Author Topic: EEVblog #184 - Open Hardware Multimeter Concept  (Read 7405 times)

0 Members and 1 Guest are viewing this topic.

Offline Fox

  • Regular Contributor
  • *
  • Posts: 54
  • Country: de
  • Electronics Engineer
EEVblog #184 - Open Hardware Multimeter Concept
« on: July 04, 2011, 04:22:45 am »
Hi Dave,

very interesting concept especialy the multichannel capability.
A few weeks ago i started designing my own Multimeter just for fun, so no multichannel here but powermeasurement, LCR capability and 5.5Digit Voltage and currentmeasurement.
I found designing a Multimeter is a very complex task depending on the accuracy one want to achieve e.g. going from 4.5Digit to5.5Digit is a very big step.
The hardest problem i encountered is the replacement of the mechanical rangeswitch with electronic switches eg. try to find an Analog switch which can isolate 1000V.
As for your Davecad drawing of the case have look at the Bopla BOS Streamline cases.
A closed Switch should have zero Ohms or less!
 

Offline ejeffrey

  • Super Contributor
  • ***
  • Posts: 1622
  • Country: us
Re: EEVblog #184 - Open Hardware Multimeter Concept
« Reply #1 on: July 04, 2011, 07:00:34 am »
I like the concept, except for the voice part.  I think that is a wank feature, but to each his own.  As long as I can turn it off I don't care.  I think some carefully designed audible alerts would be more useful.

A few more thoughts:  I am surprised that you balk at putting an opto-isolated USB input when you already have at least 4 isolation barriers.  Assuming volts-ohms-cap-diode is on the main supply, the three extra channels plus the thermocouples will need isolation, both of signal and power.  USB only needs signal unless you want to power/charge the unit via USB.  Another factor related to the isolators is power consumption.  The isolated sections will take more power, so it would be nice if "channel 1" could do all basic measurements with the isolators turned off.  This would mean an extra amps jack, or for channel 1 volts/amps to share a common terminal.

Obviously the SD card is great for logging.  It can be handy for firmware upgrades, but it can also be a bit fiddly -- it is an extra step, and it requires me to have an SD card and an SD reader handy.   

A feature that I would really like on meters is unlimited custom ranges.  A lot of bench meters come with a calibration for a PT100 or PT1000 that gives a direct readout of temperature, but what if I have a PT500 or a thermistor or an AD590?  Transforming a current/voltage/resistance to an arbitrary unit would be really useful -- this could also be handy for monitoring power when you have a voltage proportional to current, for instance if you have a hall sensor or an external shunt like the uCurrent adapter.  USB might be a handier way to load calibration curves on the meter.  For volts and current, 90% of the time you can get away with a linear + offset scale which could be entered via the onboard buttons, but a thermistor needs to have a curve loaded.

With regard to a range switch:  I am on the fence.  The standard range switch is a bit difficult on a 4/6 channel meter.  If you put one on, it will probably only select the operation of channel 1, or possibly include 'power' that uses primary volts and amps.  You might be able to have a soft range switch based on a touch sensor or a rotary encoder (think the original iPod with the mechanical wheel).  Then you could press the soft button associated with the channel you want to select and spin the wheel until it hits the right setting.  Another soft button to select manual ranging, and spin the wheel to select range, and another soft button to select custom defined scale associated with that measurement type (volts, ohms, amps, diode), and spin the wheel to select it.

A complicated measurement is going to be a bit crazy with the test leads.  In order to ease confusion and reduce the risk of error, I recommend thinking heavily about distinguishing the jacks.  Maybe take a cue from the oscilloscope manufacturers.  Have each signal channel (1-4) have their own color, and have the common terminal half black / half signal color.

I don't know how much is involved in the ohms/diode/capacitance/continuity measurement.  If it isn't too complicated, I would put all the secondary features on the second voltage channel, even if they are harder to get to in the menu system.  I can imagine that you have everything set up to measure power, and decide you want to monitor a resistor (maybe a thermistor) -- it would be nice if you could do that without changing all your probes around and resetting the meter to use channels 3 and 4 for power.
 

