Ultra Precision Reference LTZ1000

[Dr. Frank]

I have saved a copy before Fluke removed the schematics. June 1996, 39MB.
It's on the site of a voltnuts member Didier Juges, ko4bb... Or so, i just do not have access to the link right now.
There you have to work yourself through the drawings carefully.
Frank

[quantumvolt]

Thanks. I will look for it at the kobb-site.

[quarks]

http://www.ko4bb.com/Manuals/Fluke/Fluke_5700A_5720A_Multi-Function_Calibrator_Service_Manual.June_1996.pdf

[quarks]

The 5720 is just to be replaced by the brand new 5730.
https://www.eevblog.com/forum/testgear/new-fluke-5730a-calibrator/msg296208/#msg296208
Both are much too expensive if you do not earn money with them.
Therefore my favorite is the unfortunately discontinued Wavetek 4808, which playes in the same league.

[quarks]

To come back to the OT. I would like to modify a HP3458A reference board to HFL spec (if possible).

Does anyone have details about the Fluke HFL mod?

So far I found at febo/voltnuts that all the boards are suppose to be the same and only aged and selected. I also know from Dr. Frank about his mod and will do the same as a first step.

Because the HFL has not only better DCV but also better resistor accuracy, there must be more changes than only the lower heater temp.

Any hints are welcome.

[quarks]

Thanks Diligentminds,
that is a good confirmation and summary of all things I already knew and have discussed with Dr. Frank.
But I still wonder, if it is really not more to it, why HP/Agilent never offered or does not sell it anymore. Also they never have reached the resistance specs of the Fluke HFL. Besides the document you attached (which I already had), do you know of anything more documented? The only I know of is the att. selection guide.

About calibration, I have actually bought a 3458 mainly because of the artefact calibration, which I already have done and I like it very much. With my lab gear, also I still do not have a real 10V Reference like 732B (or similar 7000/4910), I can repeat the cal any time I like. So far my experiment LTZ1000 board still is the one from Babysitter (7.xxx Volt Output). 

Bye
quarks

[captbilly]

You can read more about it here
http://home.51.com/jj3055/diary/item/10053954.html

I am fairly sure the slots are there to keep the pcb from transferring mechanical stress to the LTZ1000.  Linear Technology discusses this problem in an app note, but essentially the issue is that movements of the pcb (from temperature, humidity, or even from the user applying force on the board) transfer stress to the reference IC package, as well as the die itself.  The stress effects the output voltage and, according to Linear Technology, is part of the reason that references drift over time (the stresses tend to slowly decrease over time as the board and package slightly flow).  Linear Technology recommend slotting around the reference so that movements on the rest of the board do not transfer to the reference itself.

Leakage currents could be a problem if there are high impedance paths in the circuit, but that is uncommon in voltage reference circuits.

Schematics here
http://bbs.38hot.net/read-htm-tid-36472.html

And finally here how to convert 7V to 10V
http://www.crystalradio.cn/thread-229749-1-1.html

Looking at the pcb I can't stop wondering why these slots are milled. The voltage isn't that high that you need the extra creepage distance, and right now it seems to me it's only weakening the pcb's structural integrity. Worst case it introduces unknown offsets as the pcb will bend a bit either when moved, or with temperature changes. Am I overlooking something here?

[branadic]

Board stress has no effect on IC dies packaged in hermetic TO packages-- it only applies to devices in plastic packages, and surface mount packages are more sensitive to this effect than through-hole [plastic DIP] packages are.

AFAIK, the slotted PC board design for the LTZ1000 came from the original Datron 4910 voltage reference design, and because that unit could run off of batteries, in order to reduce the battery energy depletion rate they needed to minimize heat-loss from the LTZ1000-- the slots in the PC board (as well as added insulation) help with this.  The slots are there for NO OTHER REASON.  The voltage reference design on the Chinese "volt-nut" BBS simply copies this technique without knowing why it was done in the Datron design.  If you're not going to run your device on batteries, then you don't need slots or insulation.

You guys know about the investigation that was done on the LM399 with the question what influence these slots have?

http://dg3hda.primeintrag.org/doku.php?id=lm399_thermographie

[Dr. Frank]

AFAIK, the slotted PC board design for the LTZ1000 came from the original Datron 4910 voltage reference design, and because that unit could run off of batteries, in order to reduce the battery energy depletion rate they needed to minimize heat-loss from the LTZ1000-- the slots in the PC board (as well as added insulation) help with this.  The slots are there for NO OTHER REASON.  The voltage reference design on the Chinese "volt-nut" BBS simply copies this technique without knowing why it was done in the Datron design.  If you're not going to run your device on batteries, then you don't need slots or insulation.

Note that the LTZ1000 (and ideally the support circuitry too) does need some kind of cover to prevent air currents from causing thermal EMF's generated by thermocouples created by the Kovar-leads of the TO package to the copper PC board traces.

Hi,

that's the first reasonable explanation I've read (thermal insulation for saving of battery energy).
If the LTZ1000 would run on 45°C, the circuitry would not need so much energy as in the Datron 4910, running on 60°C.

 I totally agree, that the Asians simply have copied, copied, copied and copied those stupid slots, without thinking, why they are there..
And everybody else copied from the Asians.. the slots and the "A" Version, which is copied from the 3458A and which makes sense on that 95°C temperature. (Violating the LTZ1000 as being an LM399)

Although the temperature distribution on the PCB is influenced, observable only at high stabilization temperatures as with the LM399, up to now I don't see any influence on the stability of the reference circuitry, and that'much moremost important, so those slots can be omitted easily.

Frank

[Dr. Frank]

You guys know about the investigation that was done on the LM399 with the question what influence these slots have?

Know that, sure.
But it has not been demonstrated any influence on the stability of ref amps, especially not on the LT1000 @ 45°C.

