Ultra Precision Reference LTZ1000

[Galaxyrise]

Galaxyrise, could you have damaged your LTZ1000? I's rated 0.1V max forward bias on the heater pins relative to pin 4 which would usually be at around 0.5V, ie pin 2 and 3 must always be higher than 0.5V

It's certainly possible, which is part of why I'm asking here.  But my heater has only ever been floating, or at least at 1V.  Is floating bad?

[janaf]

I do not think floating could do damage. What about at power on / off, with non-floating heater?

[janaf]

A plot of Impedance of the LTZ1000 zener only, at room temperature, no temperature stuff. Possibly some self heating but the measurements where in mS long bursts only. The impedance was approximately (edit) 33 ohm (0.2V / 6mA) for two samples. Measured with a NI 4132 precision SMU.

[Dr. Frank]

A plot of Impedance of the LTZ1000 zener only, at room temperature, no temperature stuff. Possibly some self heating but the measurements where in mS long bursts only. The impedance was approximately 0.03 ohm (6mA / 0.2V) for two samples. Measured with a NI 4132 precision SMU.

Sorry, but differential resistance R = dU/dI, not the other way round.
Therefore R= 0.2/0.006 ~ 33 Ohm, which is in accordance with the datasheet (20..60 Ohm)

Frank

[janaf]

  how embarrassing, but who cares about Ohms law anyway..... File updated.

[quarks]

While updating the 8.5 Digit DMM thread, I had the idea to show a comparision of the LTZ1000 board designs used in these DMMs.
So here is first collection. If anyone has more/other/better pics please share.

[quarks]

So, now I am curious... Why is there an LM199A [LM299A/LM399A] in the Datron/Wavetek 1271 [as well as an LTZ1000]?

1271 using LTZ1000 + LM399
1281 using 2x LTZ1000
That is the main design difference and the reason why the 1271 specs are not quite as good as the 1281.

[CaptnYellowShirt]

I've been wondering about the 'under power' aspect of the burn-in process. Does it matter if the device is powered at all during this period? Annealing would seem to be dominated by temperature effects. And only if one gets into high current densities does electromigration take effect?

[ManateeMafia]

Ken,

Are you thinking of AN42? Page 4 shows 4ma for the LTZ1000.

I also came across this link http://indico.cern.ch/event/162546/contribution/3/material/slides/1.pdf while searching. The circuit is listed on page 25.
I have not seen it before and I was wondering if this was from another publication or from an actual design.

[Vgkid]

How much do the econistors run?

[branadic]

I today received this little board, okay it didn't receive I had to collect it from the toll. It was named to be a LTZ1000 reference. The curious fact is, that the IC that used to be the reference has absolut no marking on it, but the circuit indicates a LTZ like style schematic.
Does anybody know something about such a board? It seems to be part of some test gear, beside the Linear Technology label and the date code 10/84 there is a mirrored "Superman" icon on top of the pcb. On the backside is another date code 46-84 and an icon that seems to be a growing sun?

I guess the first thing to do is to reverse engineer the schematic before powering up the beast.

[TiN]

So that was you who snatched it off my ebay bid list.
Great job 

[CaptnYellowShirt]

Very interesting... maybe someone at Linear-Tech is taking stuff out of the garbage bin and selling it on fleaBay...

Bob Dobkin's retirement plan?

[branadic]

So that was you who snatched it off my ebay bid list.
Great job 

Me, what? 

That appears to be a similar board to the one used in Linear-Tech ap-note #86.

You are damn right. Wow, I didn't expect that.

[babysitter]

Is it a LM7872? Is it a LM399? No it is a LTZ1000!

I prefer resistors to lay flat but besides that... beautiful !

[TiN]

Those prices really close to Z202's from VPG. And yes, I'm still waiting for my order.

[Vgkid]

Those are pricey, i will need to see about precisionresistor.com as well.

[Andreas]

I today received this little board,

Who grabbed it from J.W. s desk at the museum?

I priced out a 10K resistor [standard 0.01%], and here are the prices I got:
10K, 0.01% p/n "8G16A10K"

You will only need the 0.1% types they should be available at around half the price.

RS-Components has some resistors out of the UPW50 + UPW25 series which are similar.
I have them in my design. see:

https://www.eevblog.com/forum/projects/ultra-precision-reference-ltz1000/?action=dlattach;attach=49180;image

With best regards

Andreas

[ManateeMafia]

Those are pricey, i will need to see about precisionresistor.com as well.

I'm *know* that "Econistors" have an excellent track record over many years of use in military, aerospace, and automotive circuitry.  The resistors from Precision Resistor *may* be just as good [or even better!], but I have no historical information about them, and so I am hesitant to use them...  If anyone has long experience successfully using their products, please post about your experiences here!

I have purchased three of their MC-7 kits, one standard and two with values of 2.5 and 5. They were extremely helpful and quoted me accurate lead times of 10 weeks for the two non-standard sets. I also had them make a 10M 2.5W resistor that rounded out my original MC-7 kit. The prices were not cheap, but the kits were made using 1W non-inductively wire wound parts. I used them since I frequently travel near their business.

If you price out similar resistors, you will find that the 1W 5ppm resistors can cost much more. I have not priced smaller wattage units, but you can expect a quick reply from them if you send them your specs. I would imagine that the pricing would be competitive to what you will find elsewhere.


I do not have the capability to accurately verify the tempco but they measure well within tolerance on my 3458A. The quality seems to be excellent but I do not have much historical information on their performance.

[Dr. Frank]

Anyway, what I said earlier still stands-- I think one could build up an LTZ1000(A) based voltage reference using these resistors, and the circuit would be just as stable as one built with foil resistors...  Only one resistor needs to be "special order"-- the 12K5 for the LTZ1000A, [or 12K for the LTZ1000].

Regards,
Ken

As described here: https://www.eevblog.com/forum/projects/ultra-precision-reference-ltz1000/180/, we already DID it! (baysitter an me)

For the LTZ1000, the 5 needed ECONISTOR resistors of 120, 1k, 12k, 2x70k, all 0.1% are available from stock, between 6 and 10 € today, I would guess.

My two references are still stable / in accordance against HP3458A and Fluke 5442A to around 1ppm after > 5 years.

babysitters LTZ1000A reference did not move more than 0.2ppm after 1/2 year or so.

Frank

[branadic]

I've analyzed the circuit board I got, there are a few little different values for the resistors used, here is what I found:

R1: 125R Vishay KS24248L4 0.1% (datasheet value 120R)
R2: 79k7 0.1% (datasheet value 70k)
R3: 79k7 0.1% (datasheet value 70k)
R4: 11k91 0.25% (datasheet value 13k)
R5: 970R 5ppm (datasheet value 1k)

I will post the complete schematic and circuit board if completed.

[max666]

I've analyzed the circuit board I got, there are a few little different values for the resistors used, here is what I found:

R1: 125R Vishay KS24248L4 0.1% (datasheet value 120R)
R2: 79k7 0.1% (datasheet value 70k)
R3: 79k7 0.1% (datasheet value 70k)
R4: 11k91 0.25% (datasheet value 13k)
R5: 970R 5ppm (datasheet value 1k)

I will post the complete schematic and circuit board if completed.

Yikes!  Those resistors seem to have drifted a little!

I don't think branadic measured these values. I think he just read the labels on the resistors   

[retrolefty]

I've analyzed the circuit board I got, there are a few little different values for the resistors used, here is what I found:

R1: 125R Vishay KS24248L4 0.1% (datasheet value 120R)
R2: 79k7 0.1% (datasheet value 70k)
R3: 79k7 0.1% (datasheet value 70k)
R4: 11k91 0.25% (datasheet value 13k)
R5: 970R 5ppm (datasheet value 1k)

I will post the complete schematic and circuit board if completed.

Yikes!  Those resistors seem to have drifted a little!

I don't think branadic measured these values. I think he just read the labels on the resistors   

 In that case we can't blame drift. 

[branadic]

I don't think branadic measured these values. I think he just read the labels on the resistors

Correct, this are just the printed values on the parts, I didn't measure them!

[max666]

In that case we can't blame drift. 

We can blame the engineer drifting from the recommended datasheet values