Voltage Metering Issues

This forum is for specialized infomation important to the construction and safe operation of the high voltage electrical supplies and related circuitry needed for fusor operation.
User avatar
Liam David
Posts: 161
Joined: Sat Jan 25, 2014 10:30 pm
Real name: Liam David
Location: Arizona

Voltage Metering Issues

Post by Liam David » Sun Aug 24, 2014 8:43 pm

I now have all the components to make an HV voltage and current metering system, but there is a problem. The voltage metering system doesn't work.
Here is the voltage metering string.
Image

The top of the 6-100M resistor string is connected to the "backwards facing" diodes on both of the NST HV outputs. The bottom of the string with the 4-100K resistors and 1-250K potentiometer is connected in parallel with a 0-200V 3 digit voltmeter. The pot is set to 200K. The bottom of the whole assembly is connected to the case of the NST with an earth ground connection between the metering and the NST. When I crank up the voltage on the 15kV NST, the 0-200V meter does not read anything. When a much more sensitive multimeter is attached, the readings are in the millivolts. If I increase the voltage to 60V on the primary, the voltage (in millivolts) increases. But when I turn it up past 60V, the voltage decreases. When I plug the numbers in on a voltage divider calculator, I should be getting 7.4V on the meter.

Image

Anyone know what I'm doing wrong here?
-Liam David

John Futter
Posts: 1624
Joined: Thu Apr 22, 2004 2:29 am
Real name: John Futter
Contact:

Re: Voltage Metering Issues

Post by John Futter » Sun Aug 24, 2014 8:55 pm

Liam
Where is the power supply for the meter?????
you only show the red and black input wires
You cannot use a self powering meter for this exercise

User avatar
Liam David
Posts: 161
Joined: Sat Jan 25, 2014 10:30 pm
Real name: Liam David
Location: Arizona

Re: Voltage Metering Issues

Post by Liam David » Sun Aug 24, 2014 8:56 pm

The meter doesn't have a separate power supply (it doesn't need one). That would explain half of my problem. The multimeter does count as a powered meter and it's giving me strange readings nonetheless.

John Futter
Posts: 1624
Joined: Thu Apr 22, 2004 2:29 am
Real name: John Futter
Contact:

Re: Voltage Metering Issues

Post by John Futter » Sun Aug 24, 2014 9:08 pm

Liam
for your meter to work it would require you to lower all your resistance values by about a hundred
this of course will use up all your available power as it would all be dissipated in the resistors.
Find a different panel meter that uses a separate power supply to power it ie 5V or some have 9-16 volts.
It is usually better to use a 200mV full scale meter where you can put to diodes back to back across the input to protect the meter input.
you would have to adjust your divider network for 200mV full scale

Jerry Biehler
Posts: 903
Joined: Tue Nov 24, 2009 8:08 am
Real name:
Location: Beaverton, OR

Re: Voltage Metering Issues

Post by Jerry Biehler » Sun Aug 24, 2014 11:08 pm

Also watch the voltage rating of the resistors you are using. Many of those smaller resistors are only rated at a couple hundred volts max.

User avatar
Chris Bradley
Posts: 2930
Joined: Fri May 02, 2008 11:05 am
Real name:

Re: Voltage Metering Issues

Post by Chris Bradley » Sun Aug 24, 2014 11:38 pm

Jerry, I'd suggest the electrical scorch marks showing on his photo, about one resistor-body apart, already indicate Liam is aware of this! :)

Liam, together with the other points, bear in mind that these electronic devices, a DVM and the electronic display, 'process' in some way the signals they receive before displaying a value. As you are measuring from two un-smoothed and half-rectified channels, this may play havoc with whatever value these devices will try to display. Frankly, I would recommend analogue meters are used by newcomers to HV electrics, until they get a good handle on the pitfalls, as there are far fewer ways they can 'lie' to you!

The additional issue is that these devices will also have an internal resistance, which in ordinary circuits will be much higher than the impedance of the voltage measured. In this case, the meters' resistances will be comparable with those of the voltage divider circuits.

I think that's about 5 or 6 issues mentioned so far you may want to improve on with your voltage divider circuit! I'm afraid HV electronics looks simple at first but has lots of gotchas that you will eventually learn by trying things out until it makes more sense!

In regards Jerry's point, you appear to have already taken a step to cover some of the resistor legs with plastic tubing. Is this to reduce corona or to avoid arc-over? My method (for both) is to use narrow adhesive heat shrink and use that to seal the legs between the components. No need to bother to solder the legs together, that can make it difficult to get the narrow tube over, just place the legs together side-by-side inside the heat-shrink. Again, HV is 'different' - the resistance of such a contact may be several, or even 10's, of ohms by not soldering them together but that doesn't matter a squat at several kV and 10 mA.

User avatar
Rich Feldman
Posts: 1314
Joined: Mon Dec 21, 2009 11:59 pm
Real name: Rich Feldman
Location: Santa Clara County, CA, USA

Re: Voltage Metering Issues

Post by Rich Feldman » Sun Aug 24, 2014 11:57 pm

1. Yes, what Jerry said.

Let's assume you are already comfortable with Ohm's law, and voltage dividers, and calculation of power in resistors. If not, go learn them -- the Internet is your friend.

A typical resistor model comes in a wide range of ohm values, all with the same nominal power rating (which depends on the physical size and the materials) and the same voltage rating (which also depends on the physical size and materials).
For example, http://www.ohmite.com/cat/res_minimox_200_300.pdf
Those 1/4 watt resistors have a 500V rating. Another common voltage rating for 1/4 watt resistors is 350 V.

Circuit designs are supposed to comply with _both_ ratings. In low ohm values the power usually gets you first, and in the high megohms it's the voltage rating that gets you first.

You will see that FAQs at fusor.net talk about using high voltage resistors for divider circuits. Richard Hull mentions cylindrical, axial-lead types that are physically long. Another popular style is exemplified on ebay under "100MEG 20KV 2.5W" (ohmite slim-mox)
Capture.JPG
2. As others have said, a digital panel meter in this application needs separate power, often a 9 volt battery. You can probably do the job less expensively with a low-end digital multimeter, $7 to $10, that has a 200 mV range.

3. You can use an analog panel meter with high-voltage scaling resistor, and put diodes in parallel with the meter movement to protect it. For example, with 100 megohms, a 0-100 uA meter would represent 0-10 kV.
http://hyperphysics.phy-astr.gsu.edu/hb ... olmet.html

Good luck!

[edit, Sunday night] Liam, your electrical metering work already puts you ahead of most demo-fusor makers.
Are you sure there's not some sneaky little arc (or a fusor plasma) that is loading down your NST?

Here is my attempt at an answer sheet for you to check your homework.
In theory, with 100% input voltage and no load, each HV terminal of NST would have a 7500 volt RMS sinusoid. Peak voltage 10600 V. Average rectified voltage 6750 V, neglecting voltage drop in the diodes. So your 601 megohm voltage divider string would have 11.2 microamps of current. The 0.6 megohm small section would get I * R = 6.74 volts, average.

Now your digital multimeter is probably good at indicating average DC voltage, particularly rejecting ripple at multiples of 50 Hz and 60 Hz. But as Chris said, it will load down and reduce the voltage of your divider circuit small section. Most DMM's have input impedance of 10 megohms, reducing the 600K divider output to 94% of its unloaded value. Some cheap DMM's have input impedance of 1 megohm. That would reduce the divider output voltage to 5/8 of its unloaded value, so a reading of 4.2 volts would not be surprising. You can determine your meter impedance by measuring the voltage of a battery, before and after putting 600K or 1M or 10M resistor in series with the meter. Don't forget that your fingers are conductive at that level.

As others have said, how about making the small section of your divider 6000 ohms, for theoretical output voltage of 67 mV? Then your multimeter would not significantly load down the divider.

In any case, you are expecting almost 1800 volts (peak) across each of your 100 megohm resistors. That's far above their voltage rating even though the power dissipation is only 13 mW.
Hope this is food for thought.
-Rich
All models are wrong; some models are useful. -- George Box

User avatar
Richard Hull
Moderator
Posts: 12295
Joined: Fri Jun 15, 2001 1:44 pm
Real name: Richard Hull

Re: Voltage Metering Issues

Post by Richard Hull » Mon Aug 25, 2014 1:38 pm

There are too many unknowns about the original meter shown. If it came ready to read 200 volts, then there are a new class of volt meters for line voltages where only two leads are needed. (takes the guesswork out of making an AC voltmeter). Two leads, 120VAC input and you get the power and rectification and dividing in one input. Make sure to never use one of these new types. They are worthless for fusor work.

You should get a small digital panel meter that only measures 200 millivolts and start from there. These meters all have four input connections, as alrady mentioned. 2 for power and 2 for the 0-200mv input. Note* most all of these come with no wires attached, but a data sheet that discusses where to solder leads to the rear PC board, how to set the decimal point and often a bit about divider chain resistors needed for reading higher voltages. You will have to noodle all of this out on your own, hopefully, with a little electronic smarts already at hand. Even these meters come in two varieties; ones that can read their own input power voltage, (common ground) and those that cannot and demand a completely isolated power supply. Know what you are buying and how to apply it.

Rich did a great job of explaining much of this and I will not do any follow up.

Richard Hull
Progress may have been a good thing once, but it just went on too long. - Yogi Berra
Fusion is the energy of the future....and it always will be
Retired now...Doing only what I want and not what I should...every day is a saturday.

User avatar
Liam David
Posts: 161
Joined: Sat Jan 25, 2014 10:30 pm
Real name: Liam David
Location: Arizona

Re: Voltage Metering Issues

Post by Liam David » Mon Aug 25, 2014 2:03 pm

Chris-
Luckily they aren't electrical scorch marks, just the constant abuse of chemicals before I got bitten by the fusion bug. The resistors came with one of their legs covered in plastic.

Rich-
I guess I'm back on the resistor hunt. These are the resistors I used: http://www.surplus-electronics-sales.co ... cts_id=292
The only load on my NST is the metering string, no fusor plasma or arcing. I'll try cranking the potentiometer down to 6K to compensate for the multimeter imput impedence.

Richard Hull- I picked up the 0-200V meters at a hamfest, so that's why I'm using them. I've already found some 0-200mV meters that fit your description.

Thanks for all the in-depth answers!
-Liam David

User avatar
Andrew Robinson
Moderator
Posts: 687
Joined: Sun Aug 11, 2013 5:54 am
Real name: Andrew Robinson
Location: Raleigh, North Carolina
Contact:

Re: Voltage Metering Issues

Post by Andrew Robinson » Thu Sep 04, 2014 11:39 pm

How goes things Liam? How is your transformer work going? Did you get in touch with One Tesla?
I can wire anything directly into anything! I'm the professor!

Post Reply