No influence from mechanical deformations on the Output voltage, as you yourself have found out.

So, really nice IR Pictures and experiments, proving that there is no practical use for the slots.

Frank

[codeboy2k]

So, really nice IR Pictures and experiments, proving that there is no practical use for the slots.

I read the linked article with the IR pictures , and I thought the conclusion was that the slots actually help to maintain the heat at the junction of the can's leads with the copper foil, thus minimizing any delta K and thus minimizing the thermal junctions formed there (copper/kovar)

This is, in fact, still consistent with why it was done in the Datron 4910 battery based design -- to minimize heat loss giving longer battery life.

However, in non-battery applications there might be some useful reason still. 

I'm not a voltnut, and thus not an advocate for slots or no slots, just commenting on the blog I read with the IR photos.

[babysitter]

What do you guys think about my cheap connection ?

Its simple single core cooper wire that I gave a 1-minute dip in gold electrolyte...

[quarks]

Single core copper is the best possible for low EMF connection and with direct gold plating the oxidation problem should also be fixed.

Can you share how you have done the plating?

[babysitter]

I got a 3 ml bottle gold electrolyte but it has been sitting for 10+ years, and was completely dry.
Demineralized water and a night on the heater revived it obviously.

The process is easy - a victim electrode and the wire-to-be-plated in a little container with the liquid, applying about 2V (no I didnt start a high precision meter for this...  ) and 1-2 minutes later - golden goodness ! Obvious process improvements: Get fresh electrolyte, use a gold victim electrode (goldfingers of old PC cards, maybe?)

Now I will carry it in my pocket a few days to see if it degrades, I suppose cooper would turn black when exposed to warm sweat like this.

[SeanB]

If the coating is not thick enough, and if you did not put a nickel plate over the copper to provide a backing then yes it will corrode under the gold layer.  To gold plate first you need a chemically clean oil free surface, then etch it with PCB etchant for a second or two ( chemical etch to roughen surface), rinse well then plate with nickel to give an even matt finish. then rinse again twice in pure water then apply the gold plate. When it is bright then rinse and dry.

[branadic]

The etching of the copper is to remove the copper oxid, the nickel is a diffusion barrier layer and adhesion promoter for the final gold finish.

What do you want to do with such a soft gold surface? Solder? If you want to use it for probing you better have a hard gold surface.

[babysitter]

My wife was for a while working at a PCB shop, I am well aware of the Ni/Au standard process but am I allowed to play by law! This is cheap and easy enough to prepare the wires right before the experiments and dispose them as soon as they start to degrade. I am already mistreating them and I will figure out how bad they behave compared to Multi-Contact CuTe components living in happy dry storage.

@branadic: Says a guy who I suppose applies gold to plastics when I am thinking about your employer

[branadic]

@branadic: Says a guy who I suppose applies gold to plastics when I am thinking about your employer

Sure, but the golden surface of the MIDs is only for assembling technologies (soldering, glueing, bonding, FlipChip) not for mechanical use.

[babysitter]

I want to use it between those screw terminals which are made of 4mm jacks like Multicontact PK4TS i used on my reference. Ghetto connection O0

[eurofox]

"The voltage is not generated at the junction of the two metals of the thermocouple but rather along that portion of the length of the two dissimilar metals that is subjected to a temperature gradient. Because both lengths of dissimilar metals experience the same temperature gradient, the end result is a measurement of the difference in temperature between the thermocouple junction and the reference junction."

The "characteristic voltage difference (is) independent of many details (the conductors' size, length do not matter)".

http://en.wikipedia.org/wiki/Thermocouple


You can also choose whatever practically implemented "couple" coupling joint you want as long as it is made of one and the same material and its endpoints are at the same temperature (which means no net temperature gradient).

So 1 inch of thin copper wire soldered to 1 foot of thin iron wire as in the symbol < with temperature t1 to the right (both endpoints at one and the same temperature t1) and temperature t2 at the soldered junction to the left will give the same voltage as 1 yard of copper bar and 1 inch of iron bar interconnected with 2 feet of lead tube provided that the two junctions now created at the left side both are at temperature t2. This holds only in equilibrium, i.e. all connecting points and endpoints have settled.

This is imo implications from the link. Please check for yourself. A search 'thermocouple theory" gives several sources stating similar propositions.

Right 

When I have to make a quick thermocouple I just twist the 2 wires without welding and work like a charm.

eurofox

[Kjelt]

When I have to make a quick thermocouple I just twist the 2 wires without welding and work like a charm.

Yeah for awhile. Welding it is not so difficult, use a big low voltage high amp transformer and just short the wires or use a big elco that can deliver 50 Amps for a short while. Transformer works perfect.

[eurofox]

When I have to make a quick thermocouple I just twist the 2 wires without welding and work like a charm.

Yeah for awhile. Welding it is not so difficult, use a big low voltage high amp transformer and just short the wires or use a big elco that can deliver 50 Amps for a short while. Transformer works perfect.

When I say "quick" is just for some temperature testing, is used +/-1 hour and removed after the test, I did it many times to test temperature of glass in thin film solar cells production because of <> of temperature on the surface.

eurofox

[TiN]

Mind if I join this club? 

[eurofox]

Hello Volt Nuts,

I got my version on the LTZ1000A working on slotted PCB with thermal isolation.
I already put it in a box with a power supply, 24V AC outside the box.
I added a switch on the front panel to be able to select output voltage from the LTZ1000A and output of and AD706 with 2 precision resistors (0,1% 15 PPM), a trimmer to set exact 10 V output.

I still have to work on fine tuning of the trimmer part, still to sensitive to set the 10 VDC with the 10nV exactly.



eurofox

[eurofox]

Some pictures