Open Source Multimeter

[Rerouter]

isolate your adc's, using optocouplers

[glatocha]

but the reference and power supply for the ADC will be anyway same

[branadic]

I found that we have a PREMA 5017 at work. Pretty interesting stuff about that device and lots of pictures can be found here:

http://ohh.de/5017.htm

http://www.amplifier.cd/Test_Equipment/other/Prema-5017.html

Seems that they use their 26bit ADC:

http://ohh.de/5610.htm

They use a pll to lock their adc to mains frequency. Also interesting is, they use an LM399H (Linear Technology) for the 7.5 Digit device. Would be interesting to know what reference they use in the 4x better specified (stability) device 6048.

[branadic]

Any news about the project?

[Legit-Design]

Didn't see any mentioning about this project here: https://sites.google.com/site/theopensourcemultimeter/documents

The Open Source Multimeter by George Randel 
(last updated: Sep 27, 2011)

No the project is not done by me.


http://www.usc.edu/CSSF/History/2012/Projects/S0913.pdf


http://www.vcstar.com/photos/2012/apr/30/165529/ Picture of Mr Randel with his oshw multimeter board.
That pcb seems to be 12/21/2011 and different layout.


Didn't find ANY documentation whatsoever in either of the zip files.

Files are for DipTrace http://www.diptrace.com/download.php Freeware for nonprofit use, no registration no bullshit.

Pictures below are from Mulrifinalboardsr1.zip

Schematic is really messy, almost looks like it was imported from some other cad program for sharing. This guy seems to have spent alot of time with this design, too bad he hasn't shared anything else. Would have preferred to see final assembled pcb working.

http://www.linear.com/product/LTC2440

LTC2440 - 24-Bit High Speed Differential Delta Sigma ADC with Selectable Speed/Resolution


http://www.st.com/web/catalog/mmc/FM141/SC1169/SS1031/LN1565/PF189782
stm103 
QFP 48

usb isolation

2 voltage channels

2 current channels

LC meter

true rms conversion chip

mostly linear technology parts

I attempted to create a BOM with the program, but since I'm not used to using this it wouldn't let me make it properly...
BOM included below: oshwmultimeter.pdf   (same part shown multiple times as quantity of 1, I removed resistors and capacitors from bom, since it started to look even more messy)


Click for full size schematic


http://speedy.sh/khnA3/oshwmultimeter-full-sch.pdf  9MB  Full schematic for viewing pleasure.

If anyone has any knowledge about this project please share!

[branadic]

None of the so called "Open Hardware DMM" projects with more than 4 1/2 digits are well documented or finished. For everything up to 4 1/2 digits the Intersil parts (e.g. ICL7135) are still a good choice, everything else is overkill:

http://www.kmitl.ac.th/~kswichit/DVM7135/DVM7135.htm

I'm not sure if this project here will ever reach a status with a functional prototype. I fear this is due to the choosen overkill FPGA solution and some disagreements how to solve several parts.
I wonder why there is no project that uses softer goals like 5 1/2 up to 6 1/2 digits, dual-/multichannel input, a seperated board for each channel with some CPLD-based multislope adc or simple 24bit sigma-delta adc, a reference board with multichannel output, a STM32 board with a 7" LCD for displaying the several channels and an isolated pc interface maybe with SD card option.
A big LCD allows for multichannel displaying, this would be something novel and therefore a unique feature.
There is no need for a project that can't do more than commercial DVMs/DMMs as you can get them well priced as second-hand.

Personally I don't like 8-segment LED, monochrome LCD or simple dot pixel matrix displays. If you can get a 7" graphic display:

http://www.ebay.de/itm/STM32F103ZET6-Development-Board-7-Touch-Panel-SSD1963-TFT-LCD-Module-/170893762278?pt=LH_DefaultDomain_0&hash=item27ca0fdee6

you can also display additional information such as statistics, trend for each channel and spend some logging functionallity etc.
But this is just my opinion.

[airiclenz]

...a >100$ display is an overkill in my mind. There are pretty good solutions out there - smaller (up to 4") but also they stay in the range of affordability.

[ve7xen]

Those displays are available for about $30 in singles. Cheaper than the 192x64 mono OLEDs I bought a few months ago for a project...

e.g. http://www.ebay.com/itm/4-3-TFT-LCD-Module-Display-Touch-Panel-Screen-PCB-Adapter-Build-in-SSD1963-/360674009570?pt=LH_DefaultDomain_0&hash=item53f9d8a1e2

[Legit-Design]

Lets just slap in Sharp memory lcd, so we can have an awesome batterylife.

http://fi.mouser.com/ProductDetail/Sharp-Microelectronics/LS027B7DH01/?qs=%2fha2pyFaduhwKuijHwmQ%252bQF18%252b4JzroItWztWhky8OowGAyTdpoDnQ%3d%3d

5pcs : 34,28 € /each


Specifications

• Size: 2.7 inch (6.86cm)
• Pixel format: 400 × 240
• Outline dimension: 62.8 × 42.82 × 1.4
• Active viewing area: 58.8 × 35.28
• Dot pitch: 0.147 × 0.147
• Input voltage: 5V
• Interface type: 3V serial
• Reflectance: 18%
• Power consumption*: 50?W
• Operating temp range: -20 to +70ºC
• Surface hardness: - 3H or more initial
  pencil hardness

*with static image

[romovs]

You're not going to get the low end market, I'd target something like 5000$ or so unit cost (I mean you do have to recoup the dev cost as well). But you could have in it a lot of the functionality of much higher priced scopes. Sure not every hobbyist is gonna buy it ... but I'm convinced there is enough of a market for it anyway. (I'm just a hobbyist/hacker and I have a 7500$ scope ...)

No. The only possibility is to squeeze into the niche between those cheap ass pc scopes and Chinese 300$ low ends.
There is no way for an average (group) of electronics enthusiast to design a marketable scope matching 5k scopes feature and price wise.

As for the multimeter. Go for it! Kikstarter seems to be all about gaining experience and cash through the stupidity of others 

[Marco]

I wonder why there is no project that uses softer goals like 5 1/2 up to 6 1/2 digits, dual-/multichannel input, a seperated board for each channel with some CPLD-based multislope adc or simple 24bit sigma-delta adc

CS5532/34 has 23 bit noise free, you could easily do 6 1/2 with that (although INL would corrupt a few LSB). Although I'd try to use the MCP3903, cheaper and a faster SPI, with oversampling and some dither you could probably beat the cirrus.

Personally I don't like 8-segment LED, monochrome LCD or simple dot pixel matrix displays. If you can get a 7" graphic display

I'd kinda like 8 segment LED for the readability ... but you can get a 3.2" touch screen for around 10 bucks, which makes it more attractive.

All in all you could have a kit with component costs in the range of 25$, so saleable for 50.

[greatal]

Hi all
I am kind of new in electronics and eevblog and its hard to read all these pages but as I work in Electrical meter company I think that’s maybe a good idea to use Rogowski coil (no magnetic interference like CT current sensing or no limits like shunt solutions) to measure current while measuring voltage it is so acquire and it is not so expensive there is some single chip solution by ST(it is kind of include some complete isolation solution using ordinary Ethernet network filter) although it’s so hard to setup but its working so good the meter could use only and only to probes to measure current and voltage at same time.

[SixD]

I have started to make a arduino based multimeter.

[codeboy2k]

This thread pops out of the wood work occasionally.. So I read back about 6 or 7 pages to catch up today...

I'll throw my comments into the mix here...

1) regarding the FPGA:  I think a low cost iCE40 with 384 LUTS for < $2.00 (qty 1 from digikey) is good enough to time the discharge on the downslope, from top to zero-crossing.  Also, the STM32 counters might be fast enough, but not wide enough and will overflow (do you have timings done yet?  How long does a cycle last?). Each iCE40 LUT has 1 LUT, 1 FF, and carry logic, so there is enough there for a few 24 bit counters, and you might even have enough to make a voltage and current ADC if you wanted to.  Also see my #6 below regarding the need for voltage and current.

2) the comparator needs to have low offset, and reasonably fast propagation time. Presumably less than 1 clock cycle otherwise you are losing counts. If you are counting at 100Mhz (10ns) you need a comparator that can switch it's output in < 10ns. If it's slower, you get count errors.
How fast you need to count depends on your desired resolution, the charge up/charge down times, and how many conversions per second you want to achieve.

