Author Topic: Better Data Logger? Voltage, Current, Temp, Digital and Test Automation  (Read 2550 times)

0 Members and 1 Guest are viewing this topic.

Offline seanEETopic starter

  • Contributor
  • Posts: 15
  • Country: nz
Hi,
I’m an electronics engineer and I often find that I want to measure a couple parameters (voltage, current, temp, maybe a few digital pins) over a few hours or days at a relatively low data rate. I often also want simple test automation (switch a load on – drain a battery; switch load off – observe charging cycle etc.). It seems like I should be able to set up this sort of test in a few minutes, but all of the existing data acquisition solutions I can find take considerable time to configure (write code for logger, get a shunt resistor for current measurement etc.) Also, many of the existing options have limited input ranges, and their gain doesn't scale automatically with input signal amplitude. Has anyone else found this sort of thing to be a problem?

Anyway, I finally got frustrated enough and just went and made the data acquisition / test automation device I wanted. I call it “Trendeca”, you can see the details here:

http://trendeca.com/

If you have any feedback on the design, or have some other thoughts, I’m keen to hear them. Some of my engineering mates in the area wanted one, so I thought I would offer it to anyone who's interested.
Cheers,
Sean
 

Offline ovnr

  • Frequent Contributor
  • **
  • Posts: 658
  • Country: no
  • Lurker
First impressions: "Eh, it looks like a nice enough toy. Acquisition system seems to have a bit low resolution. And the form factor is a bit strange."

Then I saw the price. Hoo boy, that's not something I'd spend $1850 on. Maybe $300-400. Having to script my instruments over LXI/GPIB may suck, but it does not suck that much - plus I actually trust the results.
 

Offline beanflying

  • Super Contributor
  • ***
  • Posts: 7358
  • Country: au
  • Toys so very many Toys.
Seriously Ouch $ The Price point doesn't seem justifiable to me.

Integrated small form factor and matching software may have a place for some. WIFI why?? On board SD card and weatherproof design for remote logging or Cable is what is needed for lab work. WIFI/Bluetooth are not good options apart from maybe a data file dump.

Only K type T/C support is less than ideal in a logger in this price point. Anyone serious about Temps will want better than that. As per below strap a few RTD Maxim based boards onto an Arduino and it will kill your spec.

Voltage and Current Spec accuracy isn't going to cut it for the $ either. An Arduino, Calibrated shunt, Vref and a handful of quality Resistors would do as well or better or add some ADC's kill the accuracy spec.

I will stick to my Agilent 34970's and GPIB for a while and enjoy playing with my Arduinos.  ;)
Coffee, Food, R/C and electronics nerd in no particular order. Also CNC wannabe, 3D printer and Laser Cutter Junkie and just don't mention my TEA addiction....
 

Offline seanEETopic starter

  • Contributor
  • Posts: 15
  • Country: nz
Thanks for the feedback! Sounds like the price point was too high, I have now cut it substantially.

Regarding the second post, the device logs data directly to a USB drive in CSV format. The Wifi is included so I can check my test / operate the device from my desktop computer or at home. I have worked at a couple places where the test lab is down stairs / in another building and I was a bit tired of loosing time by going to check / modify my test.

Speaking of time, you both mentioned scripting or Arduinos. Sounds like other people have been putting together quick test solutions like I have over the years. The major motivation for making the device was I totaled up the number of hours I spent on scripts / Rpi / Arduino rigs, and that came out to $$$ way to much money (time) over just a few months.

Accuracy wise, I was aiming to make the device about equivalent to a reasonably good multi meter. Not good enough?

Any other feedback? I really appreciate receiving your comments.

Cheers,
Sean
« Last Edit: March 19, 2018, 07:05:40 pm by seanEE »
 

Online tautech

  • Super Contributor
  • ***
  • Posts: 28137
  • Country: nz
  • Taupaki Technologies Ltd. Siglent Distributor NZ.
    • Taupaki Technologies Ltd.

Any other feedback? I really appreciate receiving your comments.

Sean
Your product is up against a few products that can already do the same tasks and more at a much lower price.
Several of the Agilent /KS bench meters and a couple in the Siglent range provide multi-channel measurement and logging with GPIB, USD or LAN/LXI connectivity.
Both the 5.5 and 6.5 digit Siglent DMM's are available with a SC1016  16ch scanner card to enable exactly the data logging you're trying to provide for an additional US$ 259.
By owning the DMM and the scanner card effectively an engineer is armed with a more versatile piece of equipment and with a higher degree of accuracy.
You can see what you're up against here:
https://www.siglentamerica.com/digital-multimeters/

 
Avid Rabid Hobbyist
Siglent Youtube channel: https://www.youtube.com/@SiglentVideo/videos
 

Offline ovnr

  • Frequent Contributor
  • **
  • Posts: 658
  • Country: no
  • Lurker
I don't mind the wifi - and I expect its inclusion doesn't exactly break the bank. I mean, this is a SBC with an acquisition module strapped on top, in a pretty case.

Accuracy: Really not terribly impressed, to be honest. It's fine, but it's very much in the "here's a 16-bit ADC with an OK reference" range.

One question: why did you decide to go with the quick-connection wire terminals for all the inputs? I'd have thought some banana jacks would be more appropriate, at least for the main inputs.
 

Online thm_w

  • Super Contributor
  • ***
  • Posts: 6273
  • Country: ca
  • Non-expert
One question: why did you decide to go with the quick-connection wire terminals for all the inputs? I'd have thought some banana jacks would be more appropriate, at least for the main inputs.

Imagine how much space 24 banana jacks would take.
Also its very rare to see a thermocouple, NTC, RTD, etc. that uses banana plugs.
Profile -> Modify profile -> Look and Layout ->  Don't show users' signatures
 

Offline ovnr

  • Frequent Contributor
  • **
  • Posts: 658
  • Country: no
  • Lurker
Imagine how much space 24 banana jacks would take.
Also its very rare to see a thermocouple, NTC, RTD, etc. that uses banana plugs.

Obviously; but having six or so for the voltage and current channels would make it more useful tbh.
 

Online nctnico

  • Super Contributor
  • ***
  • Posts: 26754
  • Country: nl
    • NCT Developments

Any other feedback? I really appreciate receiving your comments.

Sean
Your product is up against a few products that can already do the same tasks and more at a much lower price.
Several of the Agilent /KS bench meters and a couple in the Siglent range provide multi-channel measurement and logging with GPIB, USD or LAN/LXI connectivity.
Both the 5.5 and 6.5 digit Siglent DMM's are available with a SC1016  16ch scanner card to enable exactly the data logging you're trying to provide for an additional US$ 259.
Perhaps but you'd typically don't leave a bench DMM somewhere to do logging and the Trendeca also has some outputs so it can also provide some test stimuli. I agree that the 1% accuracy isn't good enough. I'd be expecting at least 0.1% for the price.
There are small lies, big lies and then there is what is on the screen of your oscilloscope.
 

Offline beanflying

  • Super Contributor
  • ***
  • Posts: 7358
  • Country: au
  • Toys so very many Toys.
If you really need Banana jacks then build an external box. Screw terminals would be preferable even over quick connects.

For T/C's I do use K type polarized plugs on some of the stuff I make but screw terminals are more compact and more reliable so I use them mainly.

Re the Accuracy spec/price point and features you need to be at or better than most 4 1/2 digit meters for a start and even then pitching over $1k USD is a major stretch improving this should be a priority.

Consider even my s/hand 34970's 20 channels/card (1-3 max) 6 1/2 digit multimeter (34401A spec) fitted. They turn up on evilbay for $6-800 (USD) with a card fairly regularly and I got my two with four cards for less than that total. People serious about Lab style logging won't settle for 1%. I still have a very early (30 year old) 16ch logger in mothballs from http://www.datataker.com that would give your accuracy spec a run for it's money.
Coffee, Food, R/C and electronics nerd in no particular order. Also CNC wannabe, 3D printer and Laser Cutter Junkie and just don't mention my TEA addiction....
 

Offline joeqsmith

  • Super Contributor
  • ***
  • Posts: 11630
  • Country: us
Re: Better Data Logger? Voltage, Current, Temp, Digital and Test Automation
« Reply #10 on: March 20, 2018, 01:28:37 am »
Personally, I would rather stick with LabVIEW for a programming language on a PC.    Putting together fairly complex tests is easy.   With it all running on a PC, I can use the standard network to run it.    I do have a LabJack.  They offer several flavors that are fairly inexpensive.  They offer an open API.   Another nice thing with LabVIEW is how expandable things are and I can control pretty much anything.   Many times the controls for my testing can be more complex than a simple PID for example.   

https://youtu.be/Mkec-Eyhg7c?list=PLZSS2ajxhiQBcHhIaGpmm9GyZQfrCzqkv&t=340
https://youtu.be/04I7nHA_HxM?list=PLZSS2ajxhiQBcHhIaGpmm9GyZQfrCzqkv&t=469
https://youtu.be/q_89qoFMivg?t=2223

Offline seanEETopic starter

  • Contributor
  • Posts: 15
  • Country: nz
Re: Better Data Logger? Voltage, Current, Temp, Digital and Test Automation
« Reply #11 on: March 20, 2018, 06:37:42 pm »
Regarding the wire connectors (banana / screw terminal / other):

Yes, when I was designing this thing I really struggled to make that choice. The number of channels + banana connectors resulted in a form factor that was quite a bit bigger than what I wanted. Ultimately, I decided the idea I wanted to focus on with this design was easy of use and speed getting a test set up. Thus, I selected the push terminals as they do not need a screw driver to secure the wire. I also asked some of my engineering mates - opinion appeared to be equally divided among all solutions.

I was interested to note one of the posters divided data loggers into two camps: "hardened for outdoor use" and "high spec, comes with cable to plug into computer in lab". Regarding future versions / improvements, what do you think about:

-Hardening design for rugged outdoor use?
-Including battery in product (8 hours run time on single charge)?

Are there other features you would suggest?

Cheers,
Sean
 

Online Wolfgang

  • Super Contributor
  • ***
  • Posts: 1773
  • Country: de
  • Its great if it finally works !
    • Electronic Projects for Fun
Re: Better Data Logger? Voltage, Current, Temp, Digital and Test Automation
« Reply #12 on: August 11, 2018, 12:45:29 pm »
Hi,

looks OK, on first sight. Didi you have a look on the Rigol M300 before making this one ?
I used it for a long time test of a voltage reference (3 Voltages, 1 current, 2 temperatures).

electronicprojectsforfun.wordpress.com/silly-circuits/silly-circuits-a-heated-lm723-reference/

 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf