Author Topic: Ultra Low Noise Reference 2DW232, 2DW233, 2DW23x  (Read 234926 times)

0 Members and 1 Guest are viewing this topic.

Offline quarks

  • Frequent Contributor
  • **
  • Posts: 874
  • Country: de
Re: Ultra Low Noise Reference 2DW232, 2DW233, 2DW23x
« Reply #100 on: October 28, 2016, 02:11:13 pm »
Hello branadic,

will these caps fit to the HP3458A A9 LTZ1000 Reference Board?

thanks
quarks
 

Offline babysitter

  • Frequent Contributor
  • **
  • Posts: 893
  • Country: de
  • pushing silicon at work
Re: Ultra Low Noise Reference 2DW232, 2DW233, 2DW23x
« Reply #101 on: October 28, 2016, 03:19:35 pm »
@Branadic: If you tell me the hole distance and the location of the nose relative to the holes, I can ask this guy to apply it to his drawing. But i have defined relatively thick walls as I can have some air filled closed holes in the wall then.

We have only a prusa I3 filament printer at the club, dont ask for details, I dont know it - yet.

br
hendrik
I'm not a feature, I'm a bug! ARC DG3HDA
 

Offline branadic

  • Super Contributor
  • ***
  • Posts: 2377
  • Country: de
  • Sounds like noise
Re: Ultra Low Noise Reference 2DW232, 2DW233, 2DW23x
« Reply #102 on: October 28, 2016, 04:45:51 pm »
I guess it was TiN how published a copy of a LTZ1000 cap, a while ago. Maybe someone wants to check wether this one matches the original used in 3458A?
Otherwise we can use any other design, as long as everyone is fine with the it.
The good think about the printed material is, that you can also cut a thread into the material.
Computers exist to solve problems that we wouldn't have without them. AI exists to answer questions, we wouldn't ask without it.
 

Offline Nuno_pt

  • Frequent Contributor
  • **
  • Posts: 435
  • Country: pt
Re: Ultra Low Noise Reference 2DW232, 2DW233, 2DW23x
« Reply #103 on: October 28, 2016, 04:50:50 pm »
Maybe we can arrange a cap that fits the LTZ/LM399/2DW23x, they should be pretty close in terms of dimensions.
Nuno
CT2IRY
 

Offline pelule

  • Frequent Contributor
  • **
  • Posts: 513
  • Country: de
  • What is business? It’s other people’s money
Re: Ultra Low Noise Reference 2DW232, 2DW233, 2DW23x
« Reply #104 on: October 28, 2016, 06:02:46 pm »
I would add myself to the list of buyers of that cap-print-group, if that fit at least to LTZ1000 and 2DW2xx (thus also to LMx99 - not the H package).
You will learn something new every single day
 

Offline babysitter

  • Frequent Contributor
  • **
  • Posts: 893
  • Country: de
  • pushing silicon at work
Re: Ultra Low Noise Reference 2DW232, 2DW233, 2DW23x
« Reply #105 on: October 28, 2016, 06:21:52 pm »
Hey, great response! I guess we can make some of them at the club. But first I'd like to check the measures, try the 2DW when they arrive, my LTZ in the finished reference source is already under a (vacuum port) cap. Distribution is easy with STL files, everyone who wants one only has to find someone to print.



I'm not a feature, I'm a bug! ARC DG3HDA
 

Offline pelule

  • Frequent Contributor
  • **
  • Posts: 513
  • Country: de
  • What is business? It’s other people’s money
Re: Ultra Low Noise Reference 2DW232, 2DW233, 2DW23x
« Reply #106 on: October 28, 2016, 07:37:19 pm »
Hi Flinstone,
use that link below to add yourself to the list (but read hints/instruction first, you'll find, if you scroll up on that page a bit )
See link:
https://www.eevblog.com/forum/metrology/group-buy-of-2dw233-ultra-low-noise-zener-reference/msg1057877/#msg1057877
You will learn something new every single day
 

Offline branadic

  • Super Contributor
  • ***
  • Posts: 2377
  • Country: de
  • Sounds like noise
Re: Ultra Low Noise Reference 2DW232, 2DW233, 2DW23x
« Reply #107 on: October 31, 2016, 06:38:25 pm »
Quote
@Branadic: If you tell me the hole distance and the location of the nose relative to the holes, I can ask this guy to apply it to his drawing. But i have defined relatively thick walls as I can have some air filled closed holes in the wall then.

