New PI technology proposals.

[Carl-NC]

The center-tapped coil should be damped on each half with individual damping resistors instead of my proposal. This will give a faster decay.

Yes, that's what I'm doing.

The coil wire should have an insulation to keep the interwire capacitance low. Then the capacitive coupling between the half coils will be low.

I'm using PVC insulated wire. We have teflon as well, but I didn't have a spool handy. That's also why I wound a mono coil, for a 1-to-1 comparison.

As long as the ringing is totally in-phase, this shouldn't be a problem with differential signalling. But best would be to avoid any ringings.

Well, that's true only if the ringing perfectly cancels. But it doesn't.

If you are interested on some amazing results, please visit the Oz-forum. I just posted there.

Well, I looked but it must be in one of those hidden forums. Post the results here so everyone can see them.

- Carl

[Carl-NC]

If flyback can be damped faster with lots of resistors in coil, why not make a coil with flat spiral windings, use wire with no insulation, embedded in a fairly conductive layer of graphite? Infinite number of resistors!

Now that's clever.

[Carl-NC]

Since you were able to obtain a 1 us improvement in discharge time using two damping resistors, I suspect that the effective value of these two resistors in series across 10 turns each was a little higher than a single damping resistor across 20 turns. What were the values of the single Rd and the dual Rds?

340 ohms for the 300uH mono, and 235 ohms each for the CT coil.

What if you damped the coils at 5 turns each effectivly having the coil taps connected to 4 series damping resistors? I suspect that the combined value of this 4-resistor set might be a little higher than with 2 resistors in series thus further improving the discharge time.

I guess if you extended this to it's logical conclusion, you could have a damping resistor on each turn. That might really damp the flyback?

I think you'll have a rapidly diminishing improvement that isn't worth the extreme pain of building such a coil!

Then we could know the true role of the flyback pulse in (1) stimulating the target, (2) being a necessary artifact that must be quickly dispensed with or (3) if it produces some other measurable result related to detecting the target?

I think all this can be determined with an IB coil pair, and maybe even with an ordinary mono coil.

- Carl

[Aziz]

Hi Carl,

a MONO coil is divided into 4 sections. Each of the parts are individually critical damped. A second standard MONO coil is compared to the first one.
The coil is driven as usual on both ends (not center-tapped).

The following picture (1) shows the damping voltages with no target presented.
The second picture (2) shows the comparison of the target/no-target response.

It could be useful to find the reason for the anomaly coil current with the standard coil, which delays the decay curve. This is totally absent with the bbsailor's proposal. I am not sure, whether this effect can be measured and verified in a real practical configuration.

Aziz

[Aziz]

I think, the more coil splits with his each individual coil damping resistors are made, the more coil voltage we will have and a faster damping time. The MOSFET will breakdown anyway and will add some delays as a consequence of this.

It is very remarkable, that despite of the same transmit pulse energy (same coil current, same configuration except the coils) the flyback voltage increases with the splitted coil.

Aziz

[ZED]

Aziz i couldnt find any archived pics of any use,this anomoly that you speak of, it sounds like the flyback voltage being clipped by the diode in the mosfet,the timing looks about right,what do you think,maybe try a mosfet without a diode.

Zed

[ZED]

Carl the two receive coils are rectangular in shape with a ratio of 3 to 1,they are wound opposite to one another and connected in series and placed in a 18" transmit loop,the transmit loop has an inductance of less than 1 u/h and a resistance of less than 10 milli ohm [its half inch copper tubing] .The two receive coils are adjusted within the transmit loop so as to cancle out any effects that result from any coupling with the transmit loop,this can be done with a scope,this results in a flat line during a transmit.As targets are placed at or near a receive coil a signal can clearly be seen.The coil current rises to maximium of 45 amps within 10 u/s,this high slew rate is higher than the slew rate at switch off of many pi detectors.
All this was purely experimental,i wasnt attemting to design or build a detector,i was after imformation and data.

Zed

[Aziz]

Hi Zed,

Aziz i couldnt find any archived pics of any use,this anomoly that you speak of, it sounds like the flyback voltage being clipped by the diode in the mosfet,the timing looks about right,what do you think,maybe try a mosfet without a diode.

Zed

Finally, I found the reason for this anomaly:
Coils effective capacitance!

To test this, I have excluded the switching electronics (MOSFET and Diode) and replaced this with an ideal voltage controlled switch (Ron = 1e-12 Ohm, Roff = 1e+12 Ohm). Then I reduced the normal MONO coils capacitance from 300 pF to 30 pF. I have lower anomalies and faster decay then.

The cascaded coil damping method is in this ideal model even better.

Interesting result is the Voltage-time-product (see pictures above):
The integral U(t)*dt (t->infinite for one pulse) for both damping voltages seems to be equal, in which it represents probably the exposed magnetic energy per ampere.

The anomaly effect should be observable in real world. The more the coils capacitance, the better the observations should be.

Aziz

[Carl-NC]

Carl the two receive coils are rectangular in shape with a ratio of 3 to 1,they are wound opposite to one another and connected in series and placed in a 18" transmit loop,the transmit loop has an inductance of less than 1 u/h and a resistance of less than 10 milli ohm [its half inch copper tubing] .The two receive coils are adjusted within the transmit loop so as to cancle out any effects that result from any coupling with the transmit loop,this can be done with a scope,this results in a flat line during a transmit.As targets are placed at or near a receive coil a signal can clearly be seen.The coil current rises to maximium of 45 amps within 10 u/s,this high slew rate is higher than the slew rate at switch off of many pi detectors.

I think that explains the strong target signal during TX-ON!!

[Aziz]

Hi Carl,

could improve a twisted-pair cable (with insulation) wound to a coil and then made to a center-tapped coil the differential signal?

Aziz

[ZED]

whoops

In post 49 i stated that at the start of the transmit there was no flyback signal,i now believe this to be wrong.With the fast current growth upto 45 amps within 10 u/s there must be a high voltage signal [flyback or flyup ] induced in the receive coils at startup,because the two receive coils are inductively balanced it is not seen in the picture.
Sorry Aziz!

Zed

[KingJL]

Hey Aziz:

If flyback can be damped faster with lots of resistors in coil, why not make a coil with flat spiral windings, use wire with no insulation, embedded in a fairly conductive layer of graphite? Infinite number of resistors!

Maybe already done....

Cheers,

-SB

Yeah, and if done very carefully, you could accomplish shielding at the same time! Two birds with one stone... so to speak!!!
J. L. King

[hobbes_lives]

If flyback can be damped faster with lots of resistors in coil, why not make a coil with flat spiral windings, use wire with no insulation, embedded in a fairly conductive layer of graphite? Infinite number of resistors!

That's an interesting idea. Perhaps use an etched PCB coil painted with graphite?

Yeah, and if done very carefully, you could accomplish shielding at the same time! Two birds with one stone... so to speak!!!
J. L. King

I'd think the graphite would have to be highly conductive for shielding, but less conductive for the inter-winding resistance. That balance may be hard to accomplish, but perhaps possible.

[simonbaker]

Spice files

Hi Aziz:

Sorry to keep bothering you.

Can you post the latest ".asc" LTSpice file that you are using (and any special components needed).

I have trouble understanding some details from the pictures.

Thanks!

-SB

[KingJL]

I'd think the graphite would have to be highly conductive for shielding, but less conductive for the inter-winding resistance. That balance may be hard to accomplish, but perhaps possible.

That was with tongue-in-cheek...so to speak!