Resistor Set for LTZ1000 positive standard 7V circuit

[cellularmitosis]

Quote from Prime Technology.

[cellularmitosis]

I wrote some scripts to crawl the ebay pages of seller "hifi-szjxic" (http://stores.ebay.com/hifiaudioic/) and list all of his resistors, dividers, etc, in formats which are a bit easier to browse.

These scripts run every day just before 9AM central, so these lists will be updated daily.

Here's a list of all of his metal foil resistors, sorted by resistance value.  The format is (resistance value, description of the listing).  I haven't added support for links yet, so if you see something you want, you'll have to copy/paste the description into his store search box.

https://ssl.pepas.com/hifi/resistors.txt

Here's the above list, but only entries which contain the letter "Z":

https://ssl.pepas.com/hifi/resistors-z.txt

Here's the list of ratios (between 12:1 and 13:1) which can be made using pairs from the above list of resistors:

https://ssl.pepas.com/hifi/resistor-ratios.txt

Same thing, but where both resistors in the pair contain the letter "Z":

https://ssl.pepas.com/hifi/resistor-ratios-z.txt

Finally, here's the list of all of the dividers he sells, sorted by ratio:

https://ssl.pepas.com/hifi/dividers.txt

Hopefully these will be useful for other LTZ nuts on a budget

[TiN]

Now you just made it that easier to clear his precious resistor inventory  

[cellularmitosis]

Now you just made it that easier to clear his precious resistor inventory  

I did hesitate selfishly for a moment there

On the other hand, he has over a thousand listings of resistor values — he needs all of the help he can get .  The real rare commodity were the dividers, and he has been out of dividers useful for the LTZ for a while now.  The next most important value is one of the 70k resistors, but that value isn’t specific, so he will have options there for a long while.  The next most likely to run out is 120R, but that is only 15% as important as R4/R5, and he has a number of options left for those.

[rigrunner]

SO FAR I have  ordered 3 sets of everything.

Did you pay Rhopoint anything extra to get a response?  My quote from the beginning of January still says STATUS_REQUEST and emails to sales@ are seemingly ignored.
 

[cellularmitosis]

No, but I did have about two weeks of back-and-forth with quotes until I was finally able to place an order.  There were a few mistakes in the quotes and it took a few days for them to respond each time.  I also had to bug green after the fact about having paid with PayPal (they sent me a bill in the mail, and PayPal said my payment was still “pending” because they had not accepted it).

Everything did work out with enough persistence though!

[rigrunner]

Surprisingly, a little after posting my earlier message i had a response from a guy here in the UK.  He apologised for the delay and said he'd look into it on Monday.  So far they don't seem to be the easiest of companies to give money to 

[cellularmitosis]

Hello,

A fellow board member messaged me asking for some advice about where to find metal foil resistors for the LTZ1000, and I thought my reply might be useful to others who are just getting started, so I'll copy my reply here (and also bump this thread back to the first page

Yup, here's the deal:

The 400k resistor is a value which is used to "tune" the temperature coefficient of the whole assembly.  The board works without it (I haven't populated it yet), then you measure the temperature coefficient of the whole circuit, then try a value (maybe start with 390k), see how that works, then try a little higher like 470k, then a little lower, etc.  You don't need to worry about that resistor yet.

For the metal foil resistors (R1 through R5), there is an ebay seller in china who has a stockpile of metal foil resistors, for about 1/3 of the price of new metal foil resistors, for example look at this listing:

https://www.ebay.com/itm/1x-Z201-100K00-0-01-Vishay-Z-Foil-Series-Metal-Foil-Resistors-Y1453100K000T0L/111862789497

It is a bit of a pain to comb through all of his listings looking for the values you need, so I wrote a script which looks through all of his listings every night, then publishes a sorted list of all of them:

https://ssl.pepas.com/hifi/resistors.txt

That's a list of all of his metal foil resistors.  To buy one, you can copy a bit of the text and paste it into ebay's search bar, e.g. "RNC90Y 100K00 VR", and the listing will come up.

Here's a list of his dividers:

https://ssl.pepas.com/hifi/dividers.txt

It looks like the "12.345294:1" ratio is the only one he currently has which is useful for this circuit.