Could find the CAD files from TiN:

https://xdevs.com/doc/HP_Agilent_Keysight/3458A/cad/a9_cap_top.STEP
https://xdevs.com/doc/HP_Agilent_Keysight/3458A/cad/a9_cap_bot.STEP

There you can find the hole distance.
Computers exist to solve problems that we wouldn't have without them. AI exists to answer questions, we wouldn't ask without it.
 

Offline Kleinstein

  • Super Contributor
  • ***
  • Posts: 14016
  • Country: de
Re: Ultra Low Noise Reference 2DW232, 2DW233, 2DW23x
« Reply #108 on: October 31, 2016, 07:08:17 pm »
The cap will be only a minor detail, and usually different from the one for an LTZ1000, as the zener has no internal heater. So the zener will consume less power and good insulation may not be a good thing.
The more important point might be choosing the circuit. The simple version is something like the one used with a LM399 or LM329: scale to a higher voltage (e.g. 10 V) and than use a resistor to set the current.
 

Offline zlymexTopic starter

  • Frequent Contributor
  • **
  • Posts: 556
  • Country: cn
Re: Ultra Low Noise Reference 2DW232, 2DW233, 2DW23x
« Reply #109 on: November 01, 2016, 03:11:30 am »
The cap will be only a minor detail, and usually different from the one for an LTZ1000, as the zener has no internal heater. So the zener will consume less power and good insulation may not be a good thing.
The more important point might be choosing the circuit. The simple version is something like the one used with a LM399 or LM329: scale to a higher voltage (e.g. 10 V) and than use a resistor to set the current.

And because if it, a more 'useful' zener current of 8mA or >12mA can be applied to achieve even lower noise per device than LTZ1000 for which the data sheet only recommend 4mA.

The cap or tissue are useful only to stop the air flow around the device/legs, not for thermal insulation. However, these two are often closely related. One good example is the one used by Datron/Wavetek in their 49xx voltage reference where they use black form plus a metal outer case.

As for the step-up to 10V, yes, there is this circuit of "scale to a higher voltage (e.g. 10 V) and than use a resistor to set the current." on the first page of the thread.
 

Offline Kleinstein

  • Super Contributor
  • ***
  • Posts: 14016
  • Country: de
Re: Ultra Low Noise Reference 2DW232, 2DW233, 2DW23x
« Reply #110 on: November 01, 2016, 07:01:59 pm »
To take really advantage of the low noise of this reference type, one might want to use a lower noise than the AD707 - even though the AD707 is already quite good. As input bias / current noise is not that critical, something like an OP27 / LT1007 should be OK.

 

Offline branadic

  • Super Contributor
  • ***
  • Posts: 2377
  • Country: de
  • Sounds like noise
Re: Ultra Low Noise Reference 2DW232, 2DW233, 2DW23x
« Reply #111 on: November 01, 2016, 07:14:59 pm »
It would be good to have the second diode directly in use, maybe for some servo temperature controller.
Computers exist to solve problems that we wouldn't have without them. AI exists to answer questions, we wouldn't ask without it.
 

Offline bertik

  • Contributor
  • Posts: 30
  • Country: ch
    • W's Laser Web Site
Re: Ultra Low Noise Reference 2DW232, 2DW233, 2DW23x
« Reply #112 on: November 06, 2016, 09:12:42 am »
What about using the second zener to reduce noise instead?  Feeding it to a separate opamp and then combining reduces the noise by a factor of 1/sqrt(2). While not much this would give an easy improvement for the price of an extra opamp.

Using it just for temp sensing seems a sacrilege.. that could be easily done by gluing a thermistor on top of it.

-BK
 

Offline Kleinstein

  • Super Contributor
  • ***
  • Posts: 14016
  • Country: de
Re: Ultra Low Noise Reference 2DW232, 2DW233, 2DW23x
« Reply #113 on: November 06, 2016, 10:22:04 am »
Even with just one diode used, the noise is already really low - e.g. better than an OP07, OP177 or AD707 which are typical candidates for scaling the voltage. The current to get a low TC is already quite high (e.g 5 mA range) - more current is more of a problem it you don't want the chip to heat up too much.

Having a temperature sensor directly on the chip can really help to stabilize or just measure the temperature.  So I would consider this the more practical use for the second diode.
 
