Which devices to trust and references to get/make ?

[lordvader88]

My little lab is growing and I don't have any really high precision to see what my DMMs, scopes and LCR meters are reading.

I have a Keithley 177 bench DMM, its old and it might have a slight problem on current or was resistance, that made me think my Byrmen BM869s was doing a little better. (I have to get the Kiethley sorted out, some knobs are wonky too, I think 1's broke.)
https://www.utwente.nl/en/tnw/slt/documentation/Equipment/GeneralEquipment/keithley177.pdf

Then there's my 1980's analog and DSO tektronix scopes. I could have a serious problem of some types, and I wouldn't know, because they're my 1st scopes. They all need a tune up at least.


What cheap but good things might I buy too home calibrate my equipment with ? Like voltage references, current references, precise resistors and caps. I have copper clad boards too, so I could also buy good chips, or what ever, and make some stuff my self....if real equipment is too expensive...but making stuff is fun anyways

The Keithley could be made super accurate again I assume (if the parts are ok or replaced)

[smithnerd]

For voltage, you could investigate the AD584 based references on ebay. They have their shortcomings, but are at least useful as a poor man's transfer standard. I wouldn't trust the supplied 'calibration data' very much though.

[coromonadalix]

The minimum would be a properly calibrated meter,  with it you can start to make your own references:    resistors values, dc or ac voltages refs, your room temperature  etc...   you need to start somewhere.

With one of my calibrated 34401a,  i partially repaired an fluke 343 dc voltage standard, the 15.000 volt reference died,  just cant get to 1kv for the moment

Measured some high precision Cadock resistors at 23 degree room temp, found a few precise capacitors ...

[bob91343]

My accurate voltmeter is an HP 3456A.  It's also very good for low and medium resistance measurements.

My best bridge is a GR 1658, very accurate.  I also have its predecessor the 1650-A which isn't as accurate but certainly good enough for most work.

I have two other accurate voltmeters, an ancient EI 3000 that is good to 0.01%, and I even want to sell it.  The other is a slideback job where you turn the dials for a null.  Also very accurate, but sits in the garage unused.

My counter is an HP 5328A that goes to maybe 500 MHz.  Its time base is adjusted with my rubidium standard to an accuracy beyond reasonable for the counter resolution.

So I can measure frequency, voltage, resistance, inductance, and capacitance precisely.  I also have an HP 3466A that's pretty good.  My signal generator is extremely accurate for frequency; it's an HP 8657B that goes beyond 2 GHz.  I can measure current also, both alternating and direct.  My fastest oscilloscope goes to 500 MHz.

While all this stuff may have commanded monster prices in the catalog, it has all been picked up as bargains either on ebay or at swap meets.  My total investment is very small.  It's just that you can't decide you need something and go after it; mostly you have to keep your eyes open and grab stuff even when you don't need it and if nothing else, use it as swap material.  You also have to be prepared to make repairs.  But then, that's true even for stuff you buy at high prices.

[bitseeker]

What cheap but good things might I buy too home calibrate my equipment with ?

I'd recommend first taking a read through the calibration process for each piece of equipment that you own in order to get a little familiar with what's involved in calibrating and adjusting it. For example, here's a copy of a manual for the Keithley 177. Table 5-2 on page 5-2 summarizes the 22-step calibration procedure.

Table 5-1 on page 5-1 contains a list of all the manufacturer recommended test equipment to calibrate the meter along with example manufacturers and model numbers. You could get lower-grade equipment, but your gear will not be more accurate than the sources that are used to calibrate and adjust it. So, you'll want to determine how good is "good enough" for your uses.

Note that once you source all the equipment to do the calibration yourself, you'll first have to determine if all your calibration equipment is in calibration.

Depending on your location, it may be more cost and time effective to have a lab calibrate your gear. You don't have to do it every year, either. Again, it all depends on how good is "good enough" for your uses. Most of my test equipment is 10 or more years old and has held its calibration well.

Of course, for some, working with and maintaining calibration sources as well as tuning up your gear is fun and a hobby itself. Hopefully, this is helpful for you to get a peek at what's involved before jumping in. The metrology forum is also the place to go for even more info.

