Page 1 of 1

Metering Data Logging Advice

Posted: Tue Oct 12, 2021 11:21 pm
by Maxwell_Epstein
Hi all,

I'm not sure how I should log the current and voltage data that I collect while running my demo system. I know that there are some data logging multimeters like the Fluke 287, but even the cheap ones are somewhat expensive (especially given that I'll need two for current and voltage). I assume that there is a way to use an arduino or similar device as a multimeter (with resistors and such on a breakout board), but that can't be a good solution or else everyone would be doing it, right?

I have also acquired a Varian 804-a thermocouple gauge (along with a compatible tube) that I am going to use with my new metal chamber that has red and black knobs on the back labeled "Recorder." I assume that these connections would output the vacuum information. What format would that be in and how could I read and log it?

I would greatly appreciate some advice to steer me in the right direction as I'm not sure what sort of best practices there are for this sort of thing and couldn't find any threads discussing this in the FAQ.

Thanks!
Max

Re: Metering Data Logging Advice

Posted: Wed Oct 13, 2021 12:06 am
by Liam David
Arduinos are absolutely a viable means of data collection. I use several to monitor voltage, current, pressure, etc.. and count neutron pulses, and Joe Gayo runs his systems with Arduinos as well. In some cases the built-in ADC resolution doesn't cut it, so I use one of the Arduinos to connect to four ADS-1115 modules. I have some information in this post, and the rest of the thread: viewtopic.php?f=18&t=13542&start=20#p90361. I literally have resistors and op-amps on solderless breadboards, so it doesn't need to be anything fancy. Most people don't bother with this kind of data acquisition since it's not necessary to do fusion. Can't say much about the TC gauge controller output, but my guess is a simple analog signal.

Re: Metering Data Logging Advice

Posted: Wed Oct 13, 2021 12:36 pm
by Maxwell_Epstein
Thanks for your advice! I've read through the thread that you linked and have a much better understanding of what I can do accurately with arduinos. Right now, I am measuring current over a resistor and voltage using a high voltage probe. Am I right in saying that I could keep all of that the same and simply replace the multimeters themselves with an arduino? I have no reason to believe otherwise, but would like to be sure. Thanks again!

Re: Metering Data Logging Advice

Posted: Wed Oct 13, 2021 12:50 pm
by Liam David
For measuring current you can just replace the multimeter with an Arduino, but do NOT do that for the HV probe. Assuming it's one of those standard HV probes with a large internal resistor, the multimeter itself is the 2nd resistor in the voltage divider. The impedance of the Arduino ADC is something like 100M and is not constant, especially since there are multiplexers and other front-end circuitry built in. You'll need to emulate the multimeter's internal resistance with some other resistor, and best practice would then be to use an op-amp to buffer the signal to the Arduino.

Re: Metering Data Logging Advice

Posted: Wed Oct 13, 2021 1:06 pm
by Maxwell_Epstein
I see, it seems like I need to do some more reading to make sure I set everything up properly. It's great to know, however, that I'll be able to add data logging to my system without getting new multimeters. Thank you for your quick and informative responses, your advice will be very helpful moving forward.

Re: Metering Data Logging Advice

Posted: Wed Oct 13, 2021 6:10 pm
by Richard Hull
I really can't imagine why the 10 bit resolution on the analog ports in the Arduino are not good for any analog signal sent to the Arduino. The largest number needed to be encoded to 10 bits might be your HV value. Does any livng being working at 60,000 volts really need better than 58 volt resolution at 45,873 volts actual. Would you be happy at 45,815 or 45,931 being reported? What does a 58 volt error mean to you? At 100 volts it is intolerable. At 5,000 volts it is not all that critical. Beyond 25,000 volts it is just not worth beans. What fabulous change in fusion exists at 58 volt error readings at 45,873 volts in reality? The need for 12, 14, 16 bit precision can be absurd at some point in the amateur fusion world.

Current wise, you are far better off assuming you want to measure the milliamps very accurately. Realizing you will never hit 100ma, (absolute electron runaway in most any fusor and dead power supply fuse blowing), you need to look at 100/1024 = 970 ua +/-. Thus you can warrant correct reading to .1ma Thus, 10.2 ma or 28.6ma from the arduino is about as accurate as you will ever need.

This is also true for pressure measurement and thermal measurement of the shell.

Where the average non-hacker falls short is in critically adjusting the span of their desired signal to the analog port to 0 volts to +5 volts. Most folks should set a point at 4 volts to indicate a warning and 4.5 volts as an instant HV shut down especially as relates to current. This assumes you want automatic protection from all you analog readings to prevent damage or runaway conditins.

Be very careful with A-D conversions. Use common sense. and the arduino will suffice. $4.95 will by the Pro-mini and about $9.00 will buy either the UNO or the Nano. Slaving multiple pro-mini's via I2C will make the master free to do nothing but manage all of its children.

Richard Hull

Re: Metering Data Logging Advice

Posted: Wed Oct 13, 2021 9:09 pm
by Liam David
In most cases I absolutely agree, 10 bits is sufficient. The reason I use 16-bit ADCs is mainly because of my (full-range) pressure gauge, which has a nonlinear voltage-to-pressure relation 10^(5/3*V-11.46). With 10 bits the minimum resolution at typical fusor pressures is ~1 mtorr, but fluctuations in the LSB means it's more like 3-5 mtorr. I just didn't have good enough repeatability, especially since the fusor operates in such a narrow pressure-voltage-current regime. It's all application-dependent.

Re: Metering Data Logging Advice

Posted: Thu Oct 14, 2021 12:26 am
by Joe Gayo
The input impedance of an Arduino (328P) is relatively low (NOT 100M). From the manual ...
23.6.1 Analog Input Circuitry
The analog input circuitry for single-ended channels is illustrated in Figure 23-8 on page 212 An analog source applied to
ADCn is subjected to the pin capacitance and input leakage of that pin, regardless of whether that channel is selected as
input for the ADC. When the channel is selected, the source must drive the S/H capacitor through the series resistance
(combined resistance in the input path).

The ADC is optimized for analog signals with an output impedance of approximately 10kohm or less. If such a source is used,
the sampling time will be negligible. If a source with higher impedance is used, the sampling time will depend on how long
time the source needs to charge the S/H capacitor, with can vary widely. The user is recommended to only use low
impedance sources with slowly varying signals, since this minimizes the required charge transfer to the S/H capacitor.

Signal components higher than the Nyquist frequency (fADC/2) should not be present for either kind of channels, to avoid
distortion from unpredictable signal convolution. The user is advised to remove high-frequency components with a low-pass
filter before applying the signals as inputs to the ADC.
For high-quality measurements, buffer with a cheap op-amp and RC. As a bonus, this can allow negative voltage measurements and voltages higher than 5V. Also, use the internal voltage reference ... and turn off the digital IO functionality for unused ADC pins.

Re: Metering Data Logging Advice

Posted: Thu Oct 14, 2021 1:01 pm
by Liam David
That's the recommended source impedance for a good measurement. The internal series impedance before the S/H capacitor is given as 1-100k, and 100M is roughly that of the ADC chip itself (ignoring I_IH and I_IL). My concern for using the Arduino as R2 in a high voltage divider stems from both this high impedance, and the unknowns with the MUX and other front-end components that would make for a messy measurement. Connect the HV probe directly, and the Arduino may suddenly see a large fraction of the voltage if the ADC switches channels and the MUX input is left floating. That, and even through a 1G resistor, the 13pF capacitor gets charged to Arduino-killing voltages in just a few 10s of microseconds.

adc.PNG

Re: Metering Data Logging Advice

Posted: Thu Oct 14, 2021 4:23 pm
by Joe Gayo
They recommend a low source impedance because the input impedance isn't very high ... there's a Microchip app note about it somewhere.