pi power

[Teleno]

Where do you get the 1 ohm in the power source?

We don't know the true value so I just plugged in a sensible estimate.

[Teleno]

High Eric,

I didn't see the voltage so 12v gives much the same as your figures, 3.39A and 3.66A. As you say, not enough to make a significant improvement.

That's 8% more current. I would say it's a noticeable improvement. Power in the coil improves by 16,5% .

[ivconic]

Qiaozhi: "...Without seeing the full circuit and more details of the detector settings..."

Qiaozhi can't you recognize it? It is FelezJoo PI from another topic.
.................................................. .................................................. .....................................

Crane: "...Assuming 10v then I get around 2.83A peak coil current with one IRF840 and 3.05A peak current with
two 840s in parallel but it becomes harder to turn the FETs on as the resistance drops so it's wise to
use a push pull circuit to drive the FET otherwise it spends the early part of the pulse acting as a
resistor and this can influence the result..."

As Eric later pointed; it is 12v supply. But the rest of your post is very interesting to me!
.................................................. .................................................. .....................................

Eric: "...My conclusion is that it is not enough of a difference to make the significant improvement you suggest. There must be another factor..."

Indeed! I was thinking the same. It couldn't be just from a bit more power. Must be something else colaterall that is missed on first view.
.................................................. .................................................. .....................................

Teleno: "...That's a 7% increase in current and a 14% increase in power..."

In terms of Pulse Induction; 14% increase of power is not trivial improvement! It explains some of the differences which i saw on the bench.
.................................................. .................................................. .....................................

Eric: "...A simple solution to increase the current is to widen the TX pulse to 195uS. This would give 3.7A with the single IRF840,
which is the same as the parallel setup at 160uS..."

But two transistors on the place will be cooler than one i suppose? And than let's see what we will gain if pulse is widen to 195uS with 2 transistor setup?
.................................................. .................................................. .....................................

Many thanks for the feedback guys!

[Qiaozhi]

Qiaozhi: "...Without seeing the full circuit and more details of the detector settings..."

Qiaozhi can't you recognize it? It is FelezJoo PI from another topic.

In that case we cannot see what's happening with the main sample pulse, as it's hidden in software.

[crane]

That's 8% more current. I would say it's a noticeable improvement. Power in the coil improves by 16,5% .

8% more peak coil current would do very little to improve performance, some of the Minelab mods increase it by 20% and you have to use a bit of imagination to notice the difference, it makes the ground noise much worse in some cases though.

Like Eric said we could just increase the pulse length. Or we could wind a lower resistance coil and use a heavier cable. We would then eventually get to the point where two parallel FETs might actually do something significant. See the SD2000 circuit.

[Teleno]

8% more peak coil current would do very little to improve performance, some of the Minelab mods increase it by 20% and you have to use a bit of imagination to notice the difference, it makes the ground noise much worse in some cases though.

Minelab's niche is gold nuggets in heavy magnetic ground, not what ivconic is doing I assume.

In normal applications, the actual improvement would depend on what your starting point is. If your driving circuit can take the increase without MOSFET avalanche then most targets will benefit.

[crane]

Minelab's niche is gold nuggets in heavy magnetic ground, not what ivconic is doing I assume.

In normal applications, the actual improvement would depend on what your starting point is. If your driving circuit can take the increase without MOSFET avalanche then most targets will benefit.

But it doesn't make any difference, Minelab PI detectors are also used for relic hunting and they clamp the spike to 180v. The IRF840 avalanches at 500v so the higher avalanche point is a bit meaningless unless you are chasing a shorter delay. Knowing this though ML still clamp the "fine gold" timing to 180v and this timing can detect gold noticeably smaller than a grain.

It must be obvious they have a reason for doing this.

[Teleno]

The IRF840 avalanches at 500v so the higher avalanche point is a bit meaningless unless you are chasing a shorter delay.

For a 100us time constant, for example, doubling the flyback voltage increases target response by 3%.

Knowing this though ML still clamp the "fine gold" timing to 180v and this timing can detect gold noticeably smaller than a grain.

It must be obvious they have a reason for doing this.

My guess is this clamping has more to do with EMF and safety regulations than technical reasons.

What increases response in grain sized targets is early sampling, which forces the use of smaller currents and well balanced Tx and Rx coils. Performance for relic hunting in such a detector, however, will be worse.

[Altra]

My guess is this clamping has more to do with EMF and safety regulations than technical reasons.

