A
Anonymous
Guest
Hi,
I think it
I think it
Find's Treasure Forums
Welcome to Find's Treasure Forums, Guests!
You are viewing this forums as a guest which limits you to read only status.
Only registered members may post stories, questions, classifieds, reply to other posts, contact other members using built in messaging and use many other features found on these forums.
Why not register and join us today? It's free! (We don't share your email addresses with anyone.) We keep email addresses of our users to protect them and others from bad people posting things they shouldn't.
Click here to register!
Need Support Help?
Cannot log in?, click here to have new password emailed to you
You are using an out of date browser. It may not display this or other websites correctly.
You should upgrade or use an alternative browser.
You should upgrade or use an alternative browser.
A
Anonymous
Guest
Thomas, thanks for your very informative post. We're looking forward to hearing more from you.
On your second channel 2 graph, you show constant current, flat as the kitchen floor, during the transmit "pulse". That doesn't seem to be quite in agreement with the channel 1 and 3 data. In particular, channel 3 shows a change in induced voltage which would have to be accompanied by a change in current.
So, here's my question. Is the current really that flat; and, if so, how did you keep it that flat?
In principle, it would be possible to keep the current pretty darn flat either by active regulation, or by introducing an active negative resistance network in series with the coil which would balance out its resistance. With an induction balance receiver coil, the receiver could be turned on during this period to look at target signals. ......It seems to me that it would be easier to just short out the coil and demodulate reactive components to balance out the remaining slope during the transmit "pulse", as I described in the thing about CCPI, but actively flattening the transmit "pulse" may also be worthwhile.
You described a low power PI unit which uses symmetrical pulses at fairly high frequency, somewhat similar to the Fisher Impulse. You describe an apparent problem with increased noise. I suspect that what you are observing is not an actual increase in noise (although that's possible, if you happen to be camped on top of an interfering signal such as LORAN-C), but a degradation of S/N ratio. When you have symmetrical pulses, and start squashing them up close together, then the target decay from the positive pulses smears into the decay from the negative pulses, partially cancelling them. The effect is most pronounced on high-conductivity targets, but if you're running at a fundamental frequency of 10 kHz, then any target with a time-constant greater than 50 microseconds is going to suffer reduced sensitivity. If your circuit topology permits it, you might try disabling the negative half of the transmit/flyback cycle, and see what happens in the receiver. I predict that target detectability will improve on all but the smallest targets. ....If you do this, I hope you'll let us know how the experiment turns out.
Re: aluminum Litz: easy to describe, difficult to solder to, nearly impossible to buy! For a given conductivity, aluminum doesn't save 50% weight over copper, the improvement is more like 30% although I don't remember the exact number.
Thanks again for your post.
--Dave J.
On your second channel 2 graph, you show constant current, flat as the kitchen floor, during the transmit "pulse". That doesn't seem to be quite in agreement with the channel 1 and 3 data. In particular, channel 3 shows a change in induced voltage which would have to be accompanied by a change in current.
So, here's my question. Is the current really that flat; and, if so, how did you keep it that flat?
In principle, it would be possible to keep the current pretty darn flat either by active regulation, or by introducing an active negative resistance network in series with the coil which would balance out its resistance. With an induction balance receiver coil, the receiver could be turned on during this period to look at target signals. ......It seems to me that it would be easier to just short out the coil and demodulate reactive components to balance out the remaining slope during the transmit "pulse", as I described in the thing about CCPI, but actively flattening the transmit "pulse" may also be worthwhile.
You described a low power PI unit which uses symmetrical pulses at fairly high frequency, somewhat similar to the Fisher Impulse. You describe an apparent problem with increased noise. I suspect that what you are observing is not an actual increase in noise (although that's possible, if you happen to be camped on top of an interfering signal such as LORAN-C), but a degradation of S/N ratio. When you have symmetrical pulses, and start squashing them up close together, then the target decay from the positive pulses smears into the decay from the negative pulses, partially cancelling them. The effect is most pronounced on high-conductivity targets, but if you're running at a fundamental frequency of 10 kHz, then any target with a time-constant greater than 50 microseconds is going to suffer reduced sensitivity. If your circuit topology permits it, you might try disabling the negative half of the transmit/flyback cycle, and see what happens in the receiver. I predict that target detectability will improve on all but the smallest targets. ....If you do this, I hope you'll let us know how the experiment turns out.
Re: aluminum Litz: easy to describe, difficult to solder to, nearly impossible to buy! For a given conductivity, aluminum doesn't save 50% weight over copper, the improvement is more like 30% although I don't remember the exact number.
Thanks again for your post.
--Dave J.
A
Anonymous
Guest
Hi Dave,
thanks for your quick response. Just a few comments, it is already late and I will continue tomorrow.
You wrote:
>> On your second channel 2 graph, you show constant current, flat as the kitchen floor, during the transmit "pulse". That doesn't seem to be quite in agreement with the channel 1 and 3 data. In particular, channel 3 shows a change in induced voltage which would have to be accompanied by a change in current.
So, here's my question. Is the current really that flat; and, if so, how did you keep it that flat? <<
You already answered your questions yourself ;-) Yes, the current is really that flat, and yes, this has been achieved by by a regulation. Take a close look at the channel 1 trace of the second set: the little negative offset is the voltage drop caused by the constant current and the resistance of the TX coil. Quite simple, but a little bit tricky circuit. I can post it if you
thanks for your quick response. Just a few comments, it is already late and I will continue tomorrow.
You wrote:
>> On your second channel 2 graph, you show constant current, flat as the kitchen floor, during the transmit "pulse". That doesn't seem to be quite in agreement with the channel 1 and 3 data. In particular, channel 3 shows a change in induced voltage which would have to be accompanied by a change in current.
So, here's my question. Is the current really that flat; and, if so, how did you keep it that flat? <<
You already answered your questions yourself ;-) Yes, the current is really that flat, and yes, this has been achieved by by a regulation. Take a close look at the channel 1 trace of the second set: the little negative offset is the voltage drop caused by the constant current and the resistance of the TX coil. Quite simple, but a little bit tricky circuit. I can post it if you
A
Anonymous
Guest
I checked your numbers on aluminum vs. copper, and then calculated the same thing using wire tables rather than material properties. Yep, you're right, I was wrong. For given length and resistance, aluminum is half the weight.
For equal resistance, aluminum is 2 gages AWG/B&S fatter. For instance, you'd replace 23 AWG copper with 21 AWG aluminum. There's no corresponding simple rule for metric gages since they're not scaled logarithmically.
Crimping is not a reliable way to connect to aluminum, nor is it a reliable way to connect to Litz. If I were going to connect to aluminum on a production basis, I'd use ultrasonic welding or ultrasonic soldering, either of which requires expensive specialized equipment.
With practice, one can get good enough at soldering aluminum with special aluminum solder to construct prototypes.
In principle it would be possible to plate aluminum wire with a metal more compatible with soldering processes and which would not form unstable intermetallics. However, I don't know any such process.
Aluminum tubing is a standard industrial commodity, and might be a practical substitute for wire when constructing large searchcoils.
AN INVITATION: does anyone reading this know a reliable and inexpensive way of making connections to aluminum? Maybe an improved "aluminum solder"? A simple method of electroplating wire ends? PLEASE POST IT.
If you're willing to post your current regulator circuit, Thomas, I and no doubt several other people would be pleased to take a look at it. Who knows, it might lead to something that neither of us has thought of before.
--Dave J.
For equal resistance, aluminum is 2 gages AWG/B&S fatter. For instance, you'd replace 23 AWG copper with 21 AWG aluminum. There's no corresponding simple rule for metric gages since they're not scaled logarithmically.
Crimping is not a reliable way to connect to aluminum, nor is it a reliable way to connect to Litz. If I were going to connect to aluminum on a production basis, I'd use ultrasonic welding or ultrasonic soldering, either of which requires expensive specialized equipment.
With practice, one can get good enough at soldering aluminum with special aluminum solder to construct prototypes.
In principle it would be possible to plate aluminum wire with a metal more compatible with soldering processes and which would not form unstable intermetallics. However, I don't know any such process.
Aluminum tubing is a standard industrial commodity, and might be a practical substitute for wire when constructing large searchcoils.
AN INVITATION: does anyone reading this know a reliable and inexpensive way of making connections to aluminum? Maybe an improved "aluminum solder"? A simple method of electroplating wire ends? PLEASE POST IT.
If you're willing to post your current regulator circuit, Thomas, I and no doubt several other people would be pleased to take a look at it. Who knows, it might lead to something that neither of us has thought of before.
--Dave J.
A
Anonymous
Guest
Hi Thomas,
Glad to see you join the discussion. First, I want to thank you for the information all have provided recently, especially your scope pictures. As they say, a picture is worth a thousand words and your posted scope readings have really helped.
I have a question as to whether you have the same two sets of pictures of the transmit, receive and current signals with no target of any kind present (air test), and also, where the coils are resting on the ground with no target present, except the ground mineralization.
Obviously, the ground signals can vary from area to area, but it would be interesting to see how they vary with the different transmit pulse widths.
If you have run theses tests, I hope you would be willing to post them. I can't speak for anyone else, but I sure would like to see them.
You see, I graduated from college about 30 years ago, and in my field of industrial electronics, I am concerned about system controls as they relate to proper operation of equipment (PLC)'s, etc. As such, I have had little need for remembering the details of higher math involved or magnetic theory, let alone some of the terminology that has been discussed lately. However, I do understand what the pictures are displaying and they do help clear the cobwebs out of my mind as I try to remember the theory part.
I hope you have run the scope tests I mentioned because I feel they would add to the overall view of what is happening under different conditions commonly encountered.
Finally, I would love to see the same tests using a gold nugget instead of a silver ring as a test target. Have you ran this test? Again, I feel it would be helpful to display the results of a lower conductive target as it compares to a more conductive target like the ring.
Anyway, those are my thoughts. Once again, I want to thank you and all others who are actively participating in the latest round of discussions. These discussions have been very informative and enlightening.
Reg
Glad to see you join the discussion. First, I want to thank you for the information all have provided recently, especially your scope pictures. As they say, a picture is worth a thousand words and your posted scope readings have really helped.
I have a question as to whether you have the same two sets of pictures of the transmit, receive and current signals with no target of any kind present (air test), and also, where the coils are resting on the ground with no target present, except the ground mineralization.
Obviously, the ground signals can vary from area to area, but it would be interesting to see how they vary with the different transmit pulse widths.
If you have run theses tests, I hope you would be willing to post them. I can't speak for anyone else, but I sure would like to see them.
You see, I graduated from college about 30 years ago, and in my field of industrial electronics, I am concerned about system controls as they relate to proper operation of equipment (PