I'm working on a DIY project that requires accurate AC and mV AC measurements. In 2016, I sent my HP 3400A and another HP400 series meter for official calibration to a Metrology lab in the North East. Got the full reports, cal stickers and Void if removed stickers over screws. Wonderful. My handheld DMM and 4 1/2 digit bench top meters do not match each other and do not closely match the HP meters. Yes, all have been on for hours, etc.
So I'm debating where to send the digital meters. The Metrology lab said send them and they will most likely be able to calibrate them. However, I hate to waste shipping dollars if they can't. My meters are BK Precision 2831E and Extech MP530 (a Fluke 87 knockoff). BK wants quite a bit of money to cal and then return. I'm not getting a NIST cal, just a regular cal with documentation. Still pricey. I called Extech, which is now part of FLIR and they are not sure they handle the old product lines. I asked them to provide the calibration documentation so I can forward it to the Metrology lab. I am waiting for tier 2 support to respond to that request. I don't have good hopes on them releasing the info.
Any opinions about OEM vs Metrology lab?
Before anyone craps on the Extech (fluke knockoff), I refuse to buy Fluke products. In my industry, they have totally ruined the test equipment products by buying up other vendors, end of life their products, and then force the new over-priced & unreliable crap down our throats. It was so bad, a $20K analyzer kept loosing cal settings and was returned 3 times in a 12 month period. They refused to send a new unit in exchange for something that was in for service more than in my hands. When we spend $100K on test equipment, I now support the other vendors as much as possible. And their products are as good or better than Fluke. OK, off my soap box.
So I'm debating where to send the digital meters. The Metrology lab said send them and they will most likely be able to calibrate them. However, I hate to waste shipping dollars if they can't. My meters are BK Precision 2831E and Extech MP530 (a Fluke 87 knockoff). BK wants quite a bit of money to cal and then return. I'm not getting a NIST cal, just a regular cal with documentation. Still pricey. I called Extech, which is now part of FLIR and they are not sure they handle the old product lines. I asked them to provide the calibration documentation so I can forward it to the Metrology lab. I am waiting for tier 2 support to respond to that request. I don't have good hopes on them releasing the info.
Any opinions about OEM vs Metrology lab?
Before anyone craps on the Extech (fluke knockoff), I refuse to buy Fluke products. In my industry, they have totally ruined the test equipment products by buying up other vendors, end of life their products, and then force the new over-priced & unreliable crap down our throats. It was so bad, a $20K analyzer kept loosing cal settings and was returned 3 times in a 12 month period. They refused to send a new unit in exchange for something that was in for service more than in my hands. When we spend $100K on test equipment, I now support the other vendors as much as possible. And their products are as good or better than Fluke. OK, off my soap box.
If you are looking at accurate AC calibration you will find that's very challenging. I'm almost finished setting up my cal system and the number of places where under the best of conditions accurate devices do not say the same can be distressing.
There are a handful of vendors with good instruments and a lot of lesser products. What are you measuring (frequency & amplitude)? It will help suggest suitable products and resources.
I understand the issues with Fluke although the vintage Fluke stuff is still quite good. HP/Agilent/Keysight is doing a lot to drive customers away as well.
Let me clear up a fundamental common misunderstanding about "calibration". It does not mean "adjust", it means "checking" and in this verifying if the device meets its specifications. You need to read the specs carefully to see what they mean in the context of what you are measuring. comparing at 1 KHz pure sine wave may get different results from a 50 KHz complex signal.
One place to get a lot of input on these issues is https://www.eevblog.com/forum/testgear/multimeter-spreadsheet/
The larger independent cal services may be helpful both in learning what you can deal with and which products will be supported in the future. Factory service must be understood in the context that every new sale is an automatic new customer for the service. (Think about that the next time you take a car to the dealer and get lousy service).
There are a handful of vendors with good instruments and a lot of lesser products. What are you measuring (frequency & amplitude)? It will help suggest suitable products and resources.
I understand the issues with Fluke although the vintage Fluke stuff is still quite good. HP/Agilent/Keysight is doing a lot to drive customers away as well.
Let me clear up a fundamental common misunderstanding about "calibration". It does not mean "adjust", it means "checking" and in this verifying if the device meets its specifications. You need to read the specs carefully to see what they mean in the context of what you are measuring. comparing at 1 KHz pure sine wave may get different results from a 50 KHz complex signal.
One place to get a lot of input on these issues is https://www.eevblog.com/forum/testgear/multimeter-spreadsheet/
The larger independent cal services may be helpful both in learning what you can deal with and which products will be supported in the future. Factory service must be understood in the context that every new sale is an automatic new customer for the service. (Think about that the next time you take a car to the dealer and get lousy service).
Looking at the specs of the B&K and the Extech, for AC the B&K has specs, the Extech not really. Extech says response to 20 KHz but no accuracy indication. The B&K has more detail- .4% midband and broader at the band-ends. The HP 3400 (good instrument) is 1% midband. Its much better suited to measuring wideband low level AC than the DVM's above.
I'm measuring measuring relatively simple stuff. AC Volts 0-25 range @ 1K, AC mV @ 1K and DC Volts and mV. It is just frustrating knowing one meter shows. 17.25 VAC and the other is 17.31 VAC. DC off similarly. So which is correct and which is incorrect. It would be fine for go / no go testing which is what I generally use the hand held DMM for...like most. But here I needed a second meter...which led me to this rabbit hole of nothing matches exactly.
I want calibration. Calibrate the meter so it is correct and accurate. I do not want to get into sources and I can't afford a calibrator. That's why I'll leave it to the OEM or Metrology lab.
My big concern is the Metrology lab will call and say we don't know how to get this into XYZ mode for adjustments and the vendor will not provide the information. I know we are not just tweaking pots any longer and it is most likely a software menu to adjust 'things'. I'm not trying to be cheap, just would be nice to have everything sent to one location for a reasonable price. Especially since this is just DIY
Last edited:
Just check what you have with this: https://www.aliexpress.com/item/329....store_pc_allProduct.8148356.1.5a4f1827ShJurG
I also have a big Fluke Calibrator, and this $24 thing from China is darn close.
I also have a big Fluke Calibrator, and this $24 thing from China is darn close.
Thanks. Which one do you own? I see a few choices. I thought about making a precision reference and I saw more expensive other devices. But for a quick sanity check, this seems like a decent deal.
That's 0.3%. I don't think any affordable instrumentation, or servicing, can hold 0.3%?
BTW, that is 30 milli dB.
I have the LM399 based one. I had a cal lab here measure it and they said it was pretty close. Nothing is exact unless you are NIST.
Also, one should let this calibrator and the meter warm up for an hour or so before making any moves.
Getting down in the weeds with NIST on voltage cal is kind of interesting. Check out this paper: https://tsapps.nist.gov/publication/get_pdf.cfm?pub_id=23594
They say here they can typically calibrate a reference cell to <0.1 ppm or <0.00001%. That's as good as it gets.
.1 ppm is DC. AC is much looser at .01% or 10 PPM. Maybe 5 PPM for a very short time midband. There is an effort to build the ultimate DAC using JJ's. Its going slowly since its a really complex problem to solve https://tsapps.nist.gov/publication/get_pdf.cfm?pub_id=31547 However it would absolutely trounce any existing DAC for complexity, price and cost of operation.
Well, I eat my words and I feel like I should go purchase some lottery tickets based on some good news. Flir (which now owns the Extech brand) sent me the calibration instructions & tolerance thresholds for my meter. I sent a very nice response to the support advisor I have been communicating with at Flir.
Back when I worked for Rockwell collins about 15 years ago I had a Fluke 189 and was measuring an AC voltage
in a circuit and it measured somthing like 177mv. This just seemed wrong so I borrowed someones Fluke 287 and
it measured about 377mv. This also seemed wrong. So I sent them both of to metrology (calibration). They both
met specification.
When you are measuring non sinusoidal voltages with a digital meter, good luck!
in a circuit and it measured somthing like 177mv. This just seemed wrong so I borrowed someones Fluke 287 and
it measured about 377mv. This also seemed wrong. So I sent them both of to metrology (calibration). They both
met specification.
When you are measuring non sinusoidal voltages with a digital meter, good luck!
I'm measuring a 1K Hz sine wave out of a power amp between 1 and 20 volts; not distorted.
Same is true of analog meters. For waveforms that are complex you can get very different readings and none may be what you actually need to know. Thats when a scope makes a big difference. A recent acquisition of mine is the little battery powered scope (Sigpeak) DSO1511 (around $80) that has readouts for both RMS and peak to peak. But more you can see the actual waveform to know what you are dealing with. A quick check suggests better than 3% to 1 MHz. when I need more accuracy the options are quite limited. I have both Fluke 8506A and Fluke 931B which honestly are serious overkill but in the inventory. However with a nonsinusoidal waveform it matters a lot less.
As 1audio already wrote, a calibration is nothing more than a check against factory specs. You'll get back the meter unopened, with a pass/fail certificate. You need to supply the factory specifications along with the instrument.I want calibration. Calibrate the meter so it is correct and accurate. I do not want to get into sources and I can't afford a calibrator. That's why I'll leave it to the OEM or Metrology lab.
My big concern is the Metrology lab will call and say we don't know how to get this into XYZ mode for adjustments and the vendor will not provide the information. I know we are not just tweaking pots any longer and it is most likely a software menu to adjust 'things'. I'm not trying to be cheap, just would be nice to have everything sent to one location for a reasonable price. Especially since this is just DIY
Adjusting a meter to factory specs is a service task. If it goes out of specs sometimes the cause is a faulty component.
Referring to different DMMs showing different AC for non-sinusoidal waveforms: use a True RMS multimeter.