Lowest noise Zener Diodes?

[Joshua_G]

Which are the lowest noise Zener Diodes, either SMD or through hole, which are in current production, regardless of price?

 

[homemodder]

Temp compensated ones like lm329. There are better ones but although you say regardless of price I doubt you want to fork out 60 to 70 dollars for a single part, these are used in space exploration equipment and satelites. Read linear technology application AN82F, very good info on there. You could also make yourself one, just combine a regular zener with a diode.

 

[MarcelvdG]

I don't know, but I do know that you should use a voltage well below 5 V if you want low noise, like 3.9 V or less. Higher voltage "zeners" actually have avalanche rather than Zener breakthrough, which causes much more noise.

 

[Joshua_G]

Temp compensated ones like lm329. There are better ones but although you say regardless of price I doubt you want to fork out 60 to 70 dollars for a single part, these are used in space exploration equipment and satelites. Read linear technology application AN82F, very good info on there. You could also make yourself one, just combine a regular zener with a diode.

Thanks.
Indeed, $60-70 a piece I cannot afford (since I need quite a few of those).
How do I combine a regular diode with a Zener to lower its' noise?

 

[Wavebourn]

Theoretically the lowest noise Zeners are for about 8.2V

Also, here is some list I fond in the Net:

CMHZ4099: 6.8 V, Low Noise Zener Diode
CMHZ4100: 7.5 V, Low Noise Zener Diode
CMHZ4101: 8.2 V, Low Noise Zener Diode
CMHZ4102: 8.7 V, Low Noise Zener Diode
CMHZ4103: 9.1 V, Low Noise Zener Diode
CMHZ4104: 10 V, Low Noise Zener Diode
CMHZ4105: 11 V, Low Noise Zener Diode
CMHZ4106: 12 V, Low Noise Zener Diode
CMHZ4107: 13 V, Low Noise Zener Diode
CMHZ4108: 14 V, Low Noise Zener Diode
CMHZ4109: 15 V, Low Noise Zener Diode
CMHZ4110: 16 V, Low Noise Zener Diode
CMHZ4111: 17 V, Low Noise Zener Diode
CMHZ4112: 18 V, Low Noise Zener Diode
CMHZ4113: 19 V, Low Noise Zener Diode
CMHZ4114: 20 V, Low Noise Zener Diode
CMHZ4115: 22 V, Low Noise Zener Diode
CMHZ4116: 24 V, Low Noise Zener Diode
CMHZ4117: 25 V, Low Noise Zener Diode
CMHZ4118: 27 V, Low Noise Zener Diode
CMHZ4119: 28 V, Low Noise Zener Diode
CMHZ4120: 30 V, Low Noise Zener Diode
CMHZ4121: 33 V, Low Noise Zener Diode
CMHZ4122: 36 V, Low Noise Zener Diode
CMHZ4123: 39 V, Low Noise Zener Diode
CMHZ4124: 43 V, Low Noise Zener Diode
CMHZ4125: 47 V, Low Noise Zener Diode
CMHZ4614: 1.8 V, Low Noise Zener Diode
CMHZ4615: 2 V, Low Noise Zener Diode
CMHZ4616: 2.2 V, Low Noise Zener Diode
CMHZ4617: 2.4 V, Low Noise Zener Diode
CMHZ4618: 2.7 V, Low Noise Zener Diode
CMHZ4619: 3 V, Low Noise Zener Diode
CMHZ4620: 3.3 V, Low Noise Zener Diode
CMHZ4621: 3.6 V, Low Noise Zener Diode
CMHZ4622: 3.9 V, Low Noise Zener Diode
CMHZ4623: 4.3 V, Low Noise Zener Diode
CMHZ4624: 4.7 V, Low Noise Zener Diode
CMHZ4625: 5.1 V, Low Noise Zener Diode
CMHZ4626: 5.6 V, Low Noise Zener Diode
CMHZ4627: 6.2 V, Low Noise Zener Diode

And here are Soviet diode-compensated Zeners for 9V, $3/dozen:

D818G USSR Zener Diodes Lot of 12 | eBay

 

[SY]

How do I combine a regular diode with a Zener to lower its' noise?

You can't.

