Current transformer question [AC] / General Science and Electronics / Forums

Forums

4hv.org :: Forums :: General Science and Electronics

« Previous topic | Next topic »

Current transformer question [AC]

1 2 next

Move Thread

LAN_403

syntroniks

Sun May 10 2009, 02:37PM

Registered Member #1530 Joined: Tue Jun 10 2008, 03:34PM
Location:
Posts: 32

So my DIY Variac box is all god and all, but the ammeter doesn't work well at all. It is a DC ammeter of course >.>

Anyway: My only challenge is that I don't want to use any active components (op amp, anything like that).

I think the meter has 75mV full scale movement. Can this be done? I could probably cobble up a transformer but I see a problem with rectification... Any suggestions?

Steve Conner

Sun May 10 2009, 04:16PM

Registered Member #30 Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706

Just connect the ammeter inside a large bridge rectifier, and put the AC terminals of the rectifier in series with the line.

This will read rectified average current, not RMS, so you'll need to mentally multiply the reading by 1.11 or whatever.

syntroniks

Sun May 10 2009, 05:20PM

Registered Member #1530 Joined: Tue Jun 10 2008, 03:34PM
Location:
Posts: 32

I tried to draw up a schematic to visualize it, but won't that give me DC out? My SSTC would be fine with that, but I have standard plugs on this box and would like to be able to use normal equipment with it.

I don't mind average vs peak vs RMS, just as it is now, it is either reading ~1 amp or >10 amps :0

Right now, there are 4 turns of wire around a toroid with the current-carrying wire passing through it and a large shottky... No resistor, and I doubt I will have a class that covers this material for another 2 years O.O

Should I be looking at the transformer as a current source?

---
EDIT

Guess it does work kind of like a current source. I did some (LT)Spice which gives me an idea where to head. It looks like:

39918343

I'd tweak R3 and R2 so that the scale works alright but am I on the right track?

3l3ctrici7y

Sun May 10 2009, 06:50PM

Registered Member #1806 Joined: Sun Nov 09 2008, 04:58AM
Location: USA
Posts: 136

I think you want 170 rather than 120 in SINE()
170 is the real voltage amplitude.
120 is the RMS voltage.

klugesmith

Mon May 11 2009, 09:39PM

Registered Member #2099 Joined: Wed Apr 29 2009, 12:22AM
Location: Los Altos, California
Posts: 1714

Let's say you want to use a current transformer, as opposed to the DC-ammeter-inside-bridge as suggested by Steve.
The latter does deliver an AC sinusoidal voltage to the load, minus just two diode drops and the inconsequential voltage drop of ammeter shunt. Its main weakness is nonlinearity & distortion of voltage waveform at the extreme low end of Variac range.

Back to the current transformer. You need to allow secondary current Is = Ip / N to flow freely in both directions,
and you need enough secondary voltage Vs = Ip * N to well exceed the drop of a rectifier diode,
and you need to get rid of the current shunt in your meter.

Try this in Spice: a 100:1 current transformer with a 100 ohm resistor [called a burden] on the secondary. and no diode yet. Choose L1 = 1 uH, L2 = 10,000 uH, K = 0.99.
With your sinusoidal current source of 10 peak amperes, we expect Is = 0.1 A peak (from turns ratio),
so Vs = 10 V peak (Ohm's law) and Vp = 0.1 V peak (turns ratio again). The effective resistance on
primary side is 0.01 ohm, which is Rs / N^2.

Now you can add a rectifed voltmeter in parallel with the burden resistor.
Please let us know how that works out.

Rich

Steve Conner

Tue May 12 2009, 10:24AM

Registered Member #30 Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706

No, the trick when using a CT is to put the burden after the rectifier. This removes the non-linearity due to the diodes.

GeordieBoy

Tue May 12 2009, 11:18AM

Registered Member #1232 Joined: Wed Jan 16 2008, 10:53PM
Location: Doon tha Toon!
Posts: 881

Yes, definitely put the CT burden after the bridge rectifier. Then the terminal voltage of the CT will rise to whatever is necessary to force the sense current to flow in the burden resistor. This compensates for the voltage drop of the diodes, which may vary with temperature, manufacturer, batch etc...

If you were to put the burden resistor directly across the CT output to convert it's current to a voltage, and then follow it with a bridge rectifier, there would be a gross measurement error, particularly for small currents. No DC output voltage would appear from the rectifier until there was enough current flowing to generate at least two "diode drops" worth of voltage across the burden. Only then would the rectifier start to conduct and produce some output. That is why it is not done this way!

-Richie,

klugesmith

Tue May 12 2009, 06:06PM

Registered Member #2099 Joined: Wed Apr 29 2009, 12:22AM
Location: Los Altos, California
Posts: 1714

Yup, I figured out the [bridge] rectifier-in-series-with-burden last night. But Steve and Richie beat me to the punch, succinctly and articulately, so I'm not here editing yesterday's post.
Got no firsthand experience with CT's feeding DC instruments, but have a DIY Variac box that has served for 30 years and could use an upgrade.

I hope my example L1 and L2 values help syntroniks to get his Spice simulations on track. Did some Googling to see if they were consistent with some real CT's. Found a surprisingly close match at http://www.surplussales.com/transformers/Pdfs/tm-pe-51687.pdf -- except those are made for 20 kHz and up.

A homebrew CT at 50 or 60 Hz must have a core big enough to avoid saturation. For about 2 volts peak on the secondary (based on bridge rectifer), the 100 turn example needs about 0.02 volts per turn, so a core area around 0.1 in^2 or 0.5 cm^2. I picked a suitable ferrite toroid in a catalog. It's almost big enough to wear on my pinky finger, and has a single-turn inductance of 3.6 uH.

Rich

Steve Conner

Tue May 12 2009, 09:16PM

Registered Member #30 Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706

CTs for power frequency are made with toroidal laminated iron cores, and they're usually quite a bit bigger than pinky sized. I don't know if ferrite works well.

GeordieBoy

Tue May 12 2009, 11:20PM

Registered Member #1232 Joined: Wed Jan 16 2008, 10:53PM
Location: Doon tha Toon!
Posts: 881

Mn/Zn ferrite doesn't work well for line frequency CTs at 50/60Hz) It's lack of distributed air-gap gives rise to rapid saturation as each cycle of current begins to flow. Once the ferrite is saturated, any further increase in output would be due to "rogowski coil" action I guess.

1 2 next

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob