Page 1 of 2

Home Built Metering Panel

Posted: Sat Jun 27, 2020 4:24 pm
by Ed Meserve
Dennis made an suggestion that I detail the build for my home built metering panel as an example for other newbies, so here it is.

The first step for the uninitiated, should always be the FAQs; Read, then re-read viewtopic.php?f=29&t=4267 , viewtopic.php?f=29&t=8840 , and viewtopic.php?f=29&t=4266 . Richard Hull goes over everything you need to understand in great detail in these FAQs.

I wanted to see Voltage, Current and Pressure all on the same panel so I can capture all three neatly with a webcam and I wanted to use analog meters so I didn't need to included a separate power source for the meters.
Per Richard's FAQ advice, I purchased two 100uA meters and one 50uA meter from Amazon for about $10 each.
100uAMeter.jpg
100uA Meter on Amazon

These didn't have internal resistors which made the process of calibration simpler.


VOLTAGE:

If you want to measure voltage, your resistor block needs to be in series with your current meter. The flowing current only has one path, so if you know the resistance (selected) and the current (measured) by Ohm's law, you know the Voltage.
For the High Voltage, if you are using a 50uA meter, you need 200MegOhm per 10kV. Where this is only intended for the demo fusor, 10kV is plenty, so I have two 100 MOhm resistors, sealed with some liquid electrical tape on a stand off.

Pay attention to the meter polarity; the high voltage side of the resistor block is negative, so you will need to connect the negative terminal of the current meter to the powered side of the line and the positive terminal to the ground.
This sets the gauge on the meter to read 1kV for every 5uA.
200MOhm Resistor Block.jpg
200MOhm Resistor Block


VACUUM PRESSURE:

The deadlock vacuum gauge I'm using has a 0-10V proportional output signal for pressure; the lower the voltage the lower the pressure. Same rule applies for this as did for the high voltage as it does for the 0-10V DC signal. Select a resistor amount with the given voltage (10V) so that you get your meter full scale (100uA). R=V/I = 100kOhm. Just so happens that I had a couple 47kOhm and a 10kOhm trimpot Which allowed me to adjust to the correct reading. This made the 0-100uA gauge into a 0-10V meter, which is nice on the math so I only needed to mentally move the decimal point to get the voltage. Armed with the voltage, I then converted it to Torr per MKS' manual which instructed that the pressure can be found using a simple equation: P=10^(Vout -6).
VacuumGauge w Breakout connector.jpg
Deadlock Gauge w/ Breakout Connector


CURRENT:

If you want to measure current safely, you should divide the current between a calibrated shunt and your meter so that most of the current still flows through the shunt, but a smaller and proportional amount of current flows through the meter.
The fusor will be passing current in the mA range, and our meter is a uA so we need a calibrated shunt with the proper resistance to divert 1/1000 of the current to the meter.

I used a length of 28 gauge nichrome wire as the shunt for the current metering.
28GNichromeWire.jpg
Nicrome Wire on Amazon
In order to calibrate your shunt, you need a known current source. I used my Fluke 787 to source 20mA. You could also use a 9 V battery and 450Ohm resistor, measure the current with an ammeter and calibrate it that way.
CalibrateShunt.jpg
Calibrating Shunt
I then connected a length of the nichrome wire in parallel with the meter. If the current reading on the meter was lower than 20uA, then the wire shunt was too short. If the reading was higher than 20uA, then the wire shunt was too long. Be sure to turn off your current source before disconnecting and making adjustments to the position of the wire shunt; you don't want to accidentally hit your 100uA meter with 1000x the current it is rated for. With a little bit of trial and error, calibrating the current meter should go quickly.

Once I had the length of the nicrome wire shunt calibrated, I wound the wire around a bit on non conductive tubing and taped it in place, being careful not let the wire touch itself. I then wired it to a terminal strip, with a jumper on both end to other terminals so I could wire the meter and the ground wire from the chamber to separate terminals to make it easier if I needed to rewire later down the road.
Shunt in terminal.jpg
Shunt in terminals
Shunt Terminal Mounted.jpg
Terminal Strip Mounted

ENCLOSURE:

Once the meters themselves were set, then all is needed is a panel to mount them in. I used what scrap I had lying around for a box. I cut three holes and wired the meter to terminals on the back, again for ease of re-wiring later.
Notice the top meter for pressure has the resistor and trimpot in the enclosure with the meter.
Inside Meter Panel.jpg
Inside Meter Panel
Back Meter Panel.jpg
Back of Meter Panel
Completed Meter Panel.jpg
Completed Meter Panel

Re: Home Built Metering Panel

Posted: Sun Jun 28, 2020 3:06 am
by Richard Hull
Nice report on your effort with useful images and explanation.

Richard Hull

Re: Home Built Metering Panel

Posted: Sun Jun 28, 2020 9:34 am
by Dennis P Brown
Great report; including the previous FAQ's was a really useful idea. Making the gauge system using available meters for any fusor or demo is a necessary and important post. Your sourcing and details are essential for most newbies. This article is very much like the old Sci Am amateur scientist articles.