Best Zeners are 6.2V, which sits on the cusp of Zener and avalanche. Forward biased red LEDs are even quieter.

 

[Joshua_G]

Thanks, all.

 

[sreten]

Hi,

Parallelling cheaper devices is the obvious answer if noise is really an issue.

Theoretically 2 devices drop noise by 3dB, 4 by 6dB, 8 by 9dB etc, but
practically its never that simple and the point of no real benefit reached
for extra devices are 2 devices to about 6 devices usually.

(Might be a bigger number with devices not designed for low noise, but
that is pointless, you want cheap low noise devices in the first place.)

rgds, sreten.

 

[rsavas]

Would a TL431 be lower noise than a zener. I have read that the ones from ONSEMI are the lowest noise of the various mfgs.

 

[AndrewT]

I thought 431 were a lot noisier.

 

[AndrewT]

.............You could also make yourself one, just combine a regular zener with a diode.

how? series or parallel or ?

 

[MarcelvdG]

You can't.

Best Zeners are 6.2V, which sits on the cusp of Zener and avalanche. Forward biased red LEDs are even quieter.

Aren't you mixing up low noise with low temperature coefficient? Zener and avalanche breakdown have opposite temperature coefficients, so voltages of 5 V to 6 V give the lowest temperature coefficient.

 

[martin clark]

I think it's much less important to worry about which part, providing you ensure you pour enough current through it to drive-down the dynamic impedance. For example - a few years ago Christer posted a long series of detailed measurements on diodes, zeners and LED noise measurements, here:

http://www.diyaudio.com/forums/parts/35821-some-noise-measurements-leds-zener-diodes.html

NB from experience the LM329 isn't just a diode - it is an active part to give the resulting low effective impedance at LF. Be careful on how you use them as a result.

 

[canvas]

I am looking for low noise reference, too. The noise of TL431 varies through different manufacturers. Onsemi TL431 has low noise, but the lowest is from Sipex TL431 which has 52nv/Hz@10Hz to 38nv/Hz@100kHz noise figure. If you look through various voltage reference, this is very low noise. I use Sipex AS1431 to replace LM336-2.5 Vref on DAC's regulator (OP based). The result is very good, I hear more details.

 

[davidsrsb]

You can't.

Best Zeners are 6.2V, which sits on the cusp of Zener and avalanche. Forward biased red LEDs are even quieter.

Incorrect, the values where there is a balance of avalanche and zener processes are the noisiest. Avoid from 5V6 to about 8V2 (depends on current a bit). For values below 5V6, forward biased diodes are probably a better choice.
(but don't use blue LEDs)

 

[John Luckins]

I've just switched from an LM329 as a voltage reference on four super regs to a string of four red LED's. There's a quite definite subjective improvement, probably as a result of the lowering of the noise which the specs referred to in Martin's post above and the LM329 datasheet confirm. Sadly I don't have a way to measure the noise. The low frequency dynamic impedance of the four LED's is higher though there are ways to make that less significant in this particular application (pre-filtering the Vref feed and driving the LED's with a cascaded current source).

John

 

[AndrewT]

A cap across the LEDs will lower the AC impedance. And attenuate some of the noise from the quieter LEDs.

 

[SY]

Incorrect, the values where there is a balance of avalanche and zener processes are the noisiest. Avoid from 5V6 to about 8V2 (depends on current a bit). For values below 5V6, forward biased diodes are probably a better choice.
(but don't use blue LEDs)

Can you explain datasheets like this?

http://documentation.renesas.com/doc/products/diode/rej27g0012_zener.pdf

Also clear statements in standard texts like Pease and Horowitz/Hill?

 

[godfrey]

Best Zeners are 6.2V

Can you explain datasheets like this?

http://documentation.renesas.com/doc/products/diode/rej27g0012_zener.pdf

According to that datasheet, 6.2V isn't optimal for noise or dynamic resistance. What am I missing?

 

[SY]

What am I missing?

The data in 1.4.2. Look particularly at the first graph, which shows 1-2 orders of magnitude improvement at 5.6 and 6.2V compared to, say, 9 or 20V.

LEDs are still quieter, though- under 0.5nV/rt Hz for a cheap red.