Alternatives to Fluke 742A for Calibrating Fluke 5720A

[Hawaii596]

I'm shopping for Fluke 742A's in the values needed to calibrate ohms on 5700A/5720A.  They require (among other values) 1.9 Ohm, 19KOhm and 19 Megohm.  I already have all the 1 and 10 values, but none of the 1.9 and 19 values above.  Any other models that will work well that may be more cost effective, or inputs from experienced metrologists who have done 5700A/5720A as to what we MUST have to properly do these ohmic values.  Could be that it is only 742A's.  Just wondering if anyone has any wisdom on this.

[beadwork]

The artifact calibration on the 5700A/5720A only uses 1 ohm and 10 kohm resistors (plus the 732B).

[Hawaii596]

This is for the full cal.

[rjf1957]

What specs are you calibrating to?

We use the 99%, 1 year specs for our units.

I don't think that the 1.9 and 19 value 742s are "required." You just need standards with low enough uncertainties to meet the ratio requirement.

We currently use a Guildline temperature stabilized box or single units in a temperature controlled air bath.
We had a 1.9 & 19 value box built, but found that it really did not get us better results, and it was just one more standard to maintain.

In our case, the 0, 1, and 1.9 ohm values can be measured directly with an 8508A. Then we use 1 & 10 resistor values, and an 8508A in ratio mode for the rest. Most values are guardbanded, and it requires a very tight guardband on the 10 MOhm value.
But, an extra artifact cal or two is cheaper than sending the unit to Fluke.

We are revising our standard resistor specs to improve the ratios and are looking at using our current comparator bridge. Do I recall correctly that you have a 9975?

[Hawaii596]

I'm still looking through the 5700A/5720A procedure. I'll have to look and see how they use them.  The values called for are:  1Ω, 1.9Ω, 10Ω, 10 kΩ, 19 kΩ, 10 MΩ,
and 19 MΩ.  I have an 8508A.  So sounds like you're saying I can use the 10 value on a range for both 10 and 19 type values.  Make measurements at the 10 values, and apply a range error based on the reading from the 10 value resistor to use for certifying the 19 value resistance output of the calibrator?  I guess as long the uncertainties are still fine?

[HoggyBank]

If you have the 01 option on your 8508A, you can make a poor man's bridge.  Fluke has a white paper on it.  You just have to be sure your standard and your UUT are being measured on the same range for the best accuracy.  So your 1 Ohm standard should cover your 1 and 1.9 values, 10 will cover 10 and 19 Ohms... and so on.

[Hawaii596]

Good points. Getting ready to qualify/validate the methods for 5700A/5720A.  I do have option 01 on my 8508A's and I think I have a copy of that White Paper.  I'll just need to be sure to maintain adequate T.U.R. for that measurement.  Could also be I could use my primary resistors and get (perhaps)ratiometrically better results (Thomas 1 Ohm and ESI SR104 10K Ohm).

[rjf1957]

We even do 10 to 190 values (19:1) in some cases to get the best standard resistor uncertainty. Example: 10 kOhm standard and 190 kOhm UUT value. You just have to make sure use the correct Transfer Uncertainty for the range in use on each set of terminals.

[measure]

Hawaii, I think the ratio accuracies on the 8508A are sufficient for verifying the '1.9' values on the 57x0A calibrators, using a cardinal value standard, e.g., 1 Ohm, to transfer to its 1.9 relative. The primary concern with the 8508A when doing transfers is to keep it on the same range as the reference resistor, eliminating range-to-range uncertainty errors (which explode rapidly!). You will also likely need to exercise care at 1.9 MOhms and above, where leakage and other error sources become significant. Good Luck!

[jwilley127]

Check Low Thermal Electronics Std. Air Resistors - Model 770 Series
http://jswilley.com/files/45782800.pdf
High Quality - Reasonably Priced- Great Alternative