Online EEVblog

  • Administrator
  • *****
  • Posts: 26829
  • Country: au
    • EEVblog
Re: EEVblog #184 - Open Hardware Multimeter Concept
« Reply #2 on: July 04, 2011, 08:43:19 am »
I find USB interfaces on meters a bit silly. After all, meters only have very low sample rates, so processing power is minuscule, yet you tie the meter to a machine  that has a million times the power required? Doesn't make sense.
I prefer a meter that logs internally and then you upload the data later. No tethering to the PC required.
If you need to automatically control the meter from the PC (say an ATE system), then fair enough maybe. But it's pretty rare you need to do that, and usually you'd use nice mains powered rack mount gear for that, not a handheld meter.

Dave.
 

Offline ejeffrey

  • Super Contributor
  • ***
  • Posts: 1622
  • Country: us
Re: EEVblog #184 - Open Hardware Multimeter Concept
« Reply #3 on: July 04, 2011, 09:08:53 am »
I definitely don't want to tether the meter during use, I just generally prefer to log to internal storage, then connect via USB to offload the data.  Personal preference only, obviously.  The main factor is that I have lots and lots of USB equipment, and therefore lots of USB cables.  I don't have many SD card devices -- just a camera and a cell phone at home and a optical power meter at work. All of them also have USB so you never have to actually remove the card.  Therefore, I don't have extra cards lying around, and I don't have SD card readers.

Even better for me would be a USB host port that can upload to a USB flash drive, since I always have one of those on my keychain, but I don't know what the power consumption on those are.
 

Offline Kiriakos-GR

  • Super Contributor
  • ***
  • Banned!
  • Posts: 3526
  • Country: gr
  • User is banned.
    • Honda AX-1 rebuild
Re: EEVblog #184 - Open Hardware Multimeter Concept
« Reply #4 on: July 04, 2011, 10:04:21 am »
I had watched currently only two minutes of this video ( middle section), at the point where Dave speaks about the front inputs,
soon I will download it and watch it off line.

Currently two pictures came to my head from what I watched so far ... Test leads that looks like an octopus,
and one large Users manual, even larger from the 150 pages, of the Agilent U1272A.  :)   

I will say more when I will watch it all .
 

Online EEVblog

  • Administrator
  • *****
  • Posts: 26829
  • Country: au
    • EEVblog
Re: EEVblog #184 - Open Hardware Multimeter Concept
« Reply #5 on: July 04, 2011, 11:21:18 am »
Currently two pictures came to my head from what I watched so far ... Test leads that looks like an octopus,

Yeah, that could be a real problem  :o

Ah, details, details....

Dave.
 

Offline _Sin

  • Regular Contributor
  • *
  • Posts: 247
  • Country: gb
Re: EEVblog #184 - Open Hardware Multimeter Concept
« Reply #6 on: July 04, 2011, 08:06:10 pm »
I thought about swapping the input jacks about so the ground points were the four in the middle, allowing some combination of them to be shorted easily. However it might be even better to have a physical switch on 2, 3 and 4 which shorts them to input 1.

That might help reduce the spaghetti effect of so many probe connections.

I'm afraid I'm in the camp of liking an on-board USB connection, though I definitely like the idea that I could log directly to a memory card too. I see having USB as a good solution for not having enough on-board display capability, considering the number of possible inputs (especially with the multiple thermocouple jacks suggested).

Programmer with a soldering iron - fear me.
 

Offline Kiriakos-GR

  • Super Contributor
  • ***
  • Banned!
  • Posts: 3526
  • Country: gr
  • User is banned.
    • Honda AX-1 rebuild
Re: EEVblog #184 - Open Hardware Multimeter Concept
« Reply #7 on: July 05, 2011, 01:39:36 am »
Just watched the all thing.  :)

Nice ideas, and frankly about the voice recorder I had it in my mind before a week ago.
Many times the electricians removing the power cables from three phase motors,
and every single time you forget the way that it was connected ..
Yea, give a voice recorder on my DMM , so to just say  Blue-Red-Brown, this all that I need.  ;)   

Battery life, an truly important factor,
on the Fluke 28II ( 4 X AA ) the battery indicator is still 100% after 8 months of ownership.

My only addition to this topic would be the idea of the voice recorder.
But Dave added it, in his wish list all ready.   ;)

I would suggest an simple in function, one single button recorder (20 seconds max recording time).
One long press ( 2 secs ) = Recording start.
One single press at the recording = Recording stop
One single press  = Playback
And every new message erases the old.
 

Offline UnaClocker

  • Contributor
  • Posts: 13
Re: EEVblog #184 - Open Hardware Multimeter Concept
« Reply #8 on: July 05, 2011, 01:47:43 am »
Love the idea! Couple suggestions..
Why not put an internal Li-Ion pack in the unit? Little barrel jack on the side for occasional charging. Seems like that'd be better than a AA solution, aside from the people that never bother to charge until it's flat dead, then they can't just hot swap the battery..
Next suggestion, make sure it can read frequency. Sometimes that's the only thing I use my oscilliscope for, and if I had a slightly better multimeter, I wouldn't need to.
Coupled with that suggestion, maybe a duty cycle indicator. I can think of at least a couple sensors on automobiles that put out a signal that is both frequency and duty cycle dependant. It'd be great to read those both at the same time without having to use a scope.. :)

I like the idea of ditching the rotary knob.
 

Offline reagle

  • Supporter
  • ****
  • Posts: 524
  • Country: us
    • KuzyaTech
Re: EEVblog #184 - Open Hardware Multimeter Concept
« Reply #9 on: July 05, 2011, 02:01:32 am »
Another feature (if it's not there yet)- Amp Hour meter. Makes it nice to play with batteries and do capacity checks etc

Offline Rutger

  • Regular Contributor
  • *
  • Posts: 198
  • Country: us
Re: EEVblog #184 - Open Hardware Multimeter Concept
« Reply #10 on: July 06, 2011, 12:04:55 am »
Hi Dave,

I like the concept but I have some other ideas that might be interesting.
You mentioned Blue-Tooth in the blog and I know you are more focused on a hardware solution, but how about the following concept.

Move all the display and user interface controls to a smartphone app using the Blue-Tooth as an interface to the hardware input/output.
So you keep the input & output jacks on the multimeter box, add more battery capacity for a realtime blue-tooth communication and write an app for the iphone/android smartphones that take the raw data from the multimeter and have the software process and display all the information you need.
This would eliminate the displays and extra selector buttons, sd card interface etc and allow for a whole range of display configurations and interface skins as well and ofcourse data logging and features we haven't even thought of yet.

With my software background I could help build open source software, but the hardware side I would leave to experts like yourself.

Rutger

 

Offline James

  • Contributor
  • Posts: 8
  • Country: 00
Re: EEVblog #184 - Open Hardware Multimeter Concept
« Reply #11 on: July 07, 2011, 10:27:16 am »
Love the concept Dave! Was just thinking last night how neat it could be to pack a uCurrent into the multimeter, and as there is a constant current output and multiple channels you could even allow for 4 wire resistance measurement using the multimeter to do the maths and display the outputs!

James
All electronic devices work off smoke...Once the smoke is let out they no longer work!
 

Offline Zad

  • Super Contributor
  • ***
  • Posts: 1011
  • Country: gb
    • Digital Wizardry, Analogue Alchemy, Software Sorcery
Re: EEVblog #184 - Open Hardware Multimeter Concept
« Reply #12 on: July 07, 2011, 12:27:08 pm »
As an aside, I'm glad to see that I am not the only one who sketches stuff. I find that drawing ideas lets my brain wander around the "what ifs" whilst constraining it to what is practically realisable. I find 3D tools really good for just playing around with styles and give investors a better impression of what they are funding.

This is a mock-up I did 7 years ago for some bench test equipment. I had been asked to make it look more retro, so took the styling from an old Hameg oscilloscope. Unfortunately I can't show the final design for contractual reasons. The extra case depth enabled the use of expansion modules.





Offline NiHaoMike

  • Super Contributor
  • ***
  • Posts: 4452
  • Country: us
  • "Don't turn it on - Take it apart!"
    • Facebook Page
Re: EEVblog #184 - Open Hardware Multimeter Concept
« Reply #13 on: July 08, 2011, 01:37:58 pm »
At work, I use a setup that reads 4 Agilent multimeters using LabVIEW for measuring the efficiency of a voltage regulator while the PC sweeps parameters. That's where there's no alternative to some form of networking interface.

Using a USB UART chip and optoisolators would be enough since 4 channels at 16 bits and 20Hz update rate each is only 1280bps, easily handled by a UART even if the data formatting quadruples the data rate. Or a Bluetooth chip, but that might be unreliable in noisy environments and could possibly disturb sensitive setups. With an isolated DC/DC converter, the battery could be charged over USB.
Cryptocurrency has taught me to love math and at the same time be baffled by it.

Cryptocurrency lesson 0: Altcoins and Bitcoin are not the same thing.
 

Offline valentinc

  • Regular Contributor
  • *
  • Posts: 116
  • Country: ro
Re: EEVblog #184 - Open Hardware Multimeter Concept
« Reply #14 on: July 10, 2011, 05:05:36 am »
   I like the idea of 4 isolated channels, very interesting;

   About the SD memory card, I would prefer (but it's a subjective opinion) a bluetooth chip so that I could do some measurements at one desk and do the data logging at another PC at other desk, with no wiring or the need to take out the card, put in the card reader and copy the files ... and it's also portable, you can do data logging on a laptop or smartphone to, no need for an SD card at all ...

   About the rotary switch, I would prefer some soft buttons instead (like the BK Precision 879B LCR meter)

   The display in my opinion should be a big, clear, dot matrix, high resolution display, where you can choose to view 2 parameters or 4 at one time, graphs, and other stuff

   And the last, but not least feature I consider is a simple semiconducter parameter tester like a mini curve tracer so you can view the real characteristics of a ... let's say, tunnel diode (although it's not a common example at all, but I consider this feature very useful)
Valentin
 

Offline gregariz

  • Frequent Contributor
  • **
  • Posts: 551
  • Country: us
Re: EEVblog #184 - Open Hardware Multimeter Concept
« Reply #15 on: July 11, 2011, 01:01:46 pm »
Hi Dave,

Maybe there aren't too many of us that regularly use transistors anymore in our designs but I find the HFE test a handy one on a meter. Ironically (since we are in electronics) the only meters that seem to have HFE tests on them are the really cheap ones. Dedicated HFE meters tend to be a bit old and ratty these days so I'd prefer a quick multimeter test instead. The best meter that I could find that had a HFE test at my local Fry's was a velleman 890 - and its really not much of a meter, although they put some decent probes on it so I surprise myself and use it quite often.

http://www.vellemanusa.com/us/enu/product/view/?id=521423#

It seems to be the cheaper the meter the more likely it has a transistor test. The downside in only being able to find a HFE test on a cheap meter is that alot of the transistor sockets have pretty cheap and nasty contacts in them so you have to wiggle them around to make contact. It took me a few 'meter returns' before I got the velleman above.

The other thing I like that may seem a bit weird is to be able to turn the auto-range off and manually flick through the range settings. I find this handy if I'm looking at batches of components and I need to just see that its a eg. a 1K resistor rather than a 993.9 ohm resistor.

Lastly, and I'm sure everyone has different opinions about this but I have a few meters and I find the ones I use most are the large and heavy ones simply because they can stand up to a bit of abuse on the bench. I have a fluke and a extech but they are both pretty small and are always sliding around and falling over - so I use them if I need to take them somewheres. Ironically that velleman is a bit of a monster so I kind of like the big display. I also have another cheap Mastech 8226 that is almost as big and I like to use that one too.

True RMS is not really a deal for me, as long as its RMS on a sine measurement thats' good enough for me, I generally like to use something else than a multimeter for those measurements.

I understand your reluctance to put a USB on the meter and I think the SD card logging is a great idea. But I would still put the USB on in anycase. I got the Mastech 8226 for the sole purpose of being able to set up a PC test using Labview so that's a but hard if you only have dedicated logging.

Anyways those are just my opinions - I seem to have collected a number of meters both cheap and normal in order to do all the little things I want and itd be nice if they could all be done in one meter.
 

Offline Zad

  • Super Contributor
  • ***
  • Posts: 1011
  • Country: gb
    • Digital Wizardry, Analogue Alchemy, Software Sorcery
Re: EEVblog #184 - Open Hardware Multimeter Concept
« Reply #16 on: July 11, 2011, 02:30:08 pm »
Sounds like you need something like this http://www.peakelec.co.uk/acatalog/jz_dca55.html

Peak Electronics have some great little test devices, and pretty cheap too. The semiconductor analyser is only US$60.

Offline gregariz

  • Frequent Contributor
  • **
  • Posts: 551
  • Country: us
Re: EEVblog #184 - Open Hardware Multimeter Concept
« Reply #17 on: July 11, 2011, 03:08:33 pm »
Your quite right Zad,

I do look at them regularly, and have looked at the one you mention. I'm sure in the end I will end up getting a dedicated transistor tester. Elektor did a full PC based curve tracer a while back and admit to have looked at that on a regular basis. There's also a bunch on ebay that look a bit like the Peak one at about the same price as well. For some reason transistor curve tracers a still quite pricey on ebay, for which I'm surprised.

If your lab is like mine though.. I have trouble finding what I have...I just don't need another piece of test gear.

 

Offline johnwa

  • Regular Contributor
  • *
  • Posts: 216
  • Country: au
    • loopgain.net - a few of my projects
Re: EEVblog #184 - Open Hardware Multimeter Concept
« Reply #18 on: July 12, 2011, 07:33:48 pm »
Hi,

The open source multimeter sounds like a great idea - it will be interesting to see where this goes.

I had a somewhat similar idea a while ago, although I never did anything with it. My plan was to make something for measuring parameters that could not be measured by ordinary meters. This was in some ways going to be a collection of multimeter 'add-on' adapter circuits that had been published over the years, all in the one box with a meter. I was going to use a nice big analogue meter movement, as all the meters I had at the time were cheap digitals with slow update. Planned features were:

* High impedance FET input
* True RMS
* Capacitance / inductance (back before this was standard)
* Direct reading resistance scale
* HV leakage / megger
* Low ohms scale.
* Fast continuity
* Power
* Zener test
* Plug-in probes for environmental measurement: light, sound, temperature, humidity, etc.

While most of these are not really relevant to the current design, I think the zener test and perhaps the HV leakage test might be worthwhile inclusions.

Dave's idea of the multiple isolated input channels is a good one. We had a BWD 'Powerscope' at uni, that was designed along these lines for working with 3 phase power electronics. Although I don't think you mentioned it Dave, I presume what you were getting at was the ability to have a real time readout of the efficiency of a power supply, etc. This would let you watch the efficiency while twiddling various circuit parameters, instead of having to laboriously recalculate it after every change.

 

Offline scopeman

  • Regular Contributor
  • *
  • Posts: 143
Re: EEVblog #184 - Open Hardware Multimeter Concept
« Reply #19 on: July 13, 2011, 03:29:13 am »
Hi Dave,

In terms of increasing battery life you can always use the Energizer Ultimate Lithium AA cells. It's true they are expensive but in my Sony DSC-H5 camera they last about a year with very frequent use. Energizer also makes a AAA  (and they may even have a AAAA) version of these cells. I have started to use these in many of my hand held test equipment items. They also have a very long shelf life.

I one point they had a D cell version but they must have discontinued those as in a flashlight (is that at "torch" to you?) they would offer standby service for something like 15 years! and power a 5 cell flashlight for a very long time.

Sam
 

Online EEVblog

  • Administrator
  • *****
  • Posts: 26829
  • Country: au
    • EEVblog
Re: EEVblog #184 - Open Hardware Multimeter Concept
« Reply #20 on: July 13, 2011, 05:08:55 pm »
Although I don't think you mentioned it Dave, I presume what you were getting at was the ability to have a real time readout of the efficiency of a power supply, etc. This would let you watch the efficiency while twiddling various circuit parameters, instead of having to laboriously recalculate it after every change.

Yes, that was a given.
Once you have voltage and current at the same time, power is trivial.
And if you have two power channels, you'd naturally have the ability to display efficiency.

Dave.
 

alm

  • Guest
Re: EEVblog #184 - Open Hardware Multimeter Concept
« Reply #21 on: July 14, 2011, 05:30:42 am »
Once you have voltage and current at the same time, power is trivial.
As long as it's DC or the load is purely resistive.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf