Are those 500MOhm resistors OK for precision stuff ?

[DC1MC]

https://www.ebay.de/itm/10-St-Ableitwiderst%C3%A4nde-500-MOhm-15-M082/401101139772

 Thanks for any info,
 DC1MC

[PA0PBZ]

Not sure what you call precision or what your application is but the listing says +/-15%, that's not really precision in my book.

[DC1MC]

That's the value, I was more thinking of the tempco.

 DC1MC

[David Hess]

High valued resistors usually or always have bad temperature coefficients compared to low value precision resistors.  In practical applications, they are used in dividers with the same type of resistor element on both sides so the temperature (and voltage) coefficients cancel out.

[Magnificent Bastard]

https://www.ebay.de/itm/10-St-Ableitwiderst%C3%A4nde-500-MOhm-15-M082/401101139772

 Thanks for any info,
 DC1MC

Like many things in engineering, that depends on what you mean by "precision stuff".  Since you did not state exactly what your application is, then I will have to go with worst-case, and assume you want this for some sort of calibration of other equipment.  In general, "precision" means:

• Absolute value very close to nominal.
• Very low or zero TCR.
• Very low voltage coefficient.
• Very good stability with time.

#1 and #2 can be taken care of by (carefully) choosing some resistors out of a batch that (in series) provide the right nominal value and near zero TCR.  Any deviation from nominal can be reduced with a trim circuit.

#3 is dependent on the materials and construction.  Resistors made of metal resistance elements are in general the best for low VCR.  PWW and foil resistors are only practical for lower values-- but when you are talking about >1M-ohm, then metal foil film is your best bet.  These can be obtained at up to 10M-ohm in value, and so 50 of these (in series) would be needed to make your "ideal" 500M-ohm resistor.  With careful selection, a very good 500M-ohm resistor can be made.  Careful attention must be given to minimize leakage currents-- so the mechanical design is very important.

#4 is also dependent on materials and construction, but we have already determined at this high of a nominal value, metal-film would be the most cost effective and practical.  In this case, for good long-term stability, the final resistor must be artificially aged by baking in an oven (at max temperature for the assembly) for a minimum of 90 days, followed by many (20+) cycles of very cold and very hot.  The final trim is then adjusted and the entire assembly is placed in a hermetic package.  For this type of resistor, the easiest, lowest cost, and most practical hermetic package is a glass tube made of soda-lime glass, with Dumet wires exiting the seals at each end of the tube (like Victoreen [now Ohmite] glass encapsulated high value resistors).  The glass tube is back-filled with industrial Argon before final sealing.  Dumet wire is TCE matched to soda-lime glass, so don't use boro-silicate (Pyrex) glass.

If you are not looking for this level of accuracy, precision, and stability, then one of the RX-1M glass-encapsulated resistors from Ohmite might be "good enough" for your application-- and you can buy these off the shelf.

Hope that answers your (very vague) question.

-MB