|
Vintage Test Gear and Workshop Equipment For discussions about vintage test gear and workshop equipment such as coil winders. |
|
Thread Tools |
7th Aug 2017, 8:26 pm | #81 |
Dekatron
Join Date: Jul 2007
Location: Evesham, Worcestershire, UK.
Posts: 4,241
|
Re: Cheap multimeters
Cheers
I've been reading similar comments on EEVblog - it seems they've sold rather a lot (fuelled by Dave's video, no doubt). Let's see which of us gets their meter first - hopefully you will, as I've only been waiting for 2 weeks |
7th Aug 2017, 10:55 pm | #82 |
Rest in Peace
Join Date: Jun 2006
Location: Chard, South Somerset, UK.
Posts: 7,457
|
Re: Cheap multimeters
On the strength of the remarks in this thread, I bought an AN8008: arrived yesterday. Haven't checked it out fully yet, but it has passed all initial 'go / no go' tests. If you hear nothing more about it from me, you may safely assume that it checks out perfectly / meets its published spec.
Al. |
8th Aug 2017, 3:59 pm | #83 |
Hexode
Join Date: Dec 2005
Location: Nr. York, North Yorkshire, UK.
Posts: 348
|
Re: Cheap multimeters
Hello Mark
Congratulations on your multimeter reviews (post 79 for link) that's a lot of work there to get all that written up. You ask for comments, would it be possible, perhaps in the intro, to explain the usage of "count" in relation to meters and the significance of "true RMS". This is a useful resource for newcomers so perhaps defining terms that experianced users are used to would be useful. That's all I can think of. Will |
8th Aug 2017, 4:46 pm | #84 |
Hexode
Join Date: Aug 2010
Location: nr. Hannover, Germany
Posts: 372
|
Re: Cheap multimeters
After reading Mark's recent post I ordered an AN8008 which surprisingly arrived within about 2-3 weeks. So far I have been more than happy with it and yes, I do know when or where not to use it.
Thank you Mark for taking the time to publish your findings.
__________________
Eddie BVWS Member. Friend of the BVWTM |
8th Aug 2017, 8:16 pm | #85 | |
Dekatron
Join Date: Jul 2007
Location: Evesham, Worcestershire, UK.
Posts: 4,241
|
Re: Cheap multimeters
Quote:
Thanks for the encouragement, and for the suggestions. I started putting down some thoughts about this in Word, and already I'm up to 4 pages! On the face of it, true-RMS ought to be easy. And there are already lots of short explanations of what it is. But as you're asking the question, I'm assuming that you've already seen some of those, yet still aren't clear about what it all means? That might not be the case, of course, but in terms of writing something that will be of use - given what's already out there - that's where I feel I'd need to come from. To begin, I've explained what RMS actually is. I don't feel that's wasted words, given that I have to explain to electronics graduates what RMS actually is (invariably they can tell me the maths, but not what it really means). The question of count is much easier to explain. Are there any other points that need explanation though? How about "burden voltage" or resolution vs accuracy? Or even the basics of measurements (voltage across, current in series)? If I'm to write a decent primer, I need to see through the eyes of a beginner. While my day job requires me to be able to do that, I'd rather ask than make assumptions (which are often wrong!) Thanks in advance, Mark |
|
8th Aug 2017, 9:41 pm | #86 |
Nonode
Join Date: Dec 2012
Location: Kirk Michael, Isle of Man
Posts: 2,346
|
Re: Cheap multimeters
Mark, voltage across and current through. I recently gave a talk about motor cycle dynamos to our local VMCC group. One of our members said "tell us a bit about basics first" when he heard about it, so I started off at the beginning. I get the impression that many people talk about voltage and current without a clue what they mean. I used the old "water and pressure" comparison, but did it my way.
I did a dummy run the day before at home, and if anybody wants to criticise, here it is:- https://vimeo.com/226505541 Les. |
8th Aug 2017, 11:04 pm | #87 |
Nonode
Join Date: Oct 2014
Location: West Midlands, UK.
Posts: 2,181
|
Re: Cheap multimeters
Meters - first- these days, I'd doubt for our purposes that most digital meters are anything but accurate. I've got a few -two Flukes ,one Rapid Test and a cheapo I acquired lots of years ago at work. Few years ago I had access to meters calibrated to national standards, and I checked mine against this one on a few ranges. All of mine were within 1% of the works approved meter.
But then ,too many folks look at circuit sheets and go all scientific. It's dead easy to get led adrift on this. Sheet voltage is 220 v DC, meter reads 215/225. IMHO, from experience of a lot of faulting courses, best place to look is not at anode( valve circuit)/ collector ( transistor) ,but across cathode /emitter resistor. Voltage drop close to that on sheet- move on to next stage. Les- haven't looked at your bit,but to me, when I'm trying to explain ,using water. I treat voltage as pressure of water, and current as volume of water, with resistance as a kink in the pipe. |
9th Aug 2017, 12:24 am | #88 | |
Dekatron
Join Date: Apr 2012
Location: Worksop, Nottinghamshire, UK.
Posts: 5,549
|
Re: Cheap multimeters
Quote:
True RMS is a lot more complicated. It can be demonstrated on a scope. Display the waveform from a full wave rectifier without a smoothing capacitor on the scope with the dips on one of the lines and count the squares between the waveform and the line you set as your base. Then put all the squares and bits of squares together again to fill a space between the base line and another flat line. You now have the RMS voltage. Cheap meters are preset to estimate RMS for a sine wave. True RMS meters convert the voltage into a digital signal like a sound card does for audio with thousands of slices and calculates the average of all of them to get true RMS for other waveforms as well. |
|
9th Aug 2017, 12:03 pm | #89 |
Dekatron
Join Date: Jul 2007
Location: Evesham, Worcestershire, UK.
Posts: 4,241
|
Re: Cheap multimeters
No, a 6000-count multimeter should - in theory - go from 0 to 6000. Of course, being a general-purpose instrument, it goes from -6000 to 6000 - but it is possible to buy panel meters, etc, that don't go negative.
In practice, it's a bit less precise than that. The ubiquitous 2000-count meter doesn't reach 2000 at all - instead it goes to 1999. To reach 2000, the first digit would need to be a full seven-segment rather than just the "1", and that adds to the cost for something that would hardly ever be on. So 1999 is far more sensible, but because it's a bit of a mouthful to say "1999-count", it's almost always expressed as simply "2000-count". The argument goes that if you include 0, then that's 2000 in total. Once upon a time, we said "3.5 digits" rather than 2000-count. But then 4000-count meters appeared, which read up to 3999. For whatever reason, those were described as "3.75 digits" (don't shoot the messenger ) But along came 6000-count - what should those be called? So that's why we've moved away from the "N digits" scheme, in favour of counts. In the specific case of the AN8002 and AN860B+, it is sold as a 6000-count meter, and says as much on the front panel. But when you actually test them, they are 6200-count. And when measuring frequency or capacitance, then they are 9999-count. So really, the "headline" 6000-count specification is for marketing purposes. The Fluke 87V and Brymen BM235 goes to 6600, though the Fluke 101 really is 6000. The AN8008 is described as 9999-count, and as the display only has 4 digits, it could only ever read 9999. In this case, you have to be more precise as rounding up to 10,000 implies there is a 5 digit - albeit just "half" of one. Which again would be pointless, as you'd almost never see it used - you might as well go to 20,000-counts (19,999). So, it varies from meter to meter, and the spec needs interpretation and validation. But at least you know that a 6000-count meter can measure 5.000V, whereas a 2000-count meter can only measure 5.00V. For electronics work, where a lot of it is in the 5V region, the extra digit of resolution is very welcome indeed. Regarding RMS converters, the overwhelming majority of meters continue to use analogue converter ICs - the classic example being the AD536. Sampling the waveform to calculate RMS in the digital domain is very rare indeed. The cheap meters I've reviewed actually do do that - it's part of the DTM0660 or DM1106EN chipset that they use, and it's the only way they've been able to make true-RMS feasible at this price point - fabricating good analogue true-RMS ICs is complicated, so the ICs remain expensive. But the downside is bandwidth - good analogue RMS converters go to 100kHz or more, whereas the "cost-optimised" digital solution in the cheap chipsets has a -3dB point of 3kHz. Less of an issue for the high-end instruments (and digital oscilloscopes). In a nutshell, the non-RMS meter finds the average of the rectified incoming waveform (for a sine wave, it'll be ~64% of the peak value). Then it multiplies that by 13% so that the meter reads ~71% of the peak value. For a sine wave, that gives good results. But for a square wave, the RMS value is the same as the peak value, so how does the "average detecting, RMS reading" converter cope with that? It notionally adds 13% to the peak value, in practice over-reading by 5-13%, depending on the implementation. Obviously a square wave is an extreme example, but the point remains that anything other than a good sine wave will be reported inaccurately. An RMS converter will square the signal, then find the average of that, then find the root of that value. In other words, it'll do the maths that academics like to have their students do - the maths that tells you that for a sine wave, the magic number is root-2, and for a triangle wave, the magic number is root-3. So it will always be much more accurate, within certain limits, naturally. The best RMS converters - as used in instruments like the Fluke 8920A - use a thermal converter. Remembering that RMS is about equivalent DC heating effect, let's see how hot this signal makes something, then see how much DC current would get it to the same temperature. Implemented correctly, these can measure very high frequencies indeed, and cope with a wider range of waveforms than a good RMS converter IC can. Last edited by mhennessy; 9th Aug 2017 at 12:17 pm. |
9th Aug 2017, 12:24 pm | #90 | |
Dekatron
Join Date: Feb 2007
Location: Leominster, Herefordshire, UK.
Posts: 16,528
|
Re: Cheap multimeters
Quote:
Current is rate of flow, volume per second, firkins per fortnight etc., not just volume. The attached cartoon is quite informative as well as amusing
__________________
....__________ ....|____||__|__\_____ .=.| _---\__|__|_---_|. .........O..Chris....O |
|
9th Aug 2017, 12:57 pm | #91 | ||
Dekatron
Join Date: Jul 2007
Location: Evesham, Worcestershire, UK.
Posts: 4,241
|
Re: Cheap multimeters
Quote:
I didn't watch it all, but well done on a good presentation! Quote:
Knowing what your test gear can and can't do, however, is important. It's nice to be confident in the accuracy of your measurements, but it's also good to know the traps. For example, it's very easy to get led down a rabbit hole if you forget that your DMM only has a 1kHz AC bandwidth, or if you don't realise that the frequency function requires the signal to cross 0V when you're trying to measure a TTL signal. Those are the sorts of issues I tried to capture in my reviews - thinking about it now, I suppose with hindsight that might seem overly critical, but I was just hoping to help people avoid these traps. Sadly, there's no such thing as a perfect DMM, at any price, from any brand |
||
13th Aug 2017, 10:40 pm | #92 |
Dekatron
Join Date: Jun 2016
Location: London, UK.
Posts: 3,687
|
Re: Cheap multimeters
Did some tests on what I have on the bench and in the cupboard after seeing your post on AC bandwidth. This was done from a calibrated AWG (Agilent 33120A). AC bandwidths of what I have lying around conveniently (3dB point). I chose the AC bandwidth as a measure as it's the most variable specification:
1. Fluke 8050A - 267KHz (4.5 digit. True RMS. quoted 50KHz! £25) 2. Fluke 8600A - 90KHz (4.5 digit. Average responding. quoted 10KHz - actually reasonable to 567KHz but there is a big dip after 90KHz £25) 3. Fluke 8024B - 30.5KHz (3.5 digit. Average responding. quoted 5KHz £8) 4. Fluke 8021B - 29.8KHz (3.5 digit. Average responding. quoted 5KHz £11) 5. DT830 - 2.2KHz (3.5 digit. Average responding. unquoted - total rubbish £2.60) 6. UT61E - 25KHz (4.5 digit. True RMS. quoted 10KHz. goes mental after 25k, reading weird values £38). 7. HP427A - 3.9MHZ (Analogue. Average responding. quoted 4MHz £40) 8. AN8008 - Still MIA 6. DT830 - 2.3KHz (different to the above one. Average responding. Unquoted - total rubbish. So good it came free with something) 7. Fluke 8010A - 105KHz (3.5 digit. True RMS. quoted at 50KHz. £18) 8. HP 3478A - 580KHz (5.5 digit. True RMS. quoted at 300KHz. £78) Best meters: HP3478A, Fluke 8050A, Fluke 8010A, HP427A. And thus you get what you pay for. In my case way too many multimeters and a wife who said "so are you going to sell any of them?" which might happen as now I know this I rather just want to keep a couple of the true RMS Flukes and the 3478A. Edit: I've added paid price so you can see value for money. Last edited by MrBungle; 13th Aug 2017 at 10:45 pm. |
13th Aug 2017, 11:45 pm | #93 |
Dekatron
Join Date: Jul 2007
Location: Evesham, Worcestershire, UK.