If you don't want to use a divider, you can use two discrete resistors to make the ratio.  Here's a list of all of the resistor combinations which make ratios between 12:1 and 13:1:

https://ssl.pepas.com/hifi/resistor-ratios-12.txt

I've been using 12.3:1 or 12.5:1 in my circuits so far, which works fine if your reference will be at room-temperature.  13:1 is the default value recommended in the datasheet circuit, but that runs a little hotter, in case you need to run the circuit in a slightly hotter environment (i.e., inside of an instrument which gets a bit warm, like 35 or 40C).  In the 3458A meter, HP uses a ratio of 15:1 which runs very hot.  This means the inside of the meter can get up to around 60C and the reference still works.  However, the hotter you run the reference, the faster it drifts.  That's why I try to use 12.5:1.

This ebay seller has two different grades of resistors.  The regular metal foil (S102C, RNC90Y, etc) are going to be about 5ppm/C tempco.  The "Z-foil" resistors (Z201, RNC90Z, etc) will be under 2ppm/C tempco.

These are more expensive ($10 to $15 each for the z-foil, the regular ones are like $6 to $9 each), and I keep a separate set of lists for only z-foil resistors:

https://ssl.pepas.com/hifi/resistors-z.txt

https://ssl.pepas.com/hifi/resistor-ratios-z-12.txt

There is also an ebay seller in Poland who has a stockpile of RCK02 metal foil resistors (also by Vishay, but blue).  These are about equivalent to the "regular" resistors from the "hifi" seller (about 5ppm/C), but they are even cheaper ($4 to $6).

Here are his listings:

1%, 5ppm, 5R to 148k, $2.77 https://www.ebay.com/itm/1-5ppm-0-5W-High-Precision-Stability-Foil-resistor-RCK02-VARIOUS-VALUES-150R/132261333295

0.1%, 5ppm, 2R to 3.3k, $4.15 https://www.ebay.com/itm/0-1-5ppm-0-5W-Very-High-Precision-Vishay-SFERNICE-Foil-resistor-values-2R-3K3/132092526341

0.1%, 5ppm, 3.44k to 169k, $4.15 https://www.ebay.com/itm/0-1-5ppm-0-5W-Very-High-Precision-Vishay-SFERNICE-Foil-resistor-values-100K-/142274715625

0.05%, 5ppm, 10R to 147k, $4.85 https://www.ebay.com/itm/0-05-5ppm-0-5W-Very-High-Precision-Vishay-SFERNICE-Foil-resistor-values-10R/132121105979

0.02%, 5ppm, 6R to 99k, $5.54 https://www.ebay.com/itm/0-02-5ppm-0-5W-Very-High-Precision-Vishay-SFERNICE-Foil-resistor-VARIOUS-VALUES/132099860265

0.01%, 5ppm, 60R to 180k, $6.24 https://www.ebay.com/itm/0-01-5ppm-0-5W-Very-High-Precision-Vishay-SFERNICE-Foil-resistor-VARIOUS-VALUES/142282995078

For this circuit, the precision of the resistors doesn't matter (1%, 0.1%), only their long-term drift and tempco matters, so if you use the blue resistors, try to get the cheaper 1% ones.

Of these resistors R1 - R5, some matter more than others.  Several EEVblog members have measured how important each resistor is, and I've compiled there results here:

https://www.eevblog.com/forum/metrology/resistor-set-for-ltz1000-positive-standard-7v-circuit/msg1386014/#msg1386014

https://github.com/pepaslabs/px-ref/blob/master/kicad/releases/v2.4/README.md#r1-r2-r3-r4--r5-r45

The R4/R5 ratio's drift is attenuated by about 100:1.  So if that ratio drifts by 100ppm, the voltage of the LTZ will be moved by 1ppm.

Even with the cheap LTZ's (5ppm/C), they would have to drift by 20C to see the voltage move by 1ppm (7uV).  But maybe the two resistors of the ratio drift in opposite directions, which means only 10C would make a 7uV drift...

Next most important is R2, which is 3x less important than the ratio, then R1.  So it would have to drift like 300ppm to caus a 7uV drift.  Then R1 is less important than that, and R3 is less important than that.

So, if you aren't planning on trying to measure the temperature coefficient of your board, don't bother with the z-foil resistors, just get any metal foil resistors and you will be fine.

If you are planning on trying to dial in a very low temperature coefficient for your board, then maybe try to use a divider or z-foil for the R4/R5 ratio and for R2, then use the cheaper regular metal foil for R1 and R3.

oh, wait I just realized that divider has a weird lead spacing:

https://www.ebay.com/itm/1X-300191-16K0674-1K3015-High-Precision-Voltage-Divider-and-Network-Resistors/121921159091?epid=2095964793&hash=item1c63113fb3:g:NY8AAOSwu1VW4u~i

ok, so forget the divider.

For z-foil, use these two for the ratio ($9 each):

https://www.ebay.com/itm/1x-RNC90Z-15K000-BR-Vishay-RNC90-Series-Metal-Foil-Resistors-Y118915K0000BR0L/121779420465?hash=item1c5a9e7d31:g:mTkAAOSw4HVWErRg

https://www.ebay.com/itm/1x-RNC90Z-1K2000-BR-Vishay-RNC90-Series-Metal-Foil-Resistors-Y11891K20000BR0L/111790693166?hash=item1a073e9b2e:g:mPIAAOSwKIpWE~JL

and use this for R2 ($9):

https://www.ebay.com/itm/1x-RNC90Z-68K100-BR-Vishay-RNC90-Series-Metal-Foil-Resistors-Y118968K1000BR0L-/121804687045

Then use a regular one for R1 ($7):

https://www.ebay.com/itm/1x-RNC90Y-120R00-BR-Vishay-RNC90-Series-Metal-Foil-Resistors-Y0089120R000BR0L/111760942444?hash=item1a0578a56c:g:78MAAOSwMmBV5p7A

and for R3 ($7):

https://www.ebay.com/itm/1x-RNC90Y-71K500-BP-Vishay-RNC90-Series-Metal-Foil-Resistors-Y008971K7500BP0L/121794231241?hash=item1c5b807bc9:g:x-8AAOSwl9BWKIAL

Cheers!

Of course you can also get a custom set of Ultrohm resistors made by Edwin Pettis, which are a very good value!  (Excellent long-term drift, and the tempco will typically be under 2ppm/C, and the are usually under $8).

[hwj-d]

Hello,

A fellow board member messaged me asking for some advice about where to find metal foil resistors for the LTZ1000, ...

Maybe the same fellow ask me that too, 
and i answered him to directly search for these ebay article-numbers:

• 112337652641
  (Y0007120R000T0L)
• 121749693656
  (Y00891K00000VR0L)
• 121777259014
  (Y008913K0000BP0L)
• 121794231241
  (Y008971K7500BP0L)

But your lists are of course much more extensive 

[Andreas]

Hello,

perhaps interesting for volt nuts in EU:
I have found that Reichelt stocks some 8G16D resistors.
At least the 1K resistor and 50K (if you can live with the deviation to 70K).

https://www.reichelt.de/drahtwiderstand-axial-0-33-w-1-kohm-0-1-gen-8g16d-1k0-p233202.html?

https://www.reichelt.de/wirewound-resistor-axial-0-33-w-50-kohm-0-1-gen-8g16d-50k-p233208.html?

https://www.reichelt.de/drahtwiderstand-axial-0-33-w-100-ohm-0-1-gen-8g16d-100r-p233199.html?

with best regards

Andreas

[Andreas]

Hello,

I have asked for a quote for 5 resistor sets (120R, 1K, 12K5, 70K) + 10K/25K for the 10V transfer.
The answer was that they only sell the remaining stock and do not want to sell the resistor in future.

with best regards

Andreas

[niner_007]

I have something to contribute

Use VPG 303381, for best factory effort VHD200 12.45K/1K, with maximum 0.5ppm/C over the 20C to 30C range

[Kleinstein]

A super low TC of less than some 5 ppm/K is not really needed. The resistor TC is attenuated by about a factor of 100. The more difficult factor is the long term stability. If a resistor ratio of 1:13 and a higher resistance is OK, one could consider an array like NOMCA for the set temperature.

If it is about the 10 V output, the resistors that really matter are those for the 7 to 10 V step. Here something like the VHD200 can be justified.