.
Coincidentally, I've recently been toying (not built yet) with this "bargraph" idea.  8pin AVR, charlieplexed 10 element display:
 

Offline T3sl4co1l

  • Super Contributor
  • ***
  • Posts: 22436
  • Country: us
  • Expert, Analog Electronics, PCB Layout, EMC
    • Seven Transistor Labs
Re: LM3914 led driver chip
« Reply #9 on: August 03, 2019, 10:26:41 am »
See that's the tricky bit, isn't it?  If you're using them for dumb displays, who cares.  If it's not in the spec sheet, who cares?  (I forget if I've looked in detail, or measured it, but it's basically a bigger LM339, AFAIK, probably biased a bit lighter.)  But I've used them as ADCs proper, where the ~microsecond response time allows it to be used in a control loop (namely, that was a power DAC application, resistors driven by the 3914 and the gaps filled in with a smaller current sink).

Tim
Seven Transistor Labs, LLC
Electronic design, from concept to prototype.
Bringing a project to life?  Send me a message!
 

Offline Tegra

  • Newbie
  • Posts: 9
  • Country: ca
Re: LM3914 led driver chip
« Reply #10 on: September 06, 2019, 06:00:36 pm »
About a year ago I was informed that the DIP LM3914 was in "lifetime buy" state. Kind of like last call at the bar. Appears the SMD is still a going concern.

These are a rather pricey chip but have had good success with the "ebay versions". (by its very nature it is not a precision device)
However, as this source may also "dry up" have decided to update the design using an inexpensive 3 digit voltmeter module instead.

   T
 

Offline floobydust

  • Super Contributor
  • ***
  • Posts: 7678
  • Country: ca
Re: LM3914 led driver chip
« Reply #11 on: September 06, 2019, 07:30:30 pm »
Replicating a LM3914 in software is very difficult.
The LED outputs are not hard on/off, they can go to partial brightness. The LM3914 also has dithering.
So you would need an MCU to PWM the LED's with some random jitter added, to preserve its resolution.

Example is a 3 LED analog bargraph verses MCU driving 3 LED's. The analog bargraph can give over 6 distinct levels, but the "digital" bargraph has only 3.
 

Offline westfw

  • Super Contributor
  • ***
  • Posts: 4349
  • Country: us
Re: LM3914 led driver chip
« Reply #12 on: September 07, 2019, 02:44:17 am »
Quote
The LM3914 also has dithering.
It does?  I guess I was misled by the "simplified internal schematic" in the TI datasheet; there are apparently some tricky things you can do with the MODE and REFERENCE inputs.  But I don't see dithering, specifically?
Quote
The analog bargraph can give over 6 distinct levels [on three LEDs]
You mean like this:  https://www.pearl-hifi.com/06_Lit_Archive/02_PEARL_Arch/Vol_16/Sec_53/Pease_Lab_Notes/Dithered_Bar_Graph_Display.pdf ?
I guess.  Takes additional chips.  I'm pretty sure I can do the same thing in software, although getting the exact effect might be ... annoying.
 

Offline T3sl4co1l

  • Super Contributor
  • ***
  • Posts: 22436
  • Country: us
  • Expert, Analog Electronics, PCB Layout, EMC
    • Seven Transistor Labs
Re: LM3914 led driver chip
« Reply #13 on: September 07, 2019, 11:00:43 am »
The comparator gains aren't terrifically high and don't have hysteresis, so there is a small linear range between steps; or it may self-oscillate from coupling, or dither from noise.

A large resistor from each output, to the bottom of the resistor divider (and a small resistor from there to ground), implements hysteresis per step, if you're so inclined.

Tim
Seven Transistor Labs, LLC
Electronic design, from concept to prototype.
Bringing a project to life?  Send me a message!
 

Offline SiliconWizard

  • Super Contributor
  • ***
  • Posts: 15797
  • Country: fr
Re: LM3914 led driver chip
« Reply #14 on: September 07, 2019, 02:56:45 pm »
Replicating it exactly will be obviously difficult, all the more that some of the "special" features mentioned are either undocumented, or pretty vaguely, and some seem to rely mostly on field experience with the chip?

Now for people willing to just use this as a bargraph driver, implementing it in a small MCU is relatively easy. Even if you have to do PWM for each LED output and implement some overlap in software. It's not rocket science either. Will it be exactly like a 3914? Nope. Will it matter? Only for those using it in an extremely specific way. Probably not for a typical vu-meter.

 

Offline floobydust

  • Super Contributor
  • ***
  • Posts: 7678
  • Country: ca
Re: LM3914 led driver chip
« Reply #15 on: September 07, 2019, 11:15:53 pm »
I couldn't find proof of my fuzzy memory that an LM3914 datasheet claimed the IC has dithering. It's probably just noisy comparators. Datasheet mentions ~1mV hysteresis.
180 page Electronics Magazine circuit designer's notebook, a bit of Jim Williams circuits when he worked For National Semiconductor too:
see pdf pg. 154 AB-25 Bob Pease 'Dithering' display expands bar graph's resolution
same lab note with last page text clipped.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf