Reverse-engineering a small XRT / High Voltage / Forums

Forums

4hv.org :: Forums :: High Voltage

« Previous topic | Next topic »

Reverse-engineering a small XRT

1 2 3 next

Move Thread

LAN_403

klugesmith

Sun Nov 21 2010, 06:20PM

Registered Member #2099 Joined: Wed Apr 29 2009, 12:22AM
Location: Los Altos, California
Posts: 1716

This is to report some electrical measurements on a dental x-ray transformer (nominally 70 kVp, 7 mA, 60 Hz). They give some perspective to low-duty-cycle imperatives.

My transformer has a 5.25-inch square laminated E-E core, in two riveted sections.
I had to take it apart to rework the mounting brackets.
Took the opportunity to set aside the plastic-shrouded secondary coils. Made some electrical measurements with no load, and no high voltage in the way.

At 120 V RMS, the core is nearly saturated - I figure 1.82 T peak flux. As in a MOT, this is to get plenty of volts per turn per pound. Magnetizing current reaches about 6 amps RMS. A real wattmeter indicated that about 86 W were heating the transformer: 36 W copper loss in the 1.0 ohm primary winding, and 50 W core loss.

To get the flux, I made an inordinately fancy sense winding with 17 turns and the best flux linkage I could manage. Voltage ratio shows that the primary has exactly 174 turns, in 2 layers: from the white lead it's right hand helix up, then left hand helix back down to the black lead. We get 0.690 V RMS per turn from 2.2 in^2 (14.2 cm^2) core area.

Here is a chart of the data. Magnetizing current was measured two ways: first with an old Amprobe clip-on analog meter, then with a true-RMS Fluke DMM in series. The difference at lower voltages could be due to the clip-on meter reading low, or to a bigger core air gap when I assembled it for the series measurements. (minimizing the airgap is uncritical at such high flux levels).

For next time: secondary winding details and estimates of losses with a HV load.

Patrick

Mon Nov 22 2010, 06:19AM

Registered Member #2431 Joined: Tue Oct 13 2009, 09:47PM
Location: Chico, CA. USA
Posts: 5639

this is a great thread.... please continue to post.

Xray

Fri Nov 26 2010, 06:47PM

Registered Member #3429 Joined: Sun Nov 21 2010, 02:04AM
Location: Minnesota, USA
Posts: 288

Nice job testing the magnetic properties of the core!

I repair and rebuild dental X-ray heads for a living, and this is one brand that I have repaired MANY times over the years. It's a Gendex GX-770. I can tell you that a good working head will draw about 7 or 8 Amps from the mains. The control circuit has an "inverse voltage protection circuit" in series with the primary winding which consists of a high current diode across a low ohms (typically 10 Ohms) resistor. During the half cylcle when the tube is conducting and loading the hv secondary, the diode is forward biased, and as a result, shorts out the 10 Ohm resistor, which allows the transformer to get full voltage. During the half cycle when the tube is not conducting, the diode is reversed bias, and the voltage then flows through the resistor which limits the high voltage to a safe value so that it doesn't arc between windings. The diode/resistor circuit is only needed for X-ray applications. You won't need it if you use the transformer for a jacob's ladder or Tesla coil.

Also keep in mind that X-ray transformers, unlike neon sign transformers, are not current limited. Therefore, as the secondary load current increases, the primary current will continue to increase until either the transformer burns out, or the mains fuse blows.

Have fun!

Xray

klugesmith

Sat Dec 18 2010, 06:19AM

Registered Member #2099 Joined: Wed Apr 29 2009, 12:22AM
Location: Los Altos, California
Posts: 1716

Here's some more data.
1) Reassembled the core w/ attention to minimizing the airgap. This cut the magnetizing current in half, except near saturation where the iron permeability is relatively low.

2) captured primary current and secondary voltage waveforms, still unloaded, for RMS primary voltages of 30, 60, 90, and 114. That's as high as I could get with the isolation transformer on hand. Near saturation, it takes a large current to create the required magnetization (so the induced back-EMF almost matches the applied voltage). When the current peaks, IR drop subtracts significantly from the effective primary voltage. This distorts the secondary voltage waveform, but happens near zero crossing so we don't lose much RMS voltage. Dotted line indicates core flux for the 114V case, inferred by integrating the voltage of secondary winding.

We can plot BH curves, with current on one axis and integrated voltage on the other. This could be observed in real time using a voltage integrator circuit (fluxmeter) and an oscilloscope with X/Y mode. By factoring in the core dimensions and number of winding turns, the X axis could be labeled in oersteds or amps/meter, and the Y axis in teslas or gauss.

Finally, I took the opportunity to get some inrush current measurements when the transformer is turned on by a switch at arbitrary phase of mains voltage cycle. Purpose is to illustrate that we drop right into steady-state magnetizing current if the switch is closed at voltage PEAK. We get the highest inrush current when switch is closed at voltage zero crossing. It would be much higher still if we had a full 120 V RMS and no isolation transformer.

