pi power

[Qiaozhi]

I've long questioned whether flat-topping the coil current helps beyond a tiny bit. I have a plan to build a test circuit to objectively test this.

As a general observation, I haven't seen flat-topping make a noticeable difference. It would be good to have a definitive answer.

Most PI circuits that use a low coil resistance end up with roughly a sawtooth coil current (the early part of the exponential), where the sawtooth ramp follows di/dt=VTX/L, where VTX is the transmitter power supply voltage and L is the coil inductance. Ergo, if VTX is 12V and L=300uH then the peak current at 100us is 4A. The series resistance reduces this some, but the peak current is usually dominated by VTX/L. Inductance in coils is usually pretty tight, 2-3% is normal.

Tight tolerances are probably only true for professional metal detector manufacturers. Sod's Law says that a certain number of DIYers will make a mistake, and then we get the inevitable post stating that there was a flash/bang event, and the detector is no longer working.
I've also noticed that Eric tends to favour the use of a series resistor. It's saved me on a few occasions.

[deemon]

Coil field ( and its energy ) does depend on coil current , so if we have the same coil current before switch-off - our target response must be the same . But if we have a higher battery voltage with the same coil , then we'll need less time to reach this coil current value , so the TX pulse becomes shorter - and this short pulse won't be enough to "energize" a target with a high TC .... in this case higher battery voltage is worse .

[Tinkerer]

We could also look at the PI exciting pulse as the rate of change. Like:
25 turns at 1A for a 25cm diameter coil, about 25 Ampere-turns (AT) for 491 square centimeters of coil area or about 0.05AT per square centimeter.
Switched off to give a rate of change of 1A/us. Now try to increase this rate of change, and you will find with the usual peak Flyback voltage constraints, this is not so easy to do. So we need to find a compromise.

Compromise is the key word.

Now let`s say we increase the coil diameter to 40cm. The area increases to 1250 cm square so we need to increase the ampere turns by a factor of about 2.55 to obtain the same field density. So, lets say we keep the same amount of turns and apply 2.55A current. we got the same field density. How about the inductance and Flyback voltage? I let you do your own calculations, a very practical calculator app is: http://www.miscel.dk/MiscEl/MiscEl.zip

To get similar field strength proportions you will also need to keep the same rate of change in the switching, whereby you need to remember that the target behaves like an inductance. There you will find that the 3TC rule is well applied for charging during the ramp-up, but then you would need to measure the eddy currents also at that time. Or you can consider that these eddy currents will have to be killed by the switch off transient. (rate of change or di/dt) see post #12.

Simulations with LTSpice give you all these results with great accuracy, as long as you are careful to include all relevant parameters. Lots of simulations have been posted on the forum.

When you do your simulations, also look at the amount of energy stored in the TX coil, at the moment of switch off. All this energy will have to be dumped in the damping resistor.

[Carl-NC]

Good link thank you, another spin on that question is this. Couple ways we can set up our tx current.One way we program the series resistance so that the current in the coil builds exponentially, and we match that with our target's total charge time. 3tc-95% for example.So just as the current in the coil rises close to flat top we end the tx pulse. The other way we use more series resistance, so the current flat tops sooner. Then we maintain the constant flat top current until the 3tc-95% point of our target. Assuming all our components are rated for the currents. Is there appreciable gain from one method to the other,or does it boil down to average current through the coil?

If you want to guarantee that the target's turn-on eddies have died out, then you need to flat-top the TX current, and keep it flat-topped for 3-5 target-taus afterwards.

[Carl-NC]

If you terminate the TX-on pulse before the coil current has flattened out, eddy currents will still be present in the target. When the MOSFET is turned off, the magnetic field in the coil collapses. The eddy currents generated by this collapse will be in opposition to those flowing as a result of the TX-on pulse. So the idea of allowing the TX-on current to flat-top, is simply to allow the magnetic field collapse to fully stimulate the target. However, in practice, the benefits associated with flat-topping are small.

I've long questioned whether flat-topping the coil current helps beyond a tiny bit. I have a plan to build a test circuit to objectively test this.

There are also other [arguably more important] reasons for including a resistor in series with the coil. Firstly, it means that you don't need such a tight control on the coil parameters. If the series resistor wasn't there, the maximum coil current would be directly proportional to the coil resistance. Secondly, If you accidentally get a short-circuit across the coil pins, or the TX oscillator gets stuck in the on condition, the MOSFET will not let out the magic smoke.

Most PI circuits that use a low coil resistance end up with roughly a sawtooth coil current (the early part of the exponential), where the sawtooth ramp follows di/dt=VTX/L, where VTX is the transmitter power supply voltage and L is the coil inductance. Ergo, if VTX is 12V and L=300uH then the peak current at 100us is 4A. The series resistance reduces this some, but the peak current is usually dominated by VTX/L. Inductance in coils is usually pretty tight, 2-3% is normal.

[Brian Deese]

Picture

I am not sure my meaning is coming across. I follow everything you are saying,but I'm not sure we are on the same page. let me talk hypothetically for a minute. We know what magnetic field strength we need to stimulate our given target,in our given soil.We know all our coil specs so we can figure the current required to achieve the said field strength.So we set our series resistance to give us this field strength,so I can imagine that will be enough resistance that our flyback decay,and early sample time is a non-issue.

So maybe I have it backwards,and that is why I feel we not exactly eye to eye here. Maybe the field strength is harder to achieve than the fast flyback decay. So we have to set our series resistance to program the coil to the charge time of our target 3tc for 95% for example. Or set the resistance to get the most current through the coil that we can and still sample when we want to sample. So we are getting the most field strength that we can get.

One final example with some actual math,the gauss may be wrong came from a solenoid calc but bear with me. You come over to my shop and Im showing you my pi circuit. Its 300uH inductance coil,I want to find nickels,and gold rings so I set it up to find targets with a 20uS time constant.Yeah man I read on the geotech forum that R=L/t so I set my series resistance of my total tx circuit to 15 ohms. I set my tx pulse to 100uS so I can charge those nickels up right son! I use a 12 volt battery so I have 800mA of coil current,that gives me a B field gauss of 31.66. And you say hold up there son,what do you think you are doing using B field of 31.66 gauss? Dont you know all the iron in your blood is going to move into your toes. All you need is something more like 19.79 gauss b field strength. Oh ok man dont yell at me,I'll change to 24 ohm tx series resistance. Now I have 500mA coil current and my B field is 19.79 gauss. Then you say, look what I just did for you Deese man now your coil is super fast. I just set it up with a 12.5uS time constant. You can probably move that sample time in and get those thin gold rings now. Thanks Qiaozhi, you is da man.

I think what you are going to say is the early sample time is the imperator,and there is no such thing as too much field strength in a practical design.

[Qiaozhi]

Good link thank you, another spin on that question is this. Couple ways we can set up our tx current.One way we program the series resistance so that the current in the coil builds exponentially, and we match that with our target's total charge time. 3tc-95% for example.So just as the current in the coil rises close to flat top we end the tx pulse. The other way we use more series resistance, so the current flat tops sooner. Then we maintain the constant flat top current until the 3tc-95% point of our target. Assuming all our components are rated for the currents. Is there appreciable gain from one method to the other,or does it boil down to average current through the coil?

If you terminate the TX-on pulse before the coil current has flattened out, eddy currents will still be present in the target. When the MOSFET is turned off, the magnetic field in the coil collapses. The eddy currents generated by this collapse will be in opposition to those flowing as a result of the TX-on pulse. So the idea of allowing the TX-on current to flat-top, is simply to allow the magnetic field collapse to fully stimulate the target. However, in practice, the benefits associated with flat-topping are small.

There are also other [arguably more important] reasons for including a resistor in series with the coil. Firstly, it means that you don't need such a tight control on the coil parameters. If the series resistor wasn't there, the maximum coil current would be directly proportional to the coil resistance. Secondly, If you accidentally get a short-circuit across the coil pins, or the TX oscillator gets stuck in the on condition, the MOSFET will not let out the magic smoke.

[Brian Deese]

Good link thank you, another spin on that question is this. Couple ways we can set up our tx current.One way we program the series resistance so that the current in the coil builds exponentially, and we match that with our target's total charge time. 3tc-95% for example.So just as the current in the coil rises close to flat top we end the tx pulse. The other way we use more series resistance, so the current flat tops sooner. Then we maintain the constant flat top current until the 3tc-95% point of our target. Assuming all our components are rated for the currents. Is there appreciable gain from one method to the other,or does it boil down to average current through the coil?

[Qiaozhi]

So my question is for a given coil size, and inductance. What is a good rule of thumb for the Gauss specification? We can change the current in the coil with series resistance. What is the point where the law of diminishing returns takes over. If 5 amps does the job,whats the use using 10? There is plenty of discussion of tx time,and series resistance relating to time constants. The coil current/inductance vs Gauss must be equally important.

There is no definitive answer. As Eric points out, you also have to consider the energy stored in the magnetic field. A lot depends on what you're searching for. The requirements for gold nugget hunting are completely different to [say] a buried Russian tank.

http://www.geotech1.com/forums/showthread.php?22641-PI-Pulse-Voltage-More-Voltage-More-Depth-not-really&p=207878#post207878

[Brian Deese]

Only up to a certain point. Then you end up with more signal-to-ground noise, not to mention needing a bigger battery. Remember that the near-field strength falls off by the third power, so the target signal strength will fall off by the sixth power. In practical terms this means that you need to increase the transmitted field strength by 64x to double detection depth, and this will only be true in an air test, or ground with very low mineralization.

So my question is for a given coil size, and inductance. What is a good rule of thumb for the Gauss specification? We can change the current in the coil with series resistance. What is the point where the law of diminishing returns takes over. If 5 amps does the job,whats the use using 10? There is plenty of discussion of tx time,and series resistance relating to time constants. The coil current/inductance vs Gauss must be equally important.

[mikebg]

What will happen if I use a battery of 7.4 V (composed of two Li-ion cells) instead of three AA size cells?
I will outperform the design MP20 of Thomas Breuer. It runs for more than 70 hours on only three "AA" size alkaline batteries. Using NiCd or NiMH rechargeable batteries (3.6V) is also possible.

[Ferric Toes]

Another problem is getting rid of the energy stored in the magnetic field when the current is switched off at the end of each pulse. Higher pulse currents usually result in having to use longer delays before receiver sampling can take place. OK if you are looking for large conductive objects but no good for small nuggets.

[Qiaozhi]

Hello, dear friends all over the world! I'd like to talk with you on the issue of the pulse detector coil voltage, most pulse detector coil are driven by the 12 v, if we use 24 v and 36 v launch, the detector has higher detection range? Coil of the stronger the magnetic field intensity, the metal will enhance the reflected signal? Based on this problem, please put forward the Suggestions and comments, thank you

Only up to a certain point. Then you end up with more signal-to-ground noise, not to mention needing a bigger battery. Remember that the near-field strength falls off by the third power, so the target signal strength will fall off by the sixth power. In practical terms this means that you need to increase the transmitted field strength by 64x to double detection depth, and this will only be true in an air test, or ground with very low mineralization.

[Orbit]

Yes in my experience always!

[liudengyuan]

Hello, dear friends all over the world! I'd like to talk with you on the issue of the pulse detector coil voltage, most pulse detector coil are driven by the 12 v, if we use 24 v and 36 v launch, the detector has higher detection range? Coil of the stronger the magnetic field intensity, the metal will enhance the reflected signal? Based on this problem, please put forward the Suggestions and comments, thank you