3) there is an old trick from the Keithley meters to get higher resolutions without necessarily having to go to really high-speed counters (i.e. 100Mhz) and expensive high-speed comparators and precision op-amps..What they do is, for example, count with 10Mhz, then when the zero-crossing occurs, they take that count and save it. Then they amplify the residual on the integrating capacitor x10, reset the counter, and time the discharge of that amplified amount until the zero-crossing again. When it crosses zero, they add that count (divided by 10) to the previous total, then they take the residual of that, once more, x10 again, reset the counter and discharge that to zero. This gives two more digits of precision, using a slower counter, and less precise comparators. And of course it's slower and takes more time. But maybe it's something you want to consider too, to be able to use less expensive parts and still get good precision. If you do this residual discharge with a 100Mhz clock, you might get extra digits into the 7-1/2 to 8-1/2 regions (assuming that the noise, leakage etc, is well controlled).

4) regarding the input buffer: I think your first choice of LTC6240 is still the best choice. It's < $4 and the chopper stabilized LT1052 is more than $12.  Unless I am mistaken, the drift is not important on the input buffer (enlighten me if I am wrong about this). I think the input buffer only has to be stable for 1 conversion cycle. However, for the reference buffer you want to use a chopper stabilized opamp like the LT1052 (maybe not that one for that purpose, just pointing out that the zero-drift is more important on the reference buffer than it is on the input buffer).

5) for the display, there are plenty of 4.7" and 5" 800x400 LCD with touch displays these days. usually between $20-30.  That's not too bad and makes a nice user interface possible.

6) finally, I am working on an LCR meter design that uses a 24bit delta sigma ADC ( I am also using the LT6240 as input buffers on that project.. they *are* very nice).  You can use the STM32F4 pwm outputs to generate the sine (100Hz, 1kHz, 10Khz, 100Khz, etc), low pass filter it and send it into the device under test. Then measure the voltage across it and current through it and do the calculations. The FPU in the STM32F4 will help here.  It's really, really simple to do. I'd be happy to join up on the project and do the LCR part, but the entire AFE has to support 4 wire mode and is basically floating above ground (I measure between RED/BLACK sense leads differentially with two LTC6240s and amplify that with a 3rd difference amplifier, then send it into the ADC). Current through the DUT I measure from the black sense lead through a current sense network returning to analog ground.  For a multimeter we'd need to put the voltage divider in front of that. If you want LCR capabilities, you be best to consider it up front.

Sounds like an interesting project,
Cheers!

[mrflibble]

1) regarding the FPGA:  I think a low cost iCE40 with 384 LUTS for < $2.00 (qty 1 from digikey) is good enough to time the discharge on the downslope, from top to zero-crossing.  Also, the STM32 counters might be fast enough, but not wide enough and will overflow (do you have timings done yet?  How long does a cycle last?). Each iCE40 LUT has 1 LUT, 1 FF, and carry logic, so there is enough there for a few 24 bit counters, and you might even have enough to make a voltage and current ADC if you wanted to.  Also see my #6 below regarding the need for voltage and current.

The really small ICE40's do indeed have enough logic for the required counters + control logic. Unfortunately they do not have a PLL, so if you want to use a PLL then one of the bigger models is required. In the current design we have a spartan-6 however, but that is not necessarily because of the requirements for  counters + control logic. It also has to do with familiarity + a couple of handy features in spartan-6 that can be used for extra functionality.

2) the comparator needs to have low offset, and reasonably fast propagation time. Presumably less than 1 clock cycle otherwise you are losing counts. If you are counting at 100Mhz (10ns) you need a comparator that can switch it's output in < 10ns. If it's slower, you get count errors.
How fast you need to count depends on your desired resolution, the charge up/charge down times, and how many conversions per second you want to achieve.

The comparator doesn't have to be super fast, but it does make life easier. Just as important is that the propagation delay is reasonably constant over temperature etc.

3) there is an old trick from the Keithley meters to get higher resolutions without necessarily having to go to really high-speed counters (i.e. 100Mhz) and expensive high-speed comparators and precision op-amps..What they do is, for example, count with 10Mhz, then when the zero-crossing occurs, they take that count and save it. Then they amplify the residual on the integrating capacitor x10, reset the counter, and time the discharge of that amplified amount until the zero-crossing again. When it crosses zero, they add that count (divided by 10) to the previous total, then they take the residual of that, once more, x10 again, reset the counter and discharge that to zero. This gives two more digits of precision, using a slower counter, and less precise comparators. And of course it's slower and takes more time. But maybe it's something you want to consider too, to be able to use less expensive parts and still get good precision. If you do this residual discharge with a 100Mhz clock, you might get extra digits into the 7-1/2 to 8-1/2 regions (assuming that the noise, leakage etc, is well controlled).

That sounds interesting, how does this amplification of the residual work? Presumably not by increasing the residual charge on the capacitor, so how? Do a S&H, buffer amplify it to x10, and then store that charge on the capacitor, and then do a rundown again? Main part I don't readily see is how to amplify it x10, and then get the equivalent charge on the capacitor again so you can do a rundown again.

4) regarding the input buffer: I think your first choice of LTC6240 is still the best choice. It's < $4 and the chopper stabilized LT1052 is more than $12.  Unless I am mistaken, the drift is not important on the input buffer (enlighten me if I am wrong about this). I think the input buffer only has to be stable for 1 conversion cycle. However, for the reference buffer you want to use a chopper stabilized opamp like the LT1052 (maybe not that one for that purpose, just pointing out that the zero-drift is more important on the reference buffer than it is on the input buffer).

Agreed.

5) for the display, there are plenty of 4.7" and 5" 800x400 LCD with touch displays these days. usually between $20-30.  That's not too bad and makes a nice user interface possible.

Funky displays is waaaay at the bottom somewhere. And I hope this 800x400 lcd display comes with a decent version of android attached to it, because I for one am not interested in reinventing the gui wheel.

6) finally, I am working on an LCR meter design that uses a 24bit delta sigma ADC ( I am also using the LT6240 as input buffers on that project.. they *are* very nice).  You can use the STM32F4 pwm outputs to generate the sine (100Hz, 1kHz, 10Khz, 100Khz, etc), low pass filter it and send it into the device under test. Then measure the voltage across it and current through it and do the calculations. The FPU in the STM32F4 will help here.  It's really, really simple to do. I'd be happy to join up on the project and do the LCR part, but the entire AFE has to support 4 wire mode and is basically floating above ground (I measure between RED/BLACK sense leads differentially with two LTC6240s and amplify that with a 3rd difference amplifier, then send it into the ADC). Current through the DUT I measure from the black sense lead through a current sense network returning to analog ground.  For a multimeter we'd need to put the voltage divider in front of that. If you want LCR capabilities, you be best to consider it up front.

Sounds interesting. Do you have a sample schematic of how you think the LCR part should fit into it?

[Felicitus]

Here's a DVM the guys at the TU Berlin built in one of their lab courses. 5 1/2 digits and reasonably precise, even tough they use a simple approach. http://www.emsp.tu-berlin.de/fileadmin/fg232/Lehre/MixedSignal/Dateien/Digitalvoltmeter/Schaltplan_DVM.pdf.

I've read through the schematic above, banged my head against the wall  several times, and even did some spice simulations, however, I still don't get how they protect the ADC (or the LTC1150 opamp U100) from input overvoltage when their range switch is on the 2V range and one would feed more than 20V (or: more than 150V, because that's what the U100 opamp can handle at maximum) into it.

Is it a bad design, or have I overlooked something?

[Dave]

"Baah... Who the hell needs overload protection? You just have to be careful, that's all." Said no engineer ever.

[sync]

It's a fusible op amp. On an over load condition it pops and you just insert a new one.

Edit: On http://www.emsp.tu-berlin.de/fileadmin/fg232/Lehre/MixedSignal/Dateien/Digitalvoltmeter/] [url]http://www.emsp.tu-berlin.de/fileadmin/fg232/Lehre/MixedSignal/Dateien/Digitalvoltmeter/[/url] they wrote that it's reproducibility and linearity is on par with a 34401A. 

[mrflibble]

Here's a DVM the guys at the TU Berlin built in one of their lab courses. 5 1/2 digits and reasonably precise, even tough they use a simple approach. http://www.emsp.tu-berlin.de/fileadmin/fg232/Lehre/MixedSignal/Dateien/Digitalvoltmeter/Schaltplan_DVM.pdf.

I've read through the schematic above, banged my head against the wall  several times, and even did some spice simulations, however, I still don't get how they protect the ADC (or the LTC1150 opamp U100) from input overvoltage when their range switch is on the 2V range and one would feed more than 20V (or: more than 150V, because that's what the U100 opamp can handle at maximum) into it.

It's a simple design, which as you noticed has no input protection.

[gxti]

Also, the STM32 counters might be fast enough, but not wide enough and will overflow (do you have timings done yet?  How long does a cycle last?).

You can extend a 16 bit counter + input capture to effectively infinite width in software without much trouble, especially if the period of the signal being captured is guaranteed to be greater than the period of the 16 bit counter. I'm currently using a STM32 at 72MHz to timestamp pulse-per-second from GPS and it's been running glitch-free for quite a while. The algorithm would need adjusting to handle the case where cycles might be shorter but definitely within the realm of feasibility. Being able to just use a microcontroller without any supplemental logic would be quite nice.

Also since this thread seems to mostly be for random people to drop in and barf up some unnecessary bloatware features, allow me to indulge myself: if you're going to have an OCXO anyway, why not GPS-discipline it and have a 10MHz reference output? Haha, I'll see myself out...

[mrflibble]

Also since this thread seems to mostly be for random people to drop in and barf up some unnecessary bloatware features, allow me to indulge myself: if you're going to have an OCXO anyway, why not GPS-discipline it and have a 10MHz reference output? Haha, I'll see myself out...

Or the other way around, which at least for my use makes more sense. Have an external GPSDO with distribution amp and enable the multimeter to use an external 10 MHz reference.

[Felicitus]

It's a simple design, which as you noticed has no input protection.

Yes, but overrange "protection"? So this is clearly a really bad design.

[mrflibble]

It's a simple design, which as you noticed has no input protection.

Yes, but overrange "protection"? So this is clearly a really bad design.

*shrug* I found it useful as source of some inspiration.

[codeboy2k]

That sounds interesting, how does this amplification of the residual work? Presumably not by increasing the residual charge on the capacitor, so how? Do a S&H, buffer amplify it to x10, and then store that charge on the capacitor, and then do a rundown again? Main part I don't readily see is how to amplify it x10, and then get the equivalent charge on the capacitor again so you can do a rundown again.

I haven't caught up on this thread in a while.. basically, the residual is there already.. it's residual. On the first rundown, when the 0-crossing comparator fires, the micro switches in some analog switches that bring a x10 between the cap and the comparator, then they continue the rundown... when it crosses once more, the micro will switch in another x10 and continue running it down to get the residual until it crosses 0 once more. There is no need to sample and hold, just cross 0 three times, with the last 2 times being amplified crossings.

Do you have a sample schematic of how you think the LCR part should fit into it?

I was going to redraw the front end for this project but I haven't got to that yet... The schematic I do have now is for a client so I can't publish any parts of it.  But I'll try to find some time to put something together soon.

You can extend a 16 bit counter + input capture to effectively infinite width in software without much trouble, especially if the period of the signal being captured is guaranteed to be greater than the period of the 16 bit counter.

Agreed.  I was probably thinking that my software would be busy enough already, but you're right, just adding bits in software is not much of an extra load, so pretty easy to do.

Also since this thread seems to mostly be for random people to drop in and barf up some unnecessary bloatware features, allow me to indulge myself: if you're going to have an OCXO anyway, why not GPS-discipline it and have a 10MHz reference output? Haha, I'll see myself out...

... and discipline it with frickin' lasers too 

[mrflibble]

That sounds interesting, how does this amplification of the residual work? Presumably not by increasing the residual charge on the capacitor, so how? Do a S&H, buffer amplify it to x10, and then store that charge on the capacitor, and then do a rundown again? Main part I don't readily see is how to amplify it x10, and then get the equivalent charge on the capacitor again so you can do a rundown again.

I haven't caught up on this thread in a while.. basically, the residual is there already.. it's residual. On the first rundown, when the 0-crossing comparator fires, the micro switches in some analog switches that bring a x10 between the cap and the comparator, then they continue the rundown... when it crosses once more, the micro will switch in another x10 and continue running it down to get the residual until it crosses 0 once more. There is no need to sample and hold, just cross 0 three times, with the last 2 times being amplified crossings.

In which case I suspect I already know the approach you mean (multislope). The way you described it sounded a bit ... different. But yes, multislope is indeed the way to go. The current design is a simplified version of multislope (as in multi being two ), to limit the number of precision resistor ratios to something affordable.