Meatball

Sat Dec 18 2010, 06:31AM

Registered Member #2401 Joined: Mon Sept 28 2009, 04:25PM
Location:
Posts: 74

What are you going to do with this transformer?

EDIT: That's a lot of POT!

Sulaiman

Sat Dec 18 2010, 12:50PM

Registered Member #162 Joined: Mon Feb 13 2006, 10:25AM
Location: United Kingdom
Posts: 3141

I like the resistor//diode in the primary trick - new one to me.

I would expect a transformer to interleave the core laminations to absolutely minimise any effective airgap for minimum magnetising current, and have the primary and secondary co-axial.
Why does this x-ray transformer separate the primary and secondary windings with an airgap? A leakage flux path has been deliberately introduced....just curious.

GeordieBoy

Sat Dec 18 2010, 02:33PM

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

Some good testing of the transformer.

> Reassembled the core w/ attention to minimizing the airgap. This cut the magnetizing current in half, except near saturation where the iron permeability is relatively low.

I was going to say that the no-load current draw (or "magnetising current") will be quite dependent on how well you can get the faces of the two core halves to mate. Even the smallest of air-gap in here will increase the no-load current draw of the transformer by decreasing it's inductance. That is why most lower voltage power transformers use interleaved E and I laminations. It makes the core set virtually impossible to disassemble once its been dipped though.

The 3rd harmonic distortion you are seeing in the voltage and current waveforms is a clear sign that the core is operating well past the onset of saturation.

For transformers the minimum "inrush current" is always observed when the power is turned on at a voltage peak. This is because the transformer flux is the integral of the applied voltage. If you turn on at a voltage zero crossing, then the peak flux integrates up to twice the normal peak value and the core is pushed into saturation. In a design where the core is already barely big enough, this can cause huge current spikes if turned on at a voltage zero. This transient core flux imbalance gradually decays away after switch on depending on the DC resistance of the winding and the supply impedance. For large high power transformers they can continue to growl and draw big inrush spikes for several cycles because the DC resistance is so low.

-Richie,

Proud Mary

Sat Dec 18 2010, 03:04PM

Registered Member #543 Joined: Tue Feb 20 2007, 04:26PM
Location: UK
Posts: 4992

How about a soft starter to temper the existential threat of sudden on-ness - all those violent spikes and roaring surges searching for short cuts in every weakness.

Xray

Sat Dec 18 2010, 06:15PM

Registered Member #3429 Joined: Sun Nov 21 2010, 02:04AM
Location: Minnesota, USA
Posts: 288

Sulaiman wrote ...

I like the resistor//diode in the primary trick - new one to me.

Dental X-ray manufacturer's have been using that trick for many decades. I suspect (and I'm only guessing here) that the transformer needs to be designed in such a way that it can accept a "modified" sinewave. By that I mean, half of the sinewave has a lower peak voltage than the other half, and, because of the diode, the waveform doesn't look "normal".

Sulaiman wrote ...

Why does this x-ray transformer separate the primary and secondary windings with an airgap? A leakage flux path has been deliberately introduced....just curious.

I think the air gap is there to allow the oil to help cool the windings, and it also provides insulation between the hv secondaries, and the primary. Most 60Hz AC X-ray transformers for dental application utilize two secondary windings with each "low" end tied to ground, or one of them is sometimes used to monitor the tube current, and is brought out to a terminal (the GX-770 does this). In a 70KV tubehead, each secondary provides 35KV -- one for the cathode of the tube, and one for the anode. This design allows for smaller spacing inside the head because there is only 35KV at any one point to ground. Only the X-ray tube "sees" 70KV. That's another "trick" that is utilized in dental tubeheads.

Of course, in "DC" heads, things are quite different. The Cathode is at ground potential, and the anode get the full DC high voltage, usually from a CW multiplier.

klugesmith

Sat Dec 18 2010, 07:24PM

Registered Member #2099 Joined: Wed Apr 29 2009, 12:22AM
Location: Los Altos, California
Posts: 1716

Xray: thanks for the detail about typical current draw from the mains. You taught me the resistor||diode trick last year, after selling me the transformer. I haven't applied it yet, but you will see the waveforms when I do.

Here is another view of its function. In normal operation with self-rectifying Coolidge tube, the secondary winding carries a net DC current of 7 mA. That's about 487 ampere-turns.
Turns ratio is about 1:400, so to keep the core flux balanced we need an average DC current of 2.8 amps in the primary winding. I expect this will be in the form of 5.6 amp pulses during half of each cycle, in phase with the voltage, and will be superimposed on the magnetizing current as reported above.

Flyback converters also have net DC current in both primary in secondary, but the pulses are not concurrent, so each cycle's energy must be stored in the magnetic field, hence the airgap.

In the XRT case I think there's no deliberate airgap. The laminated core is non-interleaved to facilitate construction and service. A thin air gap in part of the contact area (due to non flatness) doesn't matter much because the flux density is so high -- the air is no longer 10000 times more reluctive than the iron. Look at the black and blue curves in my before-and-after chart.

1 2 3 next

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