Ultra Precision Reference LTZ1000

[TiN]

Based on eBay listing photos, sadly 15K/1K is used with CH chip.

[Cerebus]

I think the problem may be that your price is a little bit too high. You can buy original 3458A A9 ref boards for less quite often on ebay and they use the LTZ1000ACH.

Also on a $350 ticket item a description that includes a phrase like "The chip has been tested and appears to be working correctly in circuit, but has not been burned in for even better stability." doesn't exactly inspire confidence. Someone who's going to fork out $350 wants a guarantee that the board is working and meets specifications, which for an item like the A9 board includes a drift test over at least several months. That is the value add and that is what justifies a price that is significantly above the BOM cost.

[doktor pyta]

From ebay item's description:

'LTZ1000 heat compensated chip standard (which often goes bad)'

prove it

[Kleinstein]

I would not expect a lot of failing LTZ ref modules in the 3458. They are supposed to be well tested and burned in. So the bad ones with excessive drift or significant popcorn noise (which might often come together) should be sorted out. There still seem to be rare cases with HP refs. to show popcorn noise, but I am afraid the chances are higher with the untested replacement boards.

[martinr33]

I think that there would be some interest in the bare board from this group. That might be a way to recover your investment. You might also find some buyers for the completed board, at a price that lets you recover the parts cost.

Thanks for doing this, I know that it isn't meeting your expectations but it is nice to have this piece of engineering done. 

[VK5RC]

If I recall correctly - somewhere in this post - somebody gave mentioned a good quality socket that could be used to temporarily fit an LTZ - to try and sort some 'good ones' without a full board each.
Does any one recall the detail of the socket/source etc or did I imagine it? 

[SvanGool]

This one maybe? : https://www.eevblog.com/forum/metrology/px-reference/msg1381836/#msg1381836 by TiN

[VK5RC]

Thanks, my Google fu is improving, I think it is this one.   https://www.digikey.com.au/product-detail/en/te-connectivity-amp-connectors/8058-1G32/1-1437508-6-ND/2259044 

Thanks to BU508A in this post.  https://www.eevblog.com/forum/metrology/kx-reference/msg1274394/#msg1274394. reply#136

[branadic]

Since I received a few vintage LTZ1000 samples from lowimpedance and need some boards for comparison I decided to use the design of Dr. Frank. I modified his design to fit into a Hammond 1455C1201 case and added LT1763 (SO8) voltage regulator, giving 12V. This is how it currently looks like. The 3 blocks to the right indicate the place for Pomona 3770 binding posts.

-branadic-

[Haasje93]

It looks really nice!

[Dr. Frank]

Hello,
the loose power supply GND cable must NOT be connected to the negative reference star point, but instead to the jack farther right.
I'd recommend to separate the power supply thermally and shileding-wise.

I see no basic deviation to my LTZ design, so far, so the rest is ok.

Frank

[Andreas]

+1 for the thermal isolation between LT1763 and LTZ.
(thats why I will do a re-design of my B-revision)
In your case it might be less urgent if you have a stabilized external supply instead of battery.

with best regards

Andreas

[branadic]

Thanks for the feedback. Changed the loose wire to JP2, added a thermal slot between regulator and reference ciruit and changed two MKS 100nF to MKP2 100nF. Attached the almost final state of the board.

-branadic-

[branadic]

I will order some boards within the next days. Someone out there who wants a few boards of this layout as well?

-branadic-

[BradC]

I will order some boards within the next days. Someone out there who wants a few boards of this layout as well?

-branadic-

I'll put my hand up for one please.

[GregDunn]

Are you selling boards only, or boards with the LTZ1000?

[branadic]

Boards only...
For orders please write me a short pm with how many boards you need. I will close this offering by the end of next week and order the boards in the necessary quantities.

-branadic-

[branadic]

Wow, didn't expect such a response. I think I will order at Elecrow and pcb cost will be in the order of $2 per board or less depending on the final quantity plus shipping cost (Airmail) plus shipping to you guys (simple envelope to keep cost down). Thus I changed the outer form to a rectangle of size 120 x 50mm². PCB color will be black with immersion gold, single layer board.
Attached is the final schematic.
Please send me an official email adress with the PN too. Thanks.

-branadic-

[Kleinstein]

The pot for gain adjustment in the final stage is a little odd. Normally one avoids using such low ohms pots, as the wiper contact may not be that reliable.  The more normal way is to have something like 10-100 Ohms in parallel to a 10 K pot with 5-100 K in series.

[branadic]

Since this board is working for Frank and since I had the chance to measure one of his reference boards I wouldn't complain

-branadic-

[seec]

is it a fake?
https://ru.aliexpress.com/store/product/LTZ1000ACH/1200151_32372123768.html

[Andreas]

Does not look genuine to me

- LT-logo is missing
- printing is not correctly aligned to Pin 8 marker.

So in best case it is refurbished and re-marked.
see also:

http://volt-nuts.febo.narkive.com/FIfiYHU5/ltz1000ach-and-ltz1000ch-in-china

with best regards

Andreas

[try]

Hello Kleinstein,

The pot for gain adjustment in the final stage is a little odd. Normally one avoids using such low ohms pots, as the wiper contact may not be that reliable.  The more normal way is to have something like 10-100 Ohms in parallel to a 10 K pot with 5-100 K in series.

as far as I know every wiper contact can fail. Is there a particular issue with low ohm pots?

Obviously your proposal does not fit Frank's trimming scheme where he is basically trimming in series needing an option for a low ohm variation at the end of the trimming chain.
Could you propose a complete trimming scheme for the divider regardless of the layout available on the PCB?

[branadic]

One week left to sign up for the board. Currently there are 45 boards on the order list. Currently there are 130 boards on the order list.

6x branadic
3x SvanGool
3x BradC
4x eplpwr
3x Sprock
2x FredTW
2x Haasje93
1x retroware
2x dhillman
5x try
2x flittle
10x bck
5x F64098
2x RandallMcRee
10x MiDi
5x e61_phil
2x Le_Bassiste
5x quarks
2x Macbeth (3)
5x meandeev
5x martinr33 (10/20)
3x Hermann W
3x mikron
2x Harfner
3x pansku
2x Carrington
2x mweymarn
3x Vbemultiplier
6x vmiceli
5x ap
3x sixtimesseven
2x AlfBaz
5x FriedLogic
2x guenthert (need email address)
2x Vgkid
3x borghese


-branadic-

[Echo88]

My attempt:
LTZ1000-PCB Designed by Dr Frank and
LT1763-Low-Noise-Supply designed by Cellularmitosis



Both PCBs are made available as OSHW by Cellularmitosis (big thanks  ) and can be bought here or via any other pcb-manufacturer with the gerber-files:

https://oshpark.com/shared_projects/dGYcQAgn LTZ-Board
https://oshpark.com/shared_projects/uVr4yDgz LT1763-Board

Why i did it that way:

LTZ1000 is directly buffered via LTC1052 since i dont use the 7->10V-function for the references at the moment,
LTZ1000 protected from airflow by green lemonjuice-cap,
an EMI-Filter BNX002-01 is used at the power supply input https://www.murata.com/en-eu/api/pdfdownloadapi?cate=luNoiseSupprFilteBlockType&partno=BNX002-01,
a double sided copper-clad-board at the pomona-jacks should equalize thermal differences pretty well,
the LT1763 is located further from the LTZ1000-PCB, so it doesnt affect it by temperature variation due to varying input voltage,
TEKO 393-case is afterwards stuffed again with paper towel shreds to reduce airflow,
Resistors by Edwin Pettis