The following users thanked this post: cellularmitosis

Offline Tazz

  • Contributor
  • Posts: 25
  • Country: fr
Re: Ultra Low Noise Reference 2DW232, 2DW233, 2DW23x
« Reply #114 on: November 06, 2016, 10:31:06 am »
Kleinstein, I second you for the use of the second diode
 

Online Andreas

  • Super Contributor
  • ***
  • Posts: 3207
  • Country: de
Re: Ultra Low Noise Reference 2DW232, 2DW233, 2DW23x
« Reply #115 on: November 06, 2016, 11:47:02 am »
Zlymex:

Isnt the 2nd diode already used forward biased as temperature compensation diode?
I cannot get this clear from your description.
In this case you do not need to think about further noise reduction.

Or are there 2 temperature compensated diodes (2 zeners + 2 diodes) in one package?
In this case you cannot read the temperature. (because the zener voltage is in series to the compensation diode).

with best regards

Andreas
 

Offline plesa

  • Frequent Contributor
  • **
  • Posts: 965
  • Country: se
Re: Ultra Low Noise Reference 2DW232, 2DW233, 2DW23x
« Reply #116 on: November 06, 2016, 12:33:37 pm »
Zlymex:

Isnt the 2nd diode already used forward biased as temperature compensation diode?
I cannot get this clear from your description.
In this case you do not need to think about further noise reduction.

Or are there 2 temperature compensated diodes (2 zeners + 2 diodes) in one package?
In this case you cannot read the temperature. (because the zener voltage is in series to the compensation diode).

with best regards

Andreas

There are only two zeners, back to back and common anode connected to package.
So there is first order temperature compensation - zener avalanche  diode.
On first piece I will have I'm going to remove epoxy and check it under microscope.
https://www.eevblog.com/forum/metrology/ultra-low-noise-reference-2dw232-2dw233-2dw23x/
It is the same like used in zeners like 1N82x.
« Last Edit: November 06, 2016, 12:35:58 pm by plesa »
 

Online Andreas

  • Super Contributor
  • ***
  • Posts: 3207
  • Country: de
Re: Ultra Low Noise Reference 2DW232, 2DW233, 2DW23x
« Reply #117 on: November 06, 2016, 12:52:16 pm »
The 1N82x has a normal diode in series to the zener.

with best regards

Andreas
 

Offline plesa

  • Frequent Contributor
  • **
  • Posts: 965
  • Country: se
Re: Ultra Low Noise Reference 2DW232, 2DW233, 2DW23x
« Reply #118 on: November 06, 2016, 12:59:06 pm »
The 1N82x has a normal diode in series to the zener.

with best regards

Andreas

OK, and do you thin that doping level will make noticeable difference in tempco of forward biased diode?
 

Offline Gyro

  • Super Contributor
  • ***
  • Posts: 9392
  • Country: gb
Re: Ultra Low Noise Reference 2DW232, 2DW233, 2DW23x
« Reply #119 on: November 06, 2016, 01:25:40 pm »
Has anyone managed to find a more comprehensive datasheet for these parts? I successfully downloaded the one that zlymex linked in his first post (the 2 page Shen Xin one) but didn't have any success downloading the ones he linked later.

I've attached one pdf datasheet that I found (SJ brand), it appears to include lots of environmental test requirents but is mostly incomprehensible to me. I guess one from the 17th Radio Factory itself would be the holy grail. It, or a translation, might help answer some of the questions that are coming up.

Chris

(Edit: typo)
« Last Edit: November 06, 2016, 01:46:50 pm by Gyro »
Best Regards, Chris

"Victor Meldrew, the Crimson Avenger!"
 

Offline Kleinstein

  • Super Contributor
  • ***
  • Posts: 14016
  • Country: de
Re: Ultra Low Noise Reference 2DW232, 2DW233, 2DW23x
« Reply #120 on: November 06, 2016, 01:49:57 pm »
It is a good question whether the two zeners are just normal zener diodes (like 6.2 V reference diodes) or they are supposed to used as a pair in series as a 7 V reference more like the 1N829 or similar.

There are both ways to get a low TC reference diode at the right current: either just a suitable zener diode at about 5-6 V or a compensated version with a zener and normal diode in series like in the 1N829, usually in the 7 V range or higher.

As the voltage is only 6.2 V, I would expect a relatively normal zener. We don't really know much about the internal details - at least they seem to be much lower noise than normal zener diodes. So there is like a more or less secret detail that makes these diodes so special.

