voltage reference

[R.Lewis]

I am looking for a voltage reference (anything below 2.0V ) that is
reasonably repeatable and stable and needs work from -10 to +40c to within
10% max limit.

Any ideas for something that doesn't cost the earth - I only need a few K
pieces but price is very sensitive.

 

[Spehro Pefhany]

I am looking for a voltage reference (anything below 2.0V ) that is
reasonably repeatable and stable and needs work from -10 to +40c to within
10% max limit.

Any ideas for something that doesn't cost the earth - I only need a few K
pieces but price is very sensitive.

LMV431 $0.16/1k for 0°C~70°C , 1.24V nominal.


Best regards,
Spehro Pefhany

 

[Jan Panteltje]

I am looking for a voltage reference (anything below 2.0V ) that is
reasonably repeatable and stable and needs work from -10 to +40c to within
10% max limit.

Any ideas for something that doesn't cost the earth - I only need a few K
pieces but price is very sensitive.

LM317
1$

 

[John Popelish]

I am looking for a voltage reference (anything below 2.0V ) that is
reasonably repeatable and stable and needs work from -10 to +40c to within
10% max limit.

Any ideas for something that doesn't cost the earth - I only need a few K
pieces but price is very sensitive.

See:
http://www.national.com/catalog/AnalogReferences.html

 

[R.Lewis]

LM317
1$

Haven't got that sort of money

 

[John Fields]

On Wed, 7 Apr 2004 18:26:52 +0100, "R.Lewis" <h.lewis-not this

 

[Jan Panteltje]

Haven't got that sort of money

I am sure for a few K you can get them for 10 cents.
JP

 

[R.Lewis]

On Wed, 7 Apr 2004 18:26:52 +0100, "R.Lewis" <h.lewis-not this

I am looking (wishing) for 10 cents tops but was rather hoping someone would
recommend a Vbe of something or a forward diode that would fit my
requiremnts

 

[Jim Thompson]

On Wed, 7 Apr 2004 23:01:16 +0100, "R.Lewis" <h.lewis-not this

I am looking (wishing) for 10 cents tops but was rather hoping someone would
recommend a Vbe of something or a forward diode that would fit my
requiremnts

Dreamer ;-)

...Jim Thompson

 

[Jan Panteltje]

Dreamer ;-)

http://www.national.com/parametric/0,1850,645-c11-0,00.html
17 cent
Is that for single quantity?
Straight from the source.
JP

 

[Spehro Pefhany]

http://www.national.com/parametric/0,1850,645-c11-0,00.html
17 cent
Is that for single quantity?

1K pcs. Distributors may sell for more. My suggestion is 0.01 cheaper
(but you need a resistor) and (possibly more importantly) requires
something like 1/50 the current.

Best regards,
Spehro Pefhany

 

[Spehro Pefhany]

I am looking (wishing) for 10 cents tops but was rather hoping someone would
recommend a Vbe of something or a forward diode that would fit my
requiremnts

You could MAYBE use an LED (red) or a Vbe multiplier (2 resistors and
a BJT) and MAYBE meet +/-10% over that temperature range for half that
cost, but characterizing it probably wouldn't be worthwhile for the
meagre savings.

Best regards,
Spehro Pefhany

 

[Jim Thompson]

Why not use one of those CA3083 arrays and build your own bandgap ?

...Jim Thompson

 

[Boris Mohar]

On Wed, 7 Apr 2004 23:01:16 +0100, "R.Lewis" <h.lewis-not this

I am looking (wishing) for 10 cents tops but was rather hoping someone would
recommend a Vbe of something or a forward diode that would fit my
requiremnts

Use an LED.

 

[Jim Thompson]

$0.03 buys you a dual transistor BC847BPN, one NPN the other PNP.....in a
6-leg SOT23 package - how good a reference can you make with that? crude
current source driving temp-compensated vbe multiplier.....

How are you going to make a "temp-compensated vbe multiplier"?

What does a CA3083 sell for? I couldn't find a price on the web.

If there's interest, I'll design a bandgap around a CA3083... it's
pretty easy ;-)

...Jim Thompson

 

[R.Lewis]

On Wed, 7 Apr 2004 23:01:16 +0100, "R.Lewis" <h.lewis-not this


Use an LED.

Do you know of a led that fulfils the 10% requirement ?
I have not measured any specifically for this task but I seem to remember
getting a greater that 10% variation in Vf as they come out of the box,
plus the 2.something mV per degC tempco.

 

[Terry Given]

You could MAYBE use an LED (red) or a Vbe multiplier (2 resistors and
a BJT) and MAYBE meet +/-10% over that temperature range for half that
cost, but characterizing it probably wouldn't be worthwhile for the
meagre savings.

Best regards,
Spehro Pefhany

why stop there? Vbe multiplier with ntc (or ptc) 1st-order temperature
compensation......

 

[Terry Given]

Why not use one of those CA3083 arrays and build your own bandgap ?

...Jim Thompson

$0.03 buys you a dual transistor BC847BPN, one NPN the other PNP.....in a
6-leg SOT23 package - how good a reference can you make with that? crude
current source driving temp-compensated vbe multiplier.....

 

[Terry Given]

How are you going to make a "temp-compensated vbe multiplier"?

I was looking into a non-linear voltage divider - ntc or ptc at same (ish)
temp as bjt.

What does a CA3083 sell for? I couldn't find a price on the web.

mostly my point......can you still buy them?

If there's interest, I'll design a bandgap around a CA3083... it's
pretty easy ;-)

...Jim Thompson

I am certainly interested.....

 

[Terry Given]

I was looking into a non-linear voltage divider - ntc or ptc at same (ish)
temp as bjt.


mostly my point......can you still buy them?


I am certainly interested.....

assume an 0.65V forward drop at 25C. -10C gives about 727mV = +11.9%, 40C
gives about 617mV = -5.1%

using a Vbe multiplier with a 22k Rcb, and Rbe = 100k NTC paralleled with
27k, V = 1.349V at -10C = +1.97%, V = 1.323V at 25C and V = 1.373V at 40C =
+3.82%

I havent bothered to "design" this, just a quick fiddle with some NTC data I
happened to have in excel....shows that, over a limited temperature range,
can do OK with crappy old vbe multipliers and an ntc. If I actually designed
it (using scientific wild-assed guesses rather than wild-assed guesses) I
could probably do even better.....

If you used a BC847bpn with the pnp set up as a crude current source, this
would run on a very wide input voltage range too....and easily meet the
$0.17 price requirement