Jeff, some advice from a recovering calibrataholic.

It is good to know how to calibrate instruments, know their tolerances and how much you can trust what is displayed. However, in the quest for ever greater accuracy and precision, you can fall into the vortex of chasing calibration till it prevents actually practicing the hobby you want to use the equipment for.

How much accuracy and precision is needed for hobby work? It really depends on what you hobby is. If you rebuild tube amplifiers and vintage radios, 6.5 digits of accuracy is not very useful (and probably detracts from getting anything done) and one decimal digit will probably be enough. If you fall into the average hobbyist needs, probably 3.5 digits are fine.? If you design high accuracy test equipment, measure strain gauges, or A/D or D/A converters then more digits can be very helpful.

You need to realistically look at what you are going to need for the hobby and invest at that level. You generally need some calibration standards 10X better than what the accuracy is you are calibrating for (your actual needs). Because you have a 8.5 digit DMM (great deal or good price), does not mean you need to calibrate it to that level, if that is not what your needs are. Focus on what you need for your hobby. You don't need a micrometer to frame a wall when a measuring tape is plenty accurate.

For what I do, 3.5 digit DMMs or the ~2% accuracy I can get from a well calibrated analog meter is fine 99% of the time. I typically use a 500Mhz HP scope but often don't need anywhere near that bandwidth. I have a bunch of much higher accuracy/precision instruments I can use if I ever need them.

For calibration I have settled on a few things. I splurged and have a calibrated HP 3456A. I decided I wanted one device that was much more accurate than any other I have, to calibrate against. I have a GPS disciplined oven based oscillator with 1pps, 1Mhz and 10Mhz outputs. I also have some known cap, resistor and inductor standards I have collected over the years that are occasionally re-characterized. I only calibrate to one standard source for voltage, current, resistance, capacitance and frequency.? This removes the lingering doubt that the standards might be off from each other and assures all my equipment reads the same in relation to each other. Remember the old saying about clocks and time; if you have one clock you always know what time it is, but if you have two clocks you can never be sure.

Once your equipment is calibrated it can still seem to read wrong if you are not intimately familiar with the operation and quirks of the equipment itself, and its connectivity to what you are trying to measure/display. Frequency limitations, impedance mismatches, input capacitance, aliasing and other things need to be accounted for, or corrected if possible. The presentation mentioned above is a great source for info.

Uncertainty calculations are important and bring another level of realization that equipment published specs and displayed values aren't always what they seem.

Sorry for the long winded answer but have fun with you calibration and hobby adventures.

Yes, weak coupling of the oscillator under test to a small nearby antenna connected to a sensitive receiver is the way to tune those touchy circuits.
The receiver can be calibrated to WWV on 10MHz and 5MHz, then it is easy to get 1Hz accuracy.
I am using Airspy HF+ Discovery and HDSDR software, but other SDR HW/SW combination would work, too. Preferably ones with a TCXO, so you don't need to re-calibrate often.
My "antenna" is a 5cm central wire sticking out of an SMA male connector. The shield is stripped away.

73, Mike AF7KR

The one Goldmine sells is +/- 75mV so that is why the cost is so low.
--
Dave, AA6RE

Voltage references and calibration techniques for the home lab:

See:

Jacques?? VE2AZX

I also have the dmmplus … very pleased with it.

73,

Frank
K4FMH

---
Frank M. Howell, PhD
Ridgeland, MS
frankmhowell (at) hotmail.com
---

On Sun, Dec 11, 2022 at 07:10 PM, Robin wrote:

A good rule of thumb might be to aim for a reference whose inaccuracy is 10 times less than the accuracy that is being aimed for

I do not maintain any references to check my instruments and test equipment.. I have multiple quality variable power supplies and a number of GR L, C, and R decades plus a couple of GR L and C standards, so I depend upon a 4-1/2 digit true RMS Keithly as my standard, which goes to the cal lab annually. The accuracy of the Keithly is a factor of 10 better than any of my other instruments. My need for accuracy in L and C measurements can be met by checking their accuracy with my GR elements, and the Keithly is a sufficient reference for everything else which I own.

Dale H. Cook, Retired AM/FM/TV Chief Engineer, Roanoke/Lynchburg, VA
https://plymouthcolony.net/starcityeng/index.html

开云体育

Jeff,

that's a how long is a piece of string type question. But it's still a good one. How accurate a reference needs to be rather depends on the purpose to which the DVMs which might be checked against that reference will be put. A good rule of thumb might be to aim for a reference whose inaccuracy is 10 times less than the accuracy that is being aimed for. For example, for a meter that is accurate to 1% of measured voltage, then a reference that is accurate to 0.1% would be good. For a meter accurate to 0.3%, then a 0.1% accurate reference would be tolerable, but could do better.

But that then raises the question as to just how accurate that reference is on the day when it is used. How much does it drift with time? What is the temperature coefficient? What can one check it against? Does that checking reference have any traceability?

The AD584K is, according to the spec sheet, good to, or better than, ±0.1% on the 10?V output.

Thinking of a 3? digit DVM, that is a resolution of 0.05% at FSD, or for a 4? digit DVM 0.005% at FSD. The claimed accuracy is quite another matter, but is probably specified for the meters that you own. E.G. for a 4? digit meter of specified DC accuracy ±(0.05% + 3 LSDs) that works out, at 10?V DC to ±8?mV, or ± 0.08%. Now, in regards to the reference, the question becomes who is validating whom!