Normally the TC of a silicon diode in forward direction is not depending on details of the diode: usually one has a linear temperature dependence that extrapolates to 1.2 V at 0 K. Higher current gives a higher voltage and thus lower TC. Thus a diode with 600 mV forward drop at 300 K will have a -2 mV/K slope.

As the diodes are connected common anode, there is no way to use them in series as a 12.4 V ref. One might be able to use the second diode as a backup to check for drift - but this would likely be better with a completely separate one.

With no detailed DS one might want to do a few measurements, e.g. voltage as a function of current an temperature to get TC and differential resistance. To really get the best working conditions one might need these data for the individual units anyway.
 
The following users thanked this post: elecdonia

Online Andreas

  • Super Contributor
  • ***
  • Posts: 3207
  • Country: de
Re: Ultra Low Noise Reference 2DW232, 2DW233, 2DW23x
« Reply #121 on: November 06, 2016, 02:26:34 pm »
It is a good question whether the two zeners are just normal zener diodes (like 6.2 V reference diodes) or they are supposed to used as a pair in series as a 7 V reference more like the 1N829 or similar.

There are both ways to get a low TC reference diode at the right current: either just a suitable zener diode at about 5-6 V or a compensated version with a zener and normal diode in series like in the 1N829, usually in the 7 V range or higher.

As the voltage is only 6.2 V, I would expect a relatively normal zener. We don't really know much about the internal details - at least they seem to be much lower noise than normal zener diodes. So there is like a more or less secret detail that makes these diodes so special.

Hello,

I think you have never had a 1N829 in hand.
The 1N829 has around 6.4V. (one of the usual voltages of temperature compensated zeners).
It is a normal diode 0.7V (-2mV/K) and a Zener (around 5.7 so +2mV/K) in series.
If you measure the 1N829 zener in "forward direction" you will find the breakdown voltage of the normal diode.
(which is beyond 15 V).

with best regards

Andreas
 

Offline Gyro

  • Super Contributor
  • ***
  • Posts: 9392
  • Country: gb
Re: Ultra Low Noise Reference 2DW232, 2DW233, 2DW23x
« Reply #122 on: November 06, 2016, 03:47:26 pm »
I happened to measure a few 1N827 to 829 a few days ago, they were in the 6.2-6.3V range. The highest was a very old ex Solartron 1N829A which came out at 6.37V iirc, all at 7.5mA. It's only LM399, LTZ etc that are in the 6.95V range.

If that is in the same range as the 2DW233 then it is either a zener+series diode or a higher voltage zener with unsusual TC characteristics.  If it is a zener + diode combo then that will exclude using the second one as a sense diode... 1N82x don't conduct backwards (forwards in the conventional sense), the series diode is reverse biased.  :-\

This is the sort of thing a more comprehensive (and comprehensible?) datasheet would tell us.
« Last Edit: November 06, 2016, 03:57:40 pm by Gyro »
Best Regards, Chris

"Victor Meldrew, the Crimson Avenger!"
 

Online Andreas

  • Super Contributor
  • ***
  • Posts: 3207
  • Country: de
Re: Ultra Low Noise Reference 2DW232, 2DW233, 2DW23x
« Reply #123 on: November 06, 2016, 04:01:42 pm »
Hello,

in the 1N82x series there are also some back to back zeners available (1N824) (but without center tap)
in this case both directions will have the 5.9-6.5 V zener voltage.
http://www.microsemi.com/document-portal/doc_download/10940-sa6-3-pdf

The trick of having a forward biased diode and a zener diode is that there will be a "zero TC" current
where the diode and zener TC are compensated to near zero over a small temperature range.

with best regards

Andreas


 

Online Andreas

  • Super Contributor
  • ***
  • Posts: 3207
  • Country: de
Re: Ultra Low Noise Reference 2DW232, 2DW233, 2DW23x
« Reply #124 on: November 06, 2016, 06:22:53 pm »
Hello,

my measurements on several 1N829A (which are only specified for lesser T.C. (at 7.5mA) than 1N823) range from 2uVpp to 22uVpp 1/f noise (0.1 .. 10Hz).
This depends much on manufacturer.
I do not know if the quality nowadays is still the same...

The lesser noise (compared to higher voltage zeners in the link) might result from lower dynamic resistance.
Which is below 10 Ohms for around 6V compared to around 100 Ohms or more for higher voltage zeners.

With best regards

Andreas

 
The following users thanked this post: evb149


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf