EEVblog Electronics Community Forum
Products => Test Equipment => Topic started by: solock on January 21, 2014, 02:46:58 pm
-
It seems that with the plethora of decent vintage or near vintage test equipment on ebay, and with the typical ebay seller (at least for the best apparent bargains) doing nothing to indicate anything past it powers on if even that level that there is some potential to offer a less than pro grade calibration service.
Can anyone suggest a US vendor that is hobbyist friendly, that is not priced 2-3x the used equipment purchase price for a checkup/cal on meters, function generators, audio analyzers?
As you go back in the product lines you approach some $100 USD buys that are of unknown origin but really could go for a less that 200-300 calibration. A cal "lite" if you will.
I've seen some users mention a device that wont meet calibration, but still is relatively accurate. I would love to know the state of some of my equipment but really cant sink a ton of money into it.
Thanks in anticipation of your suggestions.
Steve
Specifically for me the need is in the audio end of the spectrum so nothing extreme, but am open to suggestions in any direction.
-
Are you familiar with http://www.voltagestandard.com (http://www.voltagestandard.com) ?
You can get a cheap reference and just check your gear with it. All depends on the accuracy you are going after of course
-
In your case, I would recommend you just buy a used Keithley 2015THD as this is a decent and versatile bench DMM that also offers DSP based audio synthesis and analysis.
Because the AF part of the meter is DSP based I would expect the accuracy (over time) of the meter to be sufficient
to let you check the accuracy and fidelity of various types of audio gear for many years.
I bought one recently for this purpose plus it has very powerful capabilities when operated remotely. I'm not so keen on using it as a general DMM but I now see it as my in house gold standard for AF based stuff. You can buy these meters for as little as £315 used on ebay UK.
So with the 2015 you get a decent (but probably OTT for most of us) DMM, plus a decent DSP based AF signal source
(calibrator) plus a decent DSP based AF voltmeter plus it can measure harmonic distortion terms and also give a THD
figure. You can use it as an accurate milliohmeter and a reasonable frequency counter at AF.
Of course, this isn't the same as having a formal calibration of your various AF test gear but I would rather have the versatility of the 2015 instead of a fancy cal certificate showing the results of a few basic cal checks that have limited merit the moment the equipment leaves the cal house. This is because it is only known to be 'in spec' on the day of the calibration.
-
I actually have a recently purchased (used)2015-P, and it was in superb condition. Physically clean, but unknown accuracy. Seems correct but again it would be nice to know how close it is to calibrated.
I have an older (obviously) 331a and a few other decent pieces that would be nice to know that I have no glaring issues with.
I looked into the voltage standards mentioned above and they are backordered temporarily.
THanks for the insight,
Steve
-
Hi Steve, I'm not really a calibration junkie like some posters on here but I did some initial THD and harmonic checks on my 2015THD a while back which you may find interesting :)
You can see the results for the square wave were very good :) I don't really need this level of precision but maybe you can see why I consider it a useful gold standard :)
https://www.eevblog.com/forum/testgear/keithley-2015thd-initial-tests/ (https://www.eevblog.com/forum/testgear/keithley-2015thd-initial-tests/)
I managed to do a few tests of the meter for THD on a really high quality square wave.
If the meter is only set to measure THD for 8 harmonics then the THD can be predicted to be -7.657dB with a simple excel spreadsheet. (2015THD meter actually measures -7.650dB THD)
If the meter is only set to measure THD for 20 harmonics then the THD can be predicted to be -6.804dB
(2015THD meter actually measures -6.797dB THD)
If the meter is only set to measure THD for 50 harmonics then the THD can be predicted to be -6.503dB
(2015THD meter actually measures -6.497dB THD)
-
What I have done over the years is to carefully buy lab grade test equipment (usually at very low cost) that I can use as long term gold standards that offer a realistic accuracy for my needs.
There's various ways to cross check these instruments for decent accuracy (often using basic physics and some simple hardware) and I feel I have no need to send any of them for formal calibration. Besides, I work at a company that doesn't trust the majority of calibration houses with our best test gear. We would rather they left the lids on the gear and just gave us a cal certificate if the equipment meets its basic specs. So this has tainted my view somewhat. There's no way I'd let a typical UK calibration house loose inside any of my generators or spectrum analysers or (heaven forbid) my network analyser.
-
For personal/hobby use, I have "reference standards" for Voltage (Weston standard cell), Resistance (10K ohm .005% Vishay precision resistor in a box), and Frequency (FE-5680A rubidium standard with programmable divider down to 1 Hz). Occasionally, I will haul one of my home standards into work, for a crosscheck against a real NIST traceable instrument.
Have never felt the need to pay for a "real" calibration for any of my hobby gear, but then again I don't do anything at home that demands very high accuracy.
-
reagle, does it make sense to test a six digit multimeter against a three digit voltage standard?
I mean, sure, it'll show you if things are WAY out of whack. If that's all you need then why not just buy a second DMM?
-
It all depends on what you are checking. Isn't DMMcheck a bit better than 3 digits though?
reagle, does it make sense to test a six digit multimeter against a three digit voltage standard?
I mean, sure, it'll show you if things are WAY out of whack. If that's all you need then why not just buy a second DMM?
-
Voltage and resistance standards are getting better but unless you luck out like Dave did and get an EDC voltage standard or a Fluke calibrator for $100 you pretty much have to make do with 3 or maybe 4 digits. There is now an AD584LH-based Chinese voltage standard for $20 on ebay that is good to 4 decimal places. There is an EEVBlog thread about them which verifies the accuracy. The other thing is to buy equipment which is known not to drift too much over time. Keithley bench DMMs, for example, tend to keep to their 5.5 or 6.5 digit accurancy over a decade or more after cal. Cross check all your DMMs with whatever known standards you have and then watch for drift. Not much else you can do unless you want to spend money on a cal.
-
All good advice. I'll look into one of the voltage standards when they are back online.
The buy well and cross check methodology is how i have been proceeding, but none of my devices has been truly cal'd in years, so the know starting point was the largest concern.
Thanks for all the insight.
Steve
-
Go on eBay and look for Ronan or Loveland Controls brand calibrators. You can find them rather inexpensively if you have the patience to wait for a deal to come along. They are far better than the little palm-sized calibrators and offer more features.
-
my take on doing some cal work is to have a voltage standard (the dmmcheck device, for example) and then a really good kelvin-varley divider (such as general radio) so that you can check linearity.
I've often thought it would be a cool idea to have a gathering (like at a hamfest or audio show, such as burning amp in san francisco) and organize a calibration 'party' of some kind. it would be informal but at least we could have a chance to compare our standards and maybe tune up some devices, while at the event.
I've never seen this happen, but it would be great if we could get enough interest at some kind of event or gathering and pull this off. a transfer standard is better than nothing ;)
-
Another generous fellow in this forum that offers reference caps for cross checking vs his 5 digits US$ priced :P LCR meter -> Reference for LCR or ESR meters (https://www.eevblog.com/forum/projects/reference-for-lcr-or-esr-meters/)
-
Sorry for raising this from the dead.
I would like to offer my calibration services to members of this forum. I have acquired a lot of metrology grade equipment, and have everything traceable to an annually cal'ed Agilent 3458A.
I have performed calibrations for hand held and 6.5 digit multimeters with this equipment. I can source 1uV to 1kV of AC and DC, have resistance standards at 3ppm, and DC/AC current up to 2amps. Also frequency from 10 Hz to 1MHz.
I will post a formal thread at a later date, but let this serve as pre-announcement and gauging interest!
Thanks!
-
I'd be very interested in getting my Keithley 197A (and perhaps my fully loaded 195A) calibrated. I think the 197 is pretty close to spot on, but the 195 may have some drift to it. Both are digital calibration based and I can provide the relevant instructions from the manuals.
Sent from my Smartphone
-
One of the best things I have come across to go it on the cheap side to have a standard for DCV, ACV, and ohms is the HP 3456A 6.5 digit multimeter. It is fairly easy to pick up a working example of this excellent meter for a hundred bucks or so, and most calibration labs only charge around a hundred or so to calibrate it which sort of surprises me. So for $200.00 (That being less than what you would pay for a 4.5 digit handheld Fluke) you can own a metrology grade calibrated standard for comparison for your shop.
I use one of the Ronan X86 process calibrators for output, and for a small unit you can pick up for around $100.00 or so it has super-stable output, more than adequate to do checks and balances on 3.5/4.5 digit multimeters.
-Mark-
-
I agree these 3465As are excellent meters! Unfortunately, a lot of them have problems and need their electrolytic caps replaced. I am almost done fixing mine up... the through-hole component replacement is quite easy, the hard part is figuring out exactly what is wrong. Service manual always helps!
The one issue I always ran into when I started maintaining "calibrated" equipment in my lab, was the lack of a source for wide range of accurate and stable voltages, currents and frequencies. It's great when you have a meter to measure something accurately, but you cannot transfer that accuracy to another meter without calibrating against a source.
Most high resolution meters (older ones such at 3456A) require several source voltages to calibrate each range.
The large majority of hand held meters usually only require 1 voltage (often around 3 volts), from which the internal ADC scales the rest of the ranges.
I have calibrated several 3456As against my standards, they are tedious no doubt... Calibration shops will often charge more to calibrate a 3456A than a more modern equivalent, due to the amount of trim pot fiddling required to make the meter accurate.
-
It surprised me when looking around that the $100.00 range seems to be where all the calibration labs are at for doing the 3456A, I thought it would be much more due to the fact it isn't "soft" calibrated. I just fixed one up and fortunately it is still accurate enough by comparison to my two Fluke 8505A's that I don't need to spend the money and can just run with it. I had so much fun fiddling with the 3456A that I bought another 3456A fixer-upper on Fleabay :)
Like I noted in my last post, the small industrial process calibrators make for very good small level DCV, DCA, and ohms standards for the hobby shop as an alternative to an expensive professional calibrator. About the cheapest Fluke calibrator one can pick up today is a 515A, and I'm sure many of those are getting out of whack at their age.
-Mark-
-
One thing to know is that calibration isn't calibration but just checking whether a piece of equipment is still within it's specifications or not (pass/fail). Adjusting a piece of equipment is an entirely different story.
-
I'm for it.
I'd be interested in whatever you can offer.
My purpose would be to reconfirm the reading of Voltagestandard and old Geller products. I have access to a true calibration facility, so I can compare results of an unofficial cal versus the real thing. I think unless you need the cal data for maintaining device drift records, low cost 'hobby' cal can work for most purposes.
Sorry for raising this from the dead.
I would like to offer my calibration services to members of this forum. I have acquired a lot of metrology grade equipment, and have everything traceable to an annually cal'ed Agilent 3458A.
I have performed calibrations for hand held and 6.5 digit multimeters with this equipment. I can source 1uV to 1kV of AC and DC, have resistance standards at 3ppm, and DC/AC current up to 2amps. Also frequency from 10 Hz to 1MHz.
I will post a formal thread at a later date, but let this serve as pre-announcement and gauging interest!
Thanks!
-
An alternative solution for US based work is obtaining "generally" stable volt, ohm etc., sources [ meaning stable for minutes to hours ] and then renting or obtaining a known calibrated meter, say a 34401a, they go for < $100/mo delivered. This is the same cost as sending the 3456a level DMM for calibration. Call for best prices to various rental suppliers such as:
https://www.microlease.com/Products/Keysight-Agilent-Technologies/Multimeters-Data-Acquisition-Counters/34401A?basemodelid=748 (https://www.microlease.com/Products/Keysight-Agilent-Technologies/Multimeters-Data-Acquisition-Counters/34401A?basemodelid=748)
You can then 'transfer' the calibration by measuring your stable sources with the calibrated meter, compare it against your DUT, adjust the DUT as needed until its in spec. It compares very well against a true formal cal. Its vital the calibrated meter and DUT reading are adjusted to be near identical to avoid the TUR effect, otherwise you'd best rent a 3458a, which is expensive.
The problem with maintaining a high quality reference for oneself is they still will not be a primary standard, so comparison against a primary standard via traceability will be needed at some point. Owning a 732a or a 3458a is great but when the time comes to do a cal of this references, even DIY, a 3458a rental will still run $500/mo and a formal cal is as expensive.
There are metrology methods to increase the interval between cal by maintaining its intrinsic stability [e.g. never turning the reference off and keep it in an environmentally stable room] and collecting drift data to plot curves versus time [ years!]. Assuming your instrument is stable, without faults, and predictable, you then correct the instrument reading with this curve, but the more you allow the instrument to drift without a formal comparison the greater uncertainty grows in your values until you resort to a comparison against a standard.
Before doing this approach note the most difficult item is obtaining a bevy of stable quality references up to 1kV VAC and VDC, AC needs to have variable frequency and then up to 10A sources for current. In the end, what I did was obtain a used $100 DMM calibrator as a reference source, but was not spec'd to calibrate a 3456a. However, in short term, that is 1 hour at minimum, its stability is superb, ~ 0 ppm and adequate for comparing against a calibrated meter.
A simple option is only calibrating ranges you need or have references available, say < 100Vdc and ohms range, and thus not worry about AC sources or values > 100V.
-
Go speak to the Electronics Department (or maybe Electrical Engineering) at your local university. Providing you do the work outside of classroom time most of them are willing to help providing the equipment being calibrated is privately owned.
-
Go speak to the Electronics Department (or maybe Electrical Engineering) at your local university. Providing you do the work outside of classroom time most of them are willing to help providing the equipment being calibrated is privately owned.
Perhaps German universities are more flexible. I tried doing this at Cambridge (where I'm a postdoc) and they have a separate department for calibration and require official billing numbers. When I asked about just doing it privately (and paying for it) this caused much consternation as if I was proposing to bribe them to undertake some sort of criminal act!
-
Sorry but I think I have been misunderstood. I was not talking about going to the calibration department but to the laboratories themselves to use the equipment that has already been calibrated. When classes are not taking place all that expensive equipment is sitting idle.
-
Sorry but I think I have been misunderstood. I was not talking about going to the calibration department but to the laboratories themselves to use the equipment that has already been calibrated. When classes are not taking place all that expensive equipment is sitting idle.
If you are a student at the university this would probably work in Sweden. If you are a company and try to offer the technician money you could face charges.
-
I wouldn't trust universities too much, knowing this because I work at one where engineering is their specialty. I would trust comparison to my own devices before anything in their labs. The items here rarely have their calibration checked, and they are blown up by students constantly. I believe the only voltage standard they have is some 200 pound Fluke thing from the '70's that probably hasn't been touched since it was new.
-Mark-
-
I wouldn't trust universities too much, knowing this because I work at one where engineering is their specialty. I would trust comparison to my own devices before anything in their labs. The items here rarely have their calibration checked, and they are blown up by students constantly. I believe the only voltage standard they have is some 200 pound Fluke thing from the '70's that probably hasn't been touched since it was new.
Eehhhummm... how would I phrase my self... Spot on :-DD
We did buy a bunch of 34461`s a year ago, from those I set an old 341 that I found (last cal in -87).