The 180v clamp is used to protect the n and p channel fets on the preamp input. It has other benefits also like lowering the power dissipation in the damping resistors.

[Teleno]

The 180v clamp is used to protect the n and p channel fets on the preamp input. It has other benefits also like lowering the power dissipation in the damping resistors.

Hardly necessary. Lots of us hve built the MPP, Barracuda etc. with 500V with no preamp or dissipation problems.

It must be something else.

[Altra]

I was referring to the preamp gating mosfets on this schematic, not internal to the op amp.

http://www.geotech1.com/forums/showt...000-schematics

[Ferric Toes]

I still question the sledgehammer approach of more current and higher voltage Mosfets when trying to improve the sensitivity, particularly to small targets. It's all about S/N ratio at the end of the day and that is one way to do it, but with the increasing risk of high voltage breakdown somewhere. I have had internal sparking in a coil from the live end to the shielding as well as other insulation problems by going this route.

Much better to have a low pulse current of 1A or less and bring the S/N back up by pulsing and sampling more often and slowing down the integrator to give the same response time. You end up with the same current draw from the power supply; keep outside of Mosfet avalanche even with 250 - 300V devices, and most importantly, because there is less magnetic field energy to dissipate on each TX pulse, you can sample earlier. That is where big gains are made on objects with fast decay TC's. Every uS helps when you are climbing up a steep exponential curve.

Eric.

[ivconic]

I still question the sledgehammer approach of more current and higher voltage Mosfets when trying to improve the sensitivity, particularly to small targets. It's all about S/N ratio at the end of the day and that is one way to do it, but with the increasing risk of high voltage breakdown somewhere. I have had internal sparking in a coil from the live end to the shielding as well as other insulation problems by going this route.

Much better to have a low pulse current of 1A or less and bring the S/N back up by pulsing and sampling more often and slowing down the integrator to give the same response time. You end up with the same current draw from the power supply; keep outside of Mosfet avalanche even with 250 - 300V devices, and most importantly, because there is less magnetic field energy to dissipate on each TX pulse, you can sample earlier. That is where big gains are made on objects with fast decay TC's. Every uS helps when you are climbing up a steep exponential curve.

Eric.

I am also not fan of lot of amperes in coil too. But in this particular case i did that because of the 1x1 meter coil which i made. It has 9 windings of 1mm litz, pvc coated cable. Overall it measures as i noted; 0.350mH and somewhat 1.5R with cable.
Such coil presents serious "drain" for only one fet, so as i noticed. Fet can get very hot on some adjustments. Especially on wider pulses. And that's what i want with that coil; wider pulses. I will search for medium sized and larger objects in soil with such coil.
So adding another fet was just sort of blind guess. But it was lucky guess, since now i have significantly better behavior.
I intend to inspect this further. Problem is in coil; it is too large to be tested indoor and now these days it is raining. Must wait.

[Teleno]

I still question the sledgehammer approach of more current and higher voltage Mosfets when trying to improve the sensitivity, particularly to small targets. It's all about S/N ratio at the end of the day and that is one way to do it, but with the increasing risk of high voltage breakdown somewhere. I have had internal sparking in a coil from the live end to the shielding as well as other insulation problems by going this route.

Much better to have a low pulse current of 1A or less and bring the S/N back up by pulsing and sampling more often and slowing down the integrator to give the same response time. You end up with the same current draw from the power supply; keep outside of Mosfet avalanche even with 250 - 300V devices, and most importantly, because there is less magnetic field energy to dissipate on each TX pulse, you can sample earlier. That is where big gains are made on objects with fast decay TC's. Every uS helps when you are climbing up a steep exponential curve.

Eric.

Another approach is to halve the no. of turns in the Tx coil and double the current. The Ampere-turns remain the same, but the coil capacitance, inductance and resistance get divided by 4. Transient decays faster, allowing for earlier sampling. If the Tx period is also halved then the power consumption remains the same.

[Ferric Toes]

Another approach is to halve the no. of turns in the Tx coil and double the current. The Ampere-turns remain the same, but the coil capacitance, inductance and resistance get divided by 4. Transient decays faster, allowing for earlier sampling. If the Tx period is also halved then the power consumption remains the same.

Yet another way is to centre tap a standard coil and drive the tap and one end with the TX and have the full coil as RX. That way you get the benefit of 4 x less inductance etc for the TX but don't lose out on the RX side.

Eric.