My own experience with one of the Chinese suppliers of AD854 units including a built-in Li-ion cell is that the measured voltage from the 10?V output is very consistent across my small fleet of DVMs, ranging from a few years old to 30 and more years. That's consistency is very encouraging and a tribute to the basic pocket multimeter design of PCB switchery and precision resistors. (And the utility of end-of-service pulls—best sort of recycling!)

HTH, 73, good question BTW,

Robin, G8DQX

开云体育

Hello Jeff,
I understood "bumble bee" capacitors to be a style of capacitor made
by Mullard, certain values have colours which are similar to? those
of a bumble bee. I see them on ebay advertised as "tropical fish"
capacitors, probably for the same reason.? One advertiser is asking ?250
for 4 of them !!? In the hi fi section of course.
If he sells them I stand to be a millionaire as I've got 100s NOS.
Regards Ken g8beq

开云体育

开云体育

开云体育

I used to supervise a service team of 14 guys and other things. Carlson would not have been one.? Servicing is about probability, experience and logical inference but it all needs to be based on sound theory, you also need to take less than a month to fix things, glacial is not an acceptable speed setting. Droning? on and on about outer foil? again and again and again shows a lack of focus on what matters at AF & LF in the crap he deals with. But it must impress someone.

Capacitors are usualy very reliable but 60 year old paper dielectrics are often way beyond their use by date, there's not much point in NOT replacing such parts unless you have a stationary exhibit in a museum. Any old electrolytic also qualifies for retirement. Why would you leave a part known to fail in place only to have to change it some time soon? For a lot of Tube gear the big issue is with AF coupling caps, there are not that many, replacing old paper caps with modern plastic film types is a no-brainer providing that you can get decent replacements. If you speak to people used to servicing particular gear they may let you know particular liable to fail parts. Stock faults are common to all kinds of electronics, there used to be particular brands which were more liable to cause problems. Long time back there were some electrolytics in plastic cases that were very unreliable. One of the causes of electrolytics failing was sweat, one trace on a foil or paper would condem the part eventually. Bead tantalum capacitors are another "Replace on Sight" part, they fail to a very short circuit. On the other hand mil spec tantalums in cans are very reliable, the flat disk "acid" types are possably the best electrolytics ever made, now at ransome price levels.

An ESR meter needs to deal with series resistances? under 0.2 ohms for larger caps ( e.g. 220-4700 uF). The failure is mainly due to ripple current at high operating temps or running cheap electrolytics where higher rated parts are needed. So a useful ESR meter needs to read down to 0.1 ohms or lower.? It also needs to be fast settling in a pro setting, waiting more than a second or two is as useful as toothache when you want to check every at risk part. I used an Electronic Design Specialists EDS 88A when I was doing it for real, that was a very effective tool. ESR meters are absolutely vital when working on switch mode supplies. Now for a True Story about ESR.

There was one product that failed after 6-18 months, we had about 800 in the field. It was a CRT monitor. It would fail with extensive damage to the Flyback Transformer and shorted Horizontal Output transistor.? If you replaced all of the failed parts it would work again then 1-30 days later you needed to replace all of the parts again. The ESR meter found a small 22uF cap in the switching supply that had high ESR. When I checked the part it turned out that the manufacturer had changed the part, the new part had a smaller can.? Smaller can = lower ripple current rating. After a while the capacitor ESR went higher when the product was warm causing the rail to the Horizontal output to increase and do extensive damage. Turns out that some "Value Study" had saved a few cents on one cap without checking the consequences. That cost the Company it's reputation that had developed in 50 years of manufacturing.

I still use the EDS88A, but also have a? Vanwell Chinese multi-function tester, it's surprisingly good for many semiconductors and will indicate ESR for electrolytics down to 0.1 0hms. It's? too slow for pro use but much better than toothache, good for hobby users and very inexpensive.But a lot of dirt cheap Chinese RCL meters that I have seen are very inaccurate. For RCL indications for rough work? I use a Peak LCR45, that has a good wide autorange for AF and RF use and is accurate.

Regards,

Alan G8LCO

--

BTW, one of the worst ones of this bunch was equivalent to a 6M resistor (tested at 300V, cap rated for 400V).

Pete

Just tried a few. Using an HP power supply that does 0-320V, with the current limit set low, and a Simpson 260 VOM on the 50V or 10V scale.
Seems to work well.
Here's a pic - good on the left, bad on the right.
Some of the 'bad' ones are perhaps marginal - a very small deflection on the meter.

BTW, I hope I didn't offend with my criticism of Carlson. He has every right to protect his I.P., to "sell himself", etc. I just exercise my right to ignore his content.

Pete

Thanks for the comments/links so far.

The "Mr. Carlson" stuff: I watched his very 'wordy' video, where he repeatedly says how awesome his tester is, how carefully he has tested his design, and he demonstrates it over and over, and shows us how the old magic-eye-based testers are dangerous and inferior. I went to the Patreon page to find out how it works, but found nothing but buttons to click to send $money$. Annoying. I don't care about seeing his Top-Secret schematic, but would anyone be able to explain how his design works? In particular the "forecast" mode - is he doing something very special/clever?

Anyway, I gather that the two methods I proposed in the first post are fine. Alan's suggestion of using a VOM sounds practical and easy.

Pete