[DaJMasta]

Unless you're willing to go full volt nut, I think a very good sanity check is just to have an instrument to compare it to.  Having a second multimeter lets you sanity-check your other meter, and a frequency range on one could be a rudimentary signal generator test.  When you get into other gear, you can do basic checks with them in similar ways - checking the RF power meter with an SA display, checking a frequency counter's oscillator with a signal generator, checking a scope's channel with a sig gen, a different scope, or even an RF generator (bandwidth testing).  It takes a very specific application to require absolute accuracy in a measurement, so cross checking gear with other instruments can confirm functionality and can show if there's a significant drift or performance issue.  Not going to check for everything, not valid for calibration, and not diving into advanced measurement techniques (and properly operating calibration equipment can be quite fascinating), but it should suffice for most hobby and troubleshooting work, and it will let you know when your gear is far enough out to really need some recalibration/repair.

Yes, it's also an excuse to amass gear, but honestly it may be cheaper than diving into proper metrology.  The odds that one piece of quality gear failing in a difficult to notice way is actually pretty low, so if you can compare to a different instrument, the odds both have failed to show the same symptom is quite low.  Also teaches you something about the limitations and operation of your equipment, though full metrology calibration sort of requires that understanding as well.

[MosherIV]

What cheap but good things might I buy too home calibrate my equipment with ?

If by 'calibrate' you mean adjust - you DO NOT!
You need equipment to accurately generate all the function, usually at 75 to 80% full scale on each range. Then you check and adjust each range.

Then modern dmm like the Brymen you have need the calibration software from Brymen. There are no calibration pots in modern dmms!

You can get cheap things like the AD584 to check dmms are working but unless the voltage reference is calibrated from a known reputable source, the vref is also unknown.

Most good dmms do not drift by much, not enough for most hobby electronics to make much difference.
Only the hobby volt nuts worry about the ppm drift on their dmms.

Probably the hardest reference is time. Even OCXOs drift. Probably the cheapest/easiest is either your national standard radio frequency broadcast channel or maybe a gps 1ppm. Never trust the frequency output of a basic gps device, it is generated by the xtal driving the gps processor where as the 1pps is generated by referencing the onboard xtal against the timing coming from the signal from the gps satalites

[Fungus]

For voltage, you could investigate the AD584 based references on ebay. They have their shortcomings, but are at least useful as a poor man's transfer standard. I wouldn't trust the supplied 'calibration data' very much though.

Some sellers do it properly (and mine seems OK).

OTOH I've heard stories where people ordered more than one and they both came with the exact same values written on the piece of paper.

There was a thread about them on EEVBLOG:

https://www.eevblog.com/forum/metrology/cheap-ebay-ad584-voltage-references-my-experiences/

You can also get used 0.01% resistors for a reasonable price on eBay:

https://www.ebay.com/sch/i.html?_nkw=vishay+metal+foil+0.01%25

I've got a few of those and they all seem to be bang-on (according to every multimeter I've ever measured them with).

[NoisyBoy]

Before going down the full adjustment route, if all you want is to check the readings of your equipment, perhaps something like the DMMCheck Plus would be a good tool to verify the reading of your equipment, and see how far off they are.  It will give you spot check on AC/DC voltage and current, as well as resistance, frequency and duty cycle.

If they are within an acceptable range of your use case, then just note it and leave it, just apply the compensation to your readings. 

If they are off enough to warrant an adjustment, then you take the next step.  You balance the equipment needed vs. the cost of paying someone to do it, and make the decision.  For the vintage equipment you have and the resolution they provide, you may even find the cost of adjustment way out-of-line and settle for the inaccuracy.  You can make the call at that point.

If all you care about is voltage, then one of those Chinese voltage reference may do the job.  I bought one out of curiosity, it came with a label with 6-digit voltage checks on a 34401a on all four voltage ranges.  I verified it against my DMMs which have all been professionally NIST cal within the last 3 months, the voltage on the label are all within 34401a accuracy tolerance.  The stability on the reference is good enough for a 4.5 digit DMM, beyond it, you will get jumpy trailing digits. 

There are multiple versions that I have seen on-line, mine is the one with acrylic case, battery, and micro-USB charging port.  The challenge for these Chinese references is quality varies greatly and it is hard to know what you actually get.  That's why the DMMCheck Plus is attractive if you don't mind only having one reference voltage.

For the Keithley 177, those are $40-50 on the used market.  Instead of trying to fix yours, you might be better off to get another working unit and use yours for parts.  Or just use the BM869, which has a better spec anyway.