Developing HV DC Pulses

I'm trying to read between the lines of the Elsevier abstract quote below.

Does this read as, the experimenter is doing electric field separation with the electrodes in the emulsion?

Is there any reason you can't have the electrodes in solution and do electric field separation of an emulsion?

How could they change the current when the Voltage is held constant? (electrode spacing?) Thanks, Mikek

"As shown in Fig. 1, experiments were carried out in a laboratory scale oil treater where the direct current (DC) electric field with 60 Hz frequency was applied. The diameter and the height of the vessel containing the brass cone are 11.5 and 20 cm, respectively. The angle of the brass cone is 80° and the width of the copper strip is 8.5 cm. Water-in-oil emulsion from tank is injected into oil treater by pumping, with rotating by the centrifugal force. When emulsion droplets pass between brass Results and discussion

Fig. 2 shows the water separation efficiency of water-in-oil emulsion as the current intensity is changed. The voltage, temperature, and residence time are kept as 5 kV, 75 °C, and 3.2 s, respectively. The current intensities are changed to 1, 2.5, 4.9 mA. The current intensity had little effect on the water separation efficiency. At the current intensity of 2.5 mA the oil treater showed the best separation efficiency of 77.2% than other current intensities."

What does that mean?

The voltage rating makes some assumptions about the field configuration, i.e. that it's constant.

In 2-D, a corner with inside angle pi/m produces a field near the vertex that goes as r**(1-m).

That is, an angle of pi (i.e. a straight line) produces a constant field, an exterior right angle (pi/2) goes like 1/sqrt(r), and anything sharper goes faster, up to 1/r.

Sharp points in 3-D produce even steeper changes in the field.

That happens in dielectrics too, so voltage rating is not the whole story.

Cheers

Phil Hobbs

My take was fullwave rectified AC without a filter capacitor. Experiments so far have shown DC, not useful vs fullwave rectified AC. i.e pulsed DC.

I take that back! The fullwave rectified DC did not produce results comparable to AC. I'm wondering if a rest period between pulses is helpful? Thanks, Mikek

Spice Model for 6BK4 in case anyone is interested in modeling the proposed circuit. John, I'm wondering if a much high voltage grid to cathode is need to pinch off the tube at such High Voltages?

.SUBCKT 6BK4 1 2 3 ; Plate Grid Cathode

• PARAMS: CCG=2.6P CGP=1P CCP=1P

• MU=2103.1 KG1=16349.49 KP=40249.51 KVB=2504.16 VCT=0.95 EX=1.4

• VGOFF=-0.6 IGA=3.25E-4 IGB=0.18 IGC=5.76 IGEX=1.46

• Vp_MAX=25000 Ip_MAX=2.5 Vg_step=2 Vg_start=0 Vg_count=7

• Rp=4000 Vg_ac=55 P_max=40 Vg_qui=-48 Vp_qui=300

• X_MIN=104 Y_MIN=62 X_SIZE=560 Y_SIZE=446 FSZ_X=1260 FSZ_Y=727 XYGrid=true

• showLoadLine=n showIp=y isDHT=n isPP=n isAsymPP=n showDissipLimit=y

• showIg1=y gridLevel2=y isInputSnapped=n

• XYProjections=n harmonicPlot=n dissipPlot=n

*---------------------------------------------------------------------------------- E1 7 0 VALUE={V(1,3)/KP*LOG(1+EXP(KP*(1/MU+(VCT+V(2,3))/SQRT(KVB+V(1,3)*V(1,3)))))} RE1 7 0 1G ; TO AVOID FLOATING NODES G1 1 3 VALUE={(PWR(V(7),EX)+PWRS(V(7),EX))/KG1} RCP 1 3 1G ; TO AVOID FLOATING NODES C1 2 3 {CCG} ; CATHODE-GRID C2 2 1 {CGP} ; GRID=PLATE C3 1 3 {CCP} ; CATHODE-PLATE RE2 2 0 1G EGC 8 0 VALUE={V(2,3)-VGOFF} ; POSITIVE GRID THRESHOLD GG 2 3 VALUE={(IGA+IGB/(IGC+V(1,3)))*(MU/KG1)*(PWR(V(8),IGEX)+PWRS(V(8),IGEX))} .ENDS *$

Once the mosfet is off, the tube pinches itself off. Wild guess, the cathode might pull itself up to +15.

My -----* suggests that a zener or something might be needed to protect the mosfet, but probably not.