Re: Home Built Metering Panel

Posted: Sun Jun 28, 2020 4:32 pm
by Rich Feldman
Nice report, Ed.

Can we see more detail about design of the HV metering resistor string?
In particular, the voltage rating of the individual R's.
Not the same as value computed from power rating and resistance; the voltage rating usually gets you first when R value is many megohms.
Not necessarily improved by oil immersion.

Applause for scratch-built and calibrated current shunt resistor, with warning about applying full current if shunt is not reliably connected.
I did a similar exercise in 2015 for an electromagnet ammeter (+/- 15 A). Coat hanger wire wasn't hard to solder to. Temperature coefficient is relatively low for nichrome and high for copper, with steel in between.
DSCN9838.JPG
I'd suggest a step 1: determine the resistance and full-scale voltage of bare meter, so the shunt design can start with an estimated target resistance.

Yay for wooden boxes and panels !

Re: Home Built Metering Panel

Posted: Sun Jun 28, 2020 4:50 pm
by Richard Hull
Rich is correct. in a 400 megohm resistor string, oil or parafin or epoxy in casement can, depending on purity of the fill, can act as a shunt resistance to 200 megohms, reducing its value.
A better deal would have been 4 - 50meg reisitors in series in air. Ideal, would be 10 - 20meg resistors in a series ladder in air. 10kv is a pretty low voltage, but the more resistors to meet the resistance, the less stress on the components and zero corona issues.

Far more important is corona suppression from the fine leads of the top most resistors. Use teflon sleeving over resistor leads or epoxy or parafin encasement of the leads, but not shunting the resistor body...Just the fine leads.

Richard Hull

Re: Home Built Metering Panel

Posted: Sun Jun 28, 2020 5:49 pm
by Ed Meserve
The resistors I used came from Digikey. They are rated for 4kV , 1W.
rnx.pdf
RNX Datasheet
(85.51 KiB) Downloaded 30 times
I soldered the two 100 Ohm resistors leads together, then slipped them into an small segment of aquarium airline tube, with just air in the tube. Then I sealed them in with Liquid Tape. Once cured, I slipped that into a 7/16" bit of tubing I had lying around and sealed that with Liquid tape, including a bit of the fine leads.
HV String Components.jpg
Resistor String Components
HV String Close-up.jpg
HV Resistor String Close Up
HV Resistor String w ref.jpg
Resistor String w/ Reference

Re: Home Built Metering Panel

Posted: Sun Jun 28, 2020 7:10 pm
by John Futter
Ed
well done
Although I would have mounted the HV dropping resistor on a bit of acrylic not plywood
At 10kV you might get away with it but not at 20kV

Re: Home Built Metering Panel

Posted: Sun Jun 28, 2020 9:38 pm
by Ed Meserve
John,
How would you rate a PEX standoff (yeah I know, plumbing, but it's what I have lying around) with Two part epoxy with a zip tie for additional strength?
The plexi would be vertical and the PEX would be horizontal standoffs.

This is a junk piece of plexi for testing. The one I would use won't have the gunk down the edges. Also the zip tie orientation would be vertical to resist the most downward force.

Any recommendation for Standoff distance from the Plexi? The Richard's FAQ calls out a 3" distance on spacing between stand offs. And judging by his photo, it looks to be about 3" off from the plexi face.


HV Standoff Strength Test1.jpg
Standoff Strength Test
HV Standoff Strength Test2.jpg
Close Up

Re: Home Built Metering Panel

Posted: Sun Jun 28, 2020 9:55 pm
by Rich Feldman
Ed, thanks for clarifying that you got special high voltage resistors.
Looks like a nice conservative design. Is the estimated bottom-line accuracy OK? (Gain can be trimmed using ordinary resistors, series or shunt, if you have access to a HV reference source or meter.)

I bet inner tube would be warmed and softened, quite tolerably, by full-scale power dissipation. A very casual experiment with 270 ohm resistor of similar size is sort of confirming that.

Re. John's comment, it looks like the molded plastic terminal blocks could handle a few thousand volts between metal and plywood.
If there's no arcing or corona visible in the dark, with full unloaded NST potential (can you peg the kilovoltmeter?) then it ought be fine with the much lower voltages of a glowing plasma.

The meter will indicate average rectified voltage, not the same as RMS or peak. I think it's fine, probably more common than not, to use average DC voltage in plasma reports and combination pictures. Greg Courville once presented plasma voltage waveforms that hardly resemble rectified sinusoids.

Re: Home Built Metering Panel

Posted: Sun Jun 28, 2020 10:00 pm
by Richard Hull
Most all plastics are good for standoff purposes. At extremely high voltages, smooth unbroken surfaces of most any substance can have arc tracking issues. Towers between resistors mounted on a flat plastic sheet are much better as the tension/strain along the surface is decreased a lot due to linear distance increase along the surface. This assumes a decent air separation in addition, of course.

Attached a rather ideal 50kv metering system. Much beyond this and corona can become a bear to deal with requiring more extreme precautions and construction.

Richard Hull