Posts: 4,241
|
Re: Cheap multimeters
Fluke's specifications are very conservative
That said, -3dB points are rarely quoted on meters because that's so far out of the sorts of spec people are used to seeing. For example, the 8050A is 5%+30 in the 20k-50kHz region. 5% is about 0.4dB. That might be enough to worry about. But then, I'd always make that sort of measurement with a 'scope, perhaps using a DMM if I need extra precision. A lot of people today seem to expect their DMMs to do things that should really be done on a 'scope or some other instrument, but a DMM won't tell you if something is clipping or hooting. I suppose that's the danger of cramming all these functions onto a DMM... The AN8008 will be 3kHz. And because that's done digitally, it'll be boringly consistently 3kHz. In my experience, that's a sure sign of the DMT0660 chipset hiding under the epoxy! I'd say you need to add a Fluke 8920A to that collection. Thermal RMS conversion, good for at least 20MHz, and a bit of a bargain when they crop up because most people don't realise what it is. My UT61E has arrived. Haven't had time to test it beyond checking it survived the journey. |
14th Aug 2017, 12:20 am | #94 |
Dekatron
Join Date: Jun 2016
Location: London, UK.
Posts: 3,687
|
Re: Cheap multimeters
Very conservative indeed. I'm quite impressed.
I will test the 8050 across the full range and see where it goes. I think some of the measurements I had could be improved with a better test rig setup. I rather like the 8050A though generally. Very useful meter. It's handy due to the relative dB measurements for filter rolloff etc. Scope isn't quite as good at accurate measurements. I will definitely add an 8920A to the collection if one makes an appearance. 20MHz would be very useful for HF stuff. |
22nd Aug 2017, 4:25 pm | #95 |
Dekatron
Join Date: Jul 2007
Location: Evesham, Worcestershire, UK.
Posts: 4,241
|
Re: Cheap multimeters
Hi again,
The first draft of my review of the UT61E is up: http://www.markhennessy.co.uk/budget...unit_ut61e.htm First impressions are pretty favourable, apart from the slow update speed. The PC interface is fun. My generic USB to RS-232 adaptor works fine (Prolific chipset). My example looks to be better built that earlier examples seen on-line, so that's good. It's not the German model with the MOVs and larger fuses though - not surprisingly at that price. I'm close to finishing the article about true RMS (in answer to an earlier question). I'm also working on an article about AC+DC bugs, which is something a lot of meters suffer from, but not everyone is aware of. These sorts of things are surprisingly time-consuming, so bear with me for now... |
22nd Aug 2017, 4:32 pm | #96 |
Octode
Join Date: Dec 2010
Location: St Ives, Cambridgeshire, UK.
Posts: 1,180
|
Re: Cheap multimeters
Mark,
Did you see Joe Smith posted a video of the 8008 in the last couple of days? Andrew
__________________
Invisible airwaves crackle with life. Or they should do. BVWS Member |
22nd Aug 2017, 4:37 pm | #97 |
Dekatron
Join Date: Jul 2007
Location: Evesham, Worcestershire, UK.
Posts: 4,241
|
Re: Cheap multimeters
Yes, indeed I did
|
30th Aug 2017, 8:35 pm | #98 |
Dekatron
Join Date: Sep 2005
Location: Seaford, East Sussex, UK.
Posts: 5,997
|
Re: Cheap multimeters
I just compared my uncalibrated water damaged and builders' dust ridden Fluke 79 mkII I rescued from the skip and repaired with a brand new Fluke 3000. A battery measures exactly 1.067 volts on both instruments...
|
30th Aug 2017, 8:47 pm | #99 |
Dekatron
Join Date: Jul 2007
Location: Evesham, Worcestershire, UK.
Posts: 4,241
|
Re: Cheap multimeters
I'm not in the least bit surprised. Mine isn't too shabby either. I've never touched the insides in all the time I've had it: http://www.vintage-radio.net/forum/s...d.php?t=124491
Most Fluke multimeters benefit from expensive thick-film divider networks that have been engineered for stability and electrical ruggedness. The resistors used in the cheap meters won't be in that class, but it'll be really interesting to see how these meters hold their calibration over the years. |
28th Sep 2017, 11:07 am | #100 |
Heptode
Join Date: Jul 2014
Location: Boston, Lincolnshire, UK.
Posts: 993
|
Re: Cheap multimeters
Just in case anyone is still looking for an AN8002 series meter - Banggood.com have the AN8002 on offer at the moment (£10.21 incl. P&P).
https://www.banggood.com/ANENG-AN800...l?rmmds=search |