The tube data sheet is not adequate to this application, so the tube should be tested. I wouldn't entitely trust the Spice model either. There's no heater!

OK, waiting on parts. Mikek

That would match with the 'DC 60 Hz' which is obtained by half-wave rectification.

Arie

Weird things can happen at very high voltages, e.g. the shielding effectiveness of the grid isn't really 100%, so the cathode current probably won't cut off altogether.

A small zener might be an excellent idea.

Cheers

Phil Hobbs

Hmm, we were about ready to try halfwave rectification and my son went a different route. I like the rest period between halfwave pulse, but don't know if it matters. Also, that could have been a mistake 60Hz of 120Hz from fuel scientists! Mikek

I think I need to add some safety, Can I ground the aluminum tube as shown in this drawing? Would it create an unwanted potential between the primary and secondary?

Thanks, Mikek

If the mosfet is open, where would any plate current go? But even a couple of uA wouldn't be a problem here.

There will certainly be some conduction between the cathode and the filament.

I guess the cathode could, maybe, rise to 100 volts or something, so a

24v zener to ground might be prudent.

Tube books don't usually address these subtelties.

Construction technique is the first thing to address concerning HV corona. No sharp points, no 90 degree bends, proper spacing etc. As a kind of double check - if it looks ugly it's probably bad HV construction. Either Kapton or teflon tape properly used is fine. Depends on what you've got on hand/ find easier to use/cost & availability considerations, etc. But don't rely on the tape to make up for bad construction. With the first proposal - a 1/4" center conductor inside an outer pipe, wrapping the center conductor with teflon tape for a snug fit is almost a "no-brainer" for getting the construction right. Where the string of 1.5meg resistors is concerned, there's a lot more chance of sharp points, sag, bent wires etc.

Ed

OK, thanks for that. I suspect first iteration will just be the resistor soldered as fat as possible with sharp points. But, if/when needed, I might try the brass tube idea. I have shown it here under the big red arrow!

I haven't solved how I'd attach the brass rod to the threaded post sticking out of the banana jack. Maybe a thru hole and put the nut on, if the threaded rod is long enough. Future developments. Thanks, Mikek

Somewhere in this thread John Larkin suggested a neon bulb to indicate HV is present. I said it probably would be enough of an attention getter. So, I put a nean tube and an LDR in shrink tube. When the neon tube comes on (about 52Vac) the LDR drops from 10s of MΩs to 7.5kΩ. Now my question is, How do I implement it? The with the 8 resistor voltage divider, seem we will run at 8KV always. So, I think I can connect it through a dropping resistor. If I scale the 47KΩ (120V) to 8kV, that says I need a 3.2MΩ resistor.

Is that how it should be done?

Then, when we go to a DC filtered supply, I'm not sure what to do, because we may repeat some pulsed experiments at lower kVolts. Then the neon bulb wouldn't strike. Can I put back to back

100v zeners across the neon bulb and lower the series resistance to the neon bulb, thinking is, it will strike at lower voltages, but at higher voltages the zeners keep the voltage drop across the neon at a safe level. I don't know what that safe level is.

Here's the circuit I put together, please critic. If it needs changes, fine. I thought the base voltage should be a little lower!

Thanks, Mikek

I'm wondering if fast edges are important, can a water bubble move through an emulsion the needed distance as fast as the rise time of a fast pulse! Ya, I know (needed) distance and (fast) pulse are unknowns. I've been concerned about rise and fall times, now wondering if it is less important. May even be, a fast rise time could be wasted because the water bubble has no time to get moving, before the pulse is steady state. Several papers say action only happens on rise and fall.

Thanks, Mikek

A water bubble won't move fast at all. Your de-emulsifying station will have to move the fluid past the electronces so that there are always a few droplets close to the electrode when the electric field is high enough to move them onto it.

This probably explains why you need a pulsed field - dirt and gunge in the fluid getting de-emulsified will also get stuck to the electrodes and will get stuck tighter than the water droplets. If you turn off the electric field some of the time, the stirring of the fluid will move the gunge away from the electrode far enough to let new water droplets in so that they can coalesce on the electrode when the field comes back.

If you leave the voltage on for long enough to let all the gunge get stuck onto the electrodes, it will block the water droplets

A neon bulb relaxation oscillator would be cool. Just a series resistor, a cap to ground, and the neon across the cap. It blinks, faster at higher voltage.

The neon-LDR thing is overkill if all you want to do is drive a remote LED or something.