Zener Q

Here is a plot I did of ye olde LTZ1000:

Shows output voltage constant to within +/- 0.1ppm or so, over 8 hours and a 20K temperature swing. Apart from that negative spike, don't know wht that was.

Not the most practical of parts though.

Very pretty! IIRC the HP 3458A uses a bunch of LTZ1000s averaged together. I have a couple in the drawer but haven't ever actually used one.

Cheers

Phil HObbs

Nah, the 3458A has just the one LTZ1000.

You may be thinking of the voltage standards where they do do that.

The one I had under test was in a "environmental chamber" (bud beer fridge) with a decent temperature swing, whereas the meter temperature was reasonably constant. Though not actively controlled. Nonetheless part of the variation, and much of the noise, was likely the meter.

say 20C to

s.

r to

art.

worry

that

are a

goes

ut

uch

ones

ie

g a C

OK we buy the LM399 for ~$5 (/100) 2ppm/K (another data point for your plot.)

George H.

.highlandtechnology.com=A0 jlarkin at highlandtechnology dot com

LTZ1000 is ~$40, 0.05ppm/K.

Hmm, should be $140 according to Larkins formula. Bargain!

Check; know about tradition; that and $5 may get you a cup of coffee at StarBucks.

The Vishay datasheet lists both of those as having the lowest (and same) temperature coefficient: ± 0.030 %/C. The 6.2V zener is listed at 0.045 %/C,and the 6.8V zener is listed at

Don't you mean that if one took a higher voltage zener (with a positive TC) and put it in series with a diode (or DCT) that one might get something rather close to zero TC?

MotorBoat made TC diodes,most were around 8V.

Visions of Billy boy on his yacht?

The Apex parts (+/-0.6ppm/K maximum) are $80 apiece in small quantities.

Best regards, Spehro Pefhany

t (say 20C to

ands.

afer to

ntor

d part.

ck.

me worry

so that

se are a

M4040's

the

about

=A0TI

y much

od ones

s die

ting a C

ld.

he

"The Journey is the reward"

eff.com- Hide quoted text -

Ouch, that's spendy. 'course all Apex stuff is. I was thinking the lm399 might be over-kill in the circuit, It's from a diode laser current driver by K. Leibricht and J Hall. RSI 1993? (I've used some RSI circuits.)

George H.

Yes, and that is exactly how "TC zeners" were made. E.g. 1N821 .. 1N829.

(say 20C

ds.

But you've got to control the current through it rather exactly to actually get that temperature coefficient - using the 1N829 was expensive. Not only did you pay through the nose for the part itself, but you needed 15ppm 0.1% resistors to control the current through it, and a good op amp to make sure that it stayed right even if the supply voltage moved around a bit.

-- Bill Sloman, Nijmegen