Author Topic: Open Source Multimeter  (Read 129064 times)

0 Members and 2 Guests are viewing this topic.

Offline Frant

  • Regular Contributor
  • *
  • Posts: 52
Re: Open Source Multimeter
« Reply #25 on: July 02, 2011, 09:41:12 PM »
I think an open source meter with the same accuracy and safety features of a Fluke would sell because people would be able to see the the quality of the components and would know that it's got the protection of a Fluke but at a lower price.

I am not sure about this because of the safety aspect. Dave's teardowns showed us that safety of a multimeter relies not only on electronic components, but also on a specific case design. It might be possible to choose an existing handheld case from a good case manufacturer if there is a product that can meet the requirements. I see no point in designing an "open source" case that can only be economically fabricated in large series. There is also a complex range switch, probably a custom LCD and a custom keyboard. There is just too many specific parts, making it pointless to be an open source project.

Perhaps a modular benchtop PC-based instrument would be a better choice for an open source project. A decent metal case can be easily found, people would build their own modules for specific functions or as an upgrade, plenty of software possibilities etc. Something like a poor man's PXI or more like Agilent USB Modular Instruments, but much cheaper.

Offline Polossatik

  • Frequent Contributor
  • **
  • Posts: 285
  • Country: be
Re: Open Source Multimeter
« Reply #26 on: July 03, 2011, 01:38:38 AM »
there is handheld open source scope "ourdso", v2 is the latest revision, Based on Cyclone EP2C8Q208, as ADC the good known AD9288 (so 2 x 100MSs).
Based on NIOS , 320x240 ... far from perfect but it works already suficient. All sources available and actually can be easy modified to
support e.g. 2 x 500MSs or even 2 x 1GSs ADC, whatever.


can't find much about the "ourdso"  not listed here: http://www.opencircuits.com/Oscilloscope#Open_Source_Oscilloscopes

http://dangerousprototypes.com/2010/01/07/open-source-digital-sampling-oscilloscope/ tried it also one time

more on topic, I do not see any real value in making an OSS handheld multimeter, there is not much functionality you can add /improve on low cost existing ones  and price is also not a factor.
Real Circuit design time in minutes= (2 + Nscopes) Testim + (40 +120 Kbrewski) Nfriends

Testim = estimated time in minutes Nscopes= number of oscilloscopes present Kbrewski = linear approx of the nonlinear beer effect Nfriends = number of circuit design friends present

Offline House91320

  • Regular Contributor
  • *
  • Posts: 176
Re: Open Source Multimeter
« Reply #27 on: July 03, 2011, 07:55:36 AM »
It is suprizing how much reaction my post has created, wasn't expecting that. Well maybe i can clear up some stuff. The meter is going to have all the functions of a normal meter, volts, amps, ect, plus an esr meter, data logging, real time graphing, tft color screen, and user customization ui. I expect the price range to be between $100 and $200.
Also its going to use a rechargeable battery and is going to have USB and a corresponding software to do pc data logging and graphing.
« Last Edit: July 03, 2011, 08:02:57 AM by House91320 »

Offline EEVblog

  • Administrator
  • *****
  • Posts: 18461
  • Country: au
    • EEVblog
Re: Open Source Multimeter
« Reply #28 on: July 03, 2011, 11:02:39 AM »
A have to agree with the many proponents that such a project hasn't got much likelihood of being funded and/or can't compete with commercial units.
It would have to offer something novel and a feature so compelling that you just have to have it.
If it's just a multimeter (but open source), I wouldn't bother.

Cool project as an exercise for sure, but that's all it is at present. It needs a killer feature.

Dave.

Offline Uncle Vernon

  • Super Contributor
  • ***
  • Posts: 1216
  • Country: au
Re: Open Source Multimeter
« Reply #29 on: July 03, 2011, 11:14:53 AM »
A have to agree with the many proponents that such a project hasn't got much likelihood of being funded and/or can't compete with commercial units.
It would have to offer something novel and a feature so compelling that you just have to have it.
If it's just a multimeter (but open source), I wouldn't bother.

Cool project as an exercise for sure, but that's all it is at present. It needs a killer feature.

Dave.

I'll give this one the Led Zeppelin stamp too. You don't have to look much further than a Jaycar catalogue to see why.
A DIY DVM kit is priced high than an equivalent fully assembled and warranted junk Chinese DVM.
Small change got rained on with his own thirty eight ....

Alex

  • Guest
Re: Open Source Multimeter
« Reply #30 on: July 03, 2011, 11:18:33 AM »
I expect the price range to be between $100 and $200.

That sounds like a truly great price for all the features! What did you consider to arrive at these figures House91320?

Offline EEVblog

  • Administrator
  • *****
  • Posts: 18461
  • Country: au
    • EEVblog
Re: Open Source Multimeter
« Reply #31 on: July 03, 2011, 11:43:28 AM »
Also its going to use a rechargeable battery and is going to have USB and a corresponding software to do pc data logging and graphing.

Any isolation with that USB?

Battery life is very important in a hand held meter, beware.

Dave.

Offline EEVblog

  • Administrator
  • *****
  • Posts: 18461
  • Country: au
    • EEVblog
Re: Open Source Multimeter
« Reply #32 on: July 03, 2011, 11:51:08 AM »
I see no point in designing an "open source" case that can only be economically fabricated in large series. There is also a complex range switch, probably a custom LCD and a custom keyboard. There is just too many specific parts, making it pointless to be an open source project.

Yeah, it gets messy. I'm just not sold yet on "yet another multimeter".
Sure it might have a ESR meter built in, and/or maybe LCR capability, but really, there is little differentiation with commercial handhelds.

It's gotta have something novel. I don't know what that is, but I'll know it if I see it!

Maybe something like a Fluke 233 with just the front end module, and the display can be anything. A watch, an iPhone, PC, whatever, or it just talks to you, or, something...

Like the iDVM concept that talks to your iPhone/iPad:


That's at least novel (but it's specs suck arse).

Stop the press!
I have a concept...
Might make a good blog.
House91320, can I show your images in the video?

Dave.
« Last Edit: July 03, 2011, 12:29:44 PM by EEVblog »

Offline House91320

  • Regular Contributor
  • *
  • Posts: 176
Re: Open Source Multimeter
« Reply #33 on: July 03, 2011, 01:29:42 PM »
Ya go ahead and use it in a video. In response to usb isolation it its going to have some opto coupling for the data lines and I am going to use a transformer and dc to ac converter for the isolation of the power.

Offline MrPlacid

  • Frequent Contributor
  • **
  • Posts: 277
  • Country: us
  • Hobby Hobbyist
Re: Open Source Multimeter
« Reply #34 on: July 03, 2011, 01:32:45 PM »
Yeah, your open source multimeter must be completed with these feature:

Minimum, since these are already on the market:
-LCR
-Talking
-Dual trigger input like function to function as mini oscilloscope.
-Storage
-Temp
-Storage
-Watt meter
-luminosity
-audio meter
-High Voltage warning

Must be added in before release:
-Temperature reading without cable.
-Timer, UP, Down, Clock, Alarm clock (Talking timer too)
-Transmit wireless readings such as audio, video, photo, or just plain vanilla reading
 (You might skip the video part,  and use e-ink technology for photos. Black and white photo is better than nothing.)
-Emit loud sound on continuity mode for a distance.
-Built in camera, photo and movie viewer (again, skip the video stuff)
-Stability feature to alert if meter is accidentally knock over or moved.

These are just the top of my head. But you don't want to waste time making an inferior product only to be wipe out one or two years from now.







Offline MrPlacid

  • Frequent Contributor
  • **
  • Posts: 277
  • Country: us
  • Hobby Hobbyist
Re: Open Source Multimeter
« Reply #35 on: July 03, 2011, 01:38:32 PM »
Battery life is very important in a hand held meter, beware.
Dave.

I agree. Rechargeable battery is a death wish. It means that meter will never last more than 20 year max. It'll be lucky to last 5-10 years. After that time, it tells everyone how crappy the hobby multimeter is that they all die out (went extinct!)

Offline House91320

  • Regular Contributor
  • *
  • Posts: 176
Re: Open Source Multimeter
« Reply #36 on: July 03, 2011, 01:54:16 PM »
The battery is going to be user replaceable.

Offline MrPlacid

  • Frequent Contributor
  • **
  • Posts: 277
  • Country: us
  • Hobby Hobbyist
Re: Open Source Multimeter
« Reply #37 on: July 03, 2011, 02:19:51 PM »
The battery is going to be user replaceable.

Is that particular battery a standard battery that will be in production x number of years from now? Plus, the multimeter will be out of work until the replacement battery is delivered. It needs to be easy as running to any local store to pick up some fresh batteries.

When you need the unit most and it's out of juice while on the field, we can't just say, "Today's work is cancel since my battery is dead. See you next week once the new battery arrived."

Offline Bored@Work

  • Super Contributor
  • ***
  • Posts: 3938
  • Country: 00
Re: Open Source Multimeter
« Reply #38 on: July 03, 2011, 06:14:28 PM »
A DIY DVM kit is priced high than an equivalent fully assembled and warranted junk Chinese DVM.

Elenco kits are such an example. They resemble fully assembled meters but the kit costs four to five times as much as the assembled junk meter. E.g. compare the Elenco M-1007K with the iconic Chinese $5 junk meter incarnations, the 830th (e.g. DT830).
I delete PMs unread. If you have something to say, say it in public.
For all else: Profile->[Modify Profile]Buddies/Ignore List->Edit Ignore List

Online mikeselectricstuff

  • Super Contributor
  • ***
  • Posts: 7109
  • Country: gb
    • Mike's Electric Stuff
Re: Open Source Multimeter
« Reply #39 on: July 03, 2011, 08:09:25 PM »
A feature which I've not seen on any sensible-cost meter, but which would be very cheap to implement is to be able to measure simultaneous DC voltage and current,  for efficiency testing voltage regulators.
I say DC as it gets much tricker at AC. For DC it's just a multiplexer.
PC software would allow connecting to two DMMs for direct readout of efficiency.
Quote
In response to usb isolation it its going to have some opto coupling for the data lines and I am going to use a transformer and dc to ac converter for the isolation of the power.
As regards optoisolation, you don't need a DC/DC converter as you will always have power available on the USB side. The method used my most meters today is probably the optimal one, i.e. the interface at the DMM is optical, and an external USB-optical link is used. This reduces the cost on the  DMM side for people who don't want it, and avoids issues of connectors that are vulnerable to moisture and damage.  Bandwidth is low so a simple LED/phototransistor provides a very low-cost solution.

Quote
tft color screen
Don't see the point in this ( apart from the 'gadget factor') as it will kill battery life. You just don't need colour on a DMM. What you do need is good visibility from no light to sunlight and long battery life

Getting an optimal display is a major challenge - a fully graphical display probably takes too much power, but an off-the-shelf fixed-segment one is not sufficiently flexible.
The ultimate answer is a custom segmented display with small matrix areas for units and function-key labels, but this means a couple of $K for LCD tooling. Unit cost is low after that though.
« Last Edit: July 03, 2011, 08:11:20 PM by mikeselectricstuff »
Mike's Electric Stuff: High voltage, vintage electronics, electrical curiosities
Day Job: UFOs, LEDs etc.
Versatronics & other low-end pick & place forum

Offline Polossatik

  • Frequent Contributor
  • **
  • Posts: 285
  • Country: be
Re: Open Source Multimeter
« Reply #40 on: July 03, 2011, 08:24:40 PM »

Quote
tft color screen
Don't see the point in this ( apart from the 'gadget factor') as it will kill battery life. You just don't need colour on a DMM. What you do need is good visibility from no light to sunlight and long battery life

Getting an optimal display is a major challenge - a fully graphical display probably takes too much power, but an off-the-shelf fixed-segment one is not sufficiently flexible.

+1

Colour lcd is just plain nonsense for a DMM seen you use a DMM not only on the bench but from the darkest corners up to full sunlight on a ladder .
Of course you might add a pixelQ screen to it, but that would be seriously overkill :)

I still use an old and silly (from specs perspective) vellman handheld scope for some stuff simply because it has a monochrome graphic lcd, hence you can use it in a lot of environments and actually *see* what is on the screen.


I would pay a premium to have an "up to date" mobile phone (so that you can use it as 3g BT modem for example) but with a non-color LCD screen that is actually readable outside like the old nokia 6310's .
well, now that's an idea :)
Real Circuit design time in minutes= (2 + Nscopes) Testim + (40 +120 Kbrewski) Nfriends

Testim = estimated time in minutes Nscopes= number of oscilloscopes present Kbrewski = linear approx of the nonlinear beer effect Nfriends = number of circuit design friends present

Offline EEVblog

  • Administrator
  • *****
  • Posts: 18461
  • Country: au
    • EEVblog
Re: Open Source Multimeter
« Reply #41 on: July 03, 2011, 09:00:52 PM »
My video suggestion is rendering now as the next blog!
But here is the DaveCad sketch!

Dave.

Offline ejeffrey

  • Super Contributor
  • ***
  • Posts: 1368
  • Country: us
Re: Open Source Multimeter
« Reply #42 on: July 03, 2011, 09:16:18 PM »
Yeah, your open source multimeter must be completed with these feature:

Minimum, since these are already on the market:
-LCR

Great idea, I don't know how technically difficult it is to make a combined multi-meter and good LCR meter, but if possible it would be nice.

Quote
-Talking
-Dual trigger input like function to function as mini oscilloscope.

Do not want.  A strip-chart style display is the most "scope like" feature I need or want from a voltmeter, but I would rather have longer battery life from a non-graphic LCD.  Talking is just a silly gimmick 99% of the time.  Maybe it can be implemented in a useful way, but adding it as a bullet-point feature is just going to suck.

Quote
-luminosity
-audio meter
-Transmit wireless readings such as audio, video, photo, or just plain vanilla reading
-Built in camera, photo and movie viewer (again, skip the video stuff)
-Stability feature to alert if meter is accidentally knock over or moved.
-Timer, UP, Down, Clock, Alarm clock (Talking timer too)

If you can add some of these features without compromising cost, battery life, usability, or form factor maybe, but I wouldn't go overboard on the kitchen sink features.  Better to make a top notch meter and add features that support the basic usage.  Every feature you add is another position on the range switch, or another entry in a menu, and turning your meter into a full-fledged computer will do a number on battery life.  Important features for a meter are things like logging, storage, and USB/wireless links.  Maybe some more flexible audible alert: rather than a talking meter, how about one where you can set a threshold voltage (or other value) and it beeps when that level is exceeded?

One thing that might be nice for a rechargeable meter is one of those fancy near-field power links.  I don't know how well they work or how much interference they cause, but it would solve the isolation issues, and allow you to use the meter in 'logging' mode simply by resting it in a cradle.

It would be really nice if you could get a useful battery life out of 3 low-discharge AA NiMH batteries.  They are cheap, replacements will be widely available for decades, last a reasonable amount of time, and can be substituted with alkaline batteries in a pinch (make sure the charging circuit doesn't try to charge alkaline batteries).  Lithium ion batteries are lighter and higher capacity, but hard to find replacements for.

Online mikeselectricstuff

  • Super Contributor
  • ***
  • Posts: 7109
  • Country: gb
    • Mike's Electric Stuff
Re: Open Source Multimeter
« Reply #43 on: July 03, 2011, 10:08:14 PM »
I think you probably still want the range switch for everyday funcitons, so maybe it should have an 'other' position for fancy stuff and multi-channel functions...
Diode test should go to about 5V, and it should be able to do 4-wire ohms, with ohms-to-frequency audio tone for tracing shorts.

Mike's Electric Stuff: High voltage, vintage electronics, electrical curiosities
Day Job: UFOs, LEDs etc.
Versatronics & other low-end pick & place forum

Online firewalker

  • Super Contributor
  • ***
  • Posts: 1875
  • Country: gr
Re: Open Source Multimeter
« Reply #44 on: July 03, 2011, 11:52:41 PM »
Maybe instruments design companies that watch this forum could contribute to the cause. I believe it would be an excellent use of social networks.

Alexander.
« Last Edit: July 03, 2011, 11:59:01 PM by firewalker »
Become a realist, stay a dreamer.


Offline EEVblog

  • Administrator
  • *****
  • Posts: 18461
  • Country: au
    • EEVblog
Re: Open Source Multimeter
« Reply #45 on: July 03, 2011, 11:56:08 PM »


Dave.

Online firewalker

  • Super Contributor
  • ***
  • Posts: 1875
  • Country: gr
Re: Open Source Multimeter
« Reply #46 on: July 03, 2011, 11:58:25 PM »
Just saw Dave's video. Really great IMHO. As an open source/hardware  it would probably need at least couple of years to hit the market.

Alexander.
Become a realist, stay a dreamer.


Offline Slobodan

  • Regular Contributor
  • *
  • Posts: 159
  • Country: cs
Re: Open Source Multimeter
« Reply #47 on: July 04, 2011, 12:55:02 AM »
@Dave

Nice video.

Online mikeselectricstuff

  • Super Contributor
  • ***
  • Posts: 7109
  • Country: gb
    • Mike's Electric Stuff
Re: Open Source Multimeter
« Reply #48 on: July 04, 2011, 01:21:29 AM »
My video suggestion is rendering now as the next blog!
But here is the DaveCad sketch!

Dave.
Thinking about it, you don't really need 4 isolated channels - for the majority of cases you only need 2 isolated channels, each doing V & I, as it will usually be possible to arrange a common point (even if it means readings are negative) . This would save a lot of cost for not much reduction in usefulness, as you need a whole front-end and behind each isolation barrier.
Also you could save more by having a 'primary' channel with all the component test stuff and a V/I channel, a secondary isolated V/I channels and isolation on the host interface.

For temperature, PT100/PT1000 would be good, and no additional hardware cost.

I'd prefer USB stick to SD, as all PCs have a USB interface but not all have an SD reader. 
However I think optoisolated serial with external USB converter gives you the most usefulness/cost ratio. Bluetooth is an option but a bit costly and power-hungry.

Another benefit of a USB host port is you could plug in a USB bluetooth stick when remote use is required.
« Last Edit: July 04, 2011, 01:31:07 AM by mikeselectricstuff »
Mike's Electric Stuff: High voltage, vintage electronics, electrical curiosities
Day Job: UFOs, LEDs etc.
Versatronics & other low-end pick & place forum

Offline Frant

  • Regular Contributor
  • *
  • Posts: 52
Re: Open Source Multimeter
« Reply #49 on: July 04, 2011, 02:21:55 AM »
@Dave

Great video! I really admire your enthusiasm!

About your suggestion to put four isolated channels... Yeah, that would be nice, but I don't really know of any isolation mechanism that could be used for this without wasting too much power for a battery powered meter. A separate ADC chip would be necessary for every channel. OK, there are low power high performance chips, such as AD779x, ADS124x or something like that, really nice. However, they would all need a dedicated isolated power supply, which is not so nice, and a separate SPI bus isolation, which is not nice at all.  Analog Devices iCoupler digital isolators with integrated isolated DC/DC converters may be used for this. However, just the mentioned parts are well over $100 already...

A modern reflective or "transflective" graphic LCD would not consume much power with the backlight off, probably 10 times less than the isolation of a single input channel.

The range switch can be replaced with software controlled relays as in bench multimeters, but that would significantly increase the overall power consumption. A combination of opto-MOS and bistable electromagnetic relays would probably be acceptable.

With all the input channels isolated and "floating", the controller side can be connected directly to a PC via the USB port. There is up to 500 mA available on a host USB port, which is more than enough for the whole thing. When using batteries, it would be essential to prevent the unused channels from drawing any supply current.


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf