13th Feb 2016, 12:10 am | #41 |
Moderator
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,894
|
Re: A precision voltage calibrator
Ah, but David, we're all doing this for fun. We're playing. Sometimes we're playing with a nice radio, sometimes it's nice test equipment. I have a couple of 6-digit DVMs and as they agree with each other, I'm inclined to trust them. But I did try them out against one at work.
I don't need this precision, but part of the playing is to get them going as well as they can be. It all hinges on why Martin wants it calibrated. If it's a business thing, then, as Jeremy said, he has to do it all the normal way so that his customers have a traceable path. If it's not, it still might be something he wants to do anyway. David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done |
13th Feb 2016, 12:29 am | #42 |
Dekatron
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 3,077
|
Re: A precision voltage calibrator
I still have an old Altai analogue meter here that I bought as a student back in the early 1980s. Prior to that I had one of the fairly hideous little 1k/V pocket meters. For a bit of fun I tried checking it against my Fluke 45 on a PSU across various voltages up to 30V.
This Altai meter has not been fiddled with at all in all of its 30-35 year life and it is still remarkably accurate. The limitation of the accuracy is dominated by the set/reset accuracy of the cheap meter movement but it could indicate 10V well within about 0.1V even allowing for the cheap movement. It is limited a bit by the 20k/V loading but I was surprised how accurate it still is on the DC volts ranges after all these years. I doubt it has any kind of calibration adjustment and I'm also impressed how accurate the mechanical meter movement has remained over all these years.
__________________
Regards, Jeremy G0HZU |
13th Feb 2016, 12:00 pm | #44 | |
Nonode
Join Date: Mar 2008
Location: Aberdeen, UK.
Posts: 2,856
|
Re: A precision voltage calibrator
Quote:
Regards, David |
|
15th Feb 2016, 5:24 pm | #45 |
Nonode
Join Date: Dec 2012
Location: Stockport, Greater Manchester, UK.
Posts: 2,113
|
Re: A precision voltage calibrator
I've only now seen this thread, but about the same time as it started I was repairing a lab PSU and I wanted to finish the job to my complete satisfaction by calibrating the meter for V & I and I ran into the dilemma of which DVM could best be trusted.
Then one of those eBay spooky coincidences happened 'you may also like...' and it presented me with a "4.096V DC precision voltage reference for arduino calibration by Fluke 8846A" - £3.99 from China, so I ordered one which is supposed to arrive this week. This looks ample to give me a calibration reference which will meet my target for a Working Standard of voltage. I am hoping to be within ±0.1%, or 10mV at 10.00V. I can't relate this product to the voltage references discussed above, nor do I understand where the Arduino or Fluke 8846A comes into the picture. The seller says that it uses a TL-431 from Texas Instruments. If we disregard the least significant figure I would still be happy with 4.10V Thanks to Jerry for the links re EPE Magazine. I am hoping these will allow me to multiply and divide the reference to give two or more additional useful voltages |
15th Feb 2016, 5:44 pm | #46 |
Nonode
Join Date: Mar 2008
Location: Virginia Water, Surrey, UK.
Posts: 2,877
|
Re: A precision voltage calibrator
The "Arduino" reference is because some of the Arduino microcontroller boards have a DAC (digital to analogue converter) that can give out nice voltage steps if presented with an accurate 4.096V (2^10 x 1mV) reference. They have "calibrated" (eg not calibrated but checked) it against a reasonable 6.5 digit dmm, the Fluke 8846A (and is that calibrated?). Since this reference is adjusted by them you can hope that it's pretty close to 4.096.
The other references mentioned earlier are NOT adjustable but have an inherent accuracy (as per the data sheet), so have fairly predictable performance (eg temperature, ageing and absolute errors) as references. The TIL-431 is just a shunt regulator not a reference, that the Chinese have trimmed to 4.096V, presumably via an on-board pot) Horses for courses, and probably more than adequate for what you need to do with it. Jeremy
__________________
Jeremy, G8MLK, BVWTVM Friend, VMARS, BVWS Secretary. www.pamphonic.co.uk www.bttt.org.uk |
15th Feb 2016, 6:41 pm | #47 |
Nonode
Join Date: Dec 2012
Location: Stockport, Greater Manchester, UK.
Posts: 2,113
|
Re: A precision voltage calibrator
Thanks, that figures. The pictures show a small PCB with what looks like a Bourns 3296 trimpot. I'll ask my former colleagues to measure it on their best calibrated DVM
|
15th Feb 2016, 6:54 pm | #48 |
Dekatron
Join Date: Feb 2003
Location: Cottingham, East Yorkshire, UK.
Posts: 5,763
|
Re: A precision voltage calibrator
There are several e-bay offerings at £10 - £15 post free from China of the ‘KKMOON High Precision Voltage Reference Module AD584kH’ with selectable outputs of 2.5, 5, 7.5 and 10V.
They’re powered from a built-in rechargeable lithium battery and can be charged from a USB port. Though the description states ‘highest accuracy’ the actual accuracy isn’t specified, so is questionable and subjective. That said, there are handwritten ‘calibration’ figures such as 5.00178, 7.50004, though no indication as to the calibration standard. More than adequate for hobbyist needs though. I can't actually think of any vintage electronics application which calls for great precision anyway, but I can understand the interest and curiosity. (I did after all, build the EPE 10V reference myself, which only served to confirm that all of the several DMMs and VVMs that I possess are pretty much spot on). The stated features are: Built-in AD584KH, highest accuracy and stability. Same port can output reference voltage: 2.5v, 7.5v, 5v, 10v. The MCU switch 4-way reference voltage output with LED indicators. Single-button operation switch reference voltage output. Built-in lithium battery, easy to carry, can be charge by DC 5V, with charge indicator. 4mm banana jack as output interface. Note: Do not charge with more than 5.5 volts DC. Please charge with a USB adapter. Specifications: Reference Voltages: 2.5V/7.5V/5V/10V Power Supply: Built-in Lithium Battery(can be charged by DC 5V) Size: 70 * 52 * 35mm / 2.76 * 2.04 * 1.38in Weight: 65g / 2.31oz
__________________
David. BVWS Member. G-QRP Club member 1339. |
17th Feb 2016, 3:05 pm | #49 |
Dekatron
Join Date: Apr 2005
Location: Bewdley, Worcestershire, UK.
Posts: 4,748
|
Re: A precision voltage calibrator
I have a cheap and tiny TS6001 chip on a bit of Veroboard, powered by a 3 volt Lithium coin cell. It produces a 2.5 volt reference potential and delivers up to 500uA. More than accurate enough to check the calibration of most analogue and digital meters. Given that the DC accuracy of the uniquitous Avo 8 is only specified as 2% of FSD, I can usually get these to better than 0.2%. As has been said, for hobby usage on vintage electronics this is more than sufficient.
http://www.farnell.com/datasheets/1774446.pdf
__________________
Phil Optimist [n]: One who is not in possession of the full facts Last edited by Phil G4SPZ; 17th Feb 2016 at 3:07 pm. Reason: Link added |
17th Feb 2016, 8:48 pm | #50 |
Hexode
Join Date: Oct 2014
Location: Wiltshire, UK.
Posts: 381
|
Re: A precision voltage calibrator
I bought a non-working Solartron 7150, replaced several electrolytics and wondered how good the calibration was so I got one of the K K Moon voltage Standards'. The ratio of K K Moon to Solartron is fairly constant at 1.00070, who knows which is more accurate but at least I feel a lot happier with the Solartron. Feeling even more happy could cost a lot more! You pay your money ...
Roger |
29th Feb 2016, 12:21 pm | #51 | |
Rest in Peace
Join Date: Jun 2006
Location: Chard, South Somerset, UK.
Posts: 7,457
|
Re: A precision voltage calibrator
Update, Feb. 29, 2016.
I bought a precision voltage reference from a supplier in China (refer to post # 23). It was remarkably cheap - but upon test, it was dead. Eventually, I received a refund. Undeterred, I then purchased one of these: Quote:
Al. |
|
29th Feb 2016, 1:32 pm | #52 |
Nonode
Join Date: Dec 2012
Location: Stockport, Greater Manchester, UK.
Posts: 2,113
|
Re: A precision voltage calibrator
What are you doing to them, Al?
The 4.096V ref using the TI TL-431 (see post #45) at last arrived this morning (slightly over 3 weeks after my order) and does what it says on the tin so I shall make this my home reference. While waiting for this, I recalled that I have a small box of ceramic packaged precision reference ICs with their own tiny ovenised device, so I've been looking for them ever since Googling for the device finds that they are LM399, readily available on eBay I also came across an entertaining and valuable App Note by Mitchell Lee of Linear Understanding and Applying Voltage References, and another by Brendan Whelan How to Choose a Voltage Reference which I commend to those who want to pursue the topic further |
29th Feb 2016, 5:59 pm | #53 | |
Rest in Peace
Join Date: Jun 2006
Location: Chard, South Somerset, UK.
Posts: 7,457
|
Re: A precision voltage calibrator
Quote:
Where appropriate, I am testing them as per the supplier's instructions - plus measuring the output voltages, in turn, with a Fluke 73, a Fluke 75, a Fluke 77, an AVO 8 Mk. V, an AVO 8 Mk. VI and a home-brew analogue multimeter (which was calibrated against a professional standard some few years ago). Now there may be minor calibration errors in those instruments, but when the mean reading of them is calculated, (thus minimizing the error), all the output voltages from these 'precision voltage reference' pcbs are substantially different from that calculated value. And I said above, one of them was D.O.A: no O/P voltages at all. Al. |
|
29th Feb 2016, 6:05 pm | #54 | |
Rest in Peace
Join Date: Jun 2006
Location: Chard, South Somerset, UK.
Posts: 7,457
|
Re: A precision voltage calibrator
Quote:
Hmm. Pass the milk, please: looks like a bowl of spikey breakfast cereal! Al. Last edited by Skywave; 29th Feb 2016 at 6:07 pm. Reason: Add image |
|
29th Feb 2016, 8:33 pm | #55 |
Heptode
Join Date: Mar 2009
Location: Portsmouth, Hampshire, UK.
Posts: 674
|
Re: A precision voltage calibrator
My AD584L arrived about a week ago, I was a bit suspicious when the 'calibration certificate' had printed results.
So far I've only performed some preliminary testing, but apart from initial voltage tolerances (which I cannot measure with sufficient accuracy to be sure of) the stability with temperature, load and line seem pretty good. Temperature testing was done using a hair dryer, the effect of thermal transients was surprisingly low. I should have access to a calibrated HP-33401A in a couple of weeks so I should be able to refine the results then. It's good enough what I want, and for under £4 I'm happy. Jim |
29th Feb 2016, 9:24 pm | #56 |
Moderator
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,894
|
Re: A precision voltage calibrator
I had a look at the ADI datasheet on their website to see what the spec is. There are a lot of versions on it, but I can't find an 'LH' suffix. The list is on page 12 of 12 on the AD584 datasheet.
David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done |
29th Feb 2016, 9:35 pm | #57 |
Octode
Join Date: May 2009
Location: Linkoping, Sweden
Posts: 1,465
|
Re: A precision voltage calibrator
The H-suffix is used for "Hermetic metal can", says so on page 2 in the A-rev of the datasheet. It is an AD584L in TO99 metal can.
__________________
Martin, Sweden |
29th Feb 2016, 9:37 pm | #58 |
Heptode
Join Date: Mar 2009
Location: Portsmouth, Hampshire, UK.
Posts: 674
|
Re: A precision voltage calibrator
The 'L' version of the AD584 was discontinued around 2012, it is in rev. B of the datasheet but was dropped from rev. C.
The date code on mine is 1015 (2010 week 15). Jim I see Martin posted while I was typing. Last edited by jimmc101; 29th Feb 2016 at 9:44 pm. Reason: ps added |
4th Mar 2016, 1:21 am | #59 | |
Rest in Peace
Join Date: Jun 2006
Location: Chard, South Somerset, UK.
Posts: 7,457
|
Re: A precision voltage calibrator
Extract from post #51, dated Feb. 29, q.v. . . .
I purchased one of these: Quote:
-------------------------------- Update, March 4. 2016. Fortunately, that matter is now resolved, but not without a struggle with the supplier. I eventually received a refund. So that's now two precision voltage reference units - from different suppliers - that I have here, both faulty. One is simply a pcb with an AD584 plus a few passive components; the other also has several SMT devices on its pcb, in addition to the AD584. Now on the basis that there may a reasonable chance that only one AD584 chip is defective and since I have the datasheet for the AD584, in the established time-honoured tradition, I might be able to make a good one out of the two. After all, what have got to lose? A few hours of my time at worst. I'll report back later. Al. |
|
4th Mar 2016, 8:06 am | #60 |
Octode
Join Date: May 2009
Location: Linkoping, Sweden
Posts: 1,465
|
Re: A precision voltage calibrator
Al, maybe you can go and buy AD584's from a proper source and populate your boards with them. I sometimes buy things cheaply knowing that they will probably be fakes, or populated with fake ic's, but if the PCB is nice I simply replace the fake ic's and get a good PCB cheap. With fake I mean both real fakes and re-used chips.
__________________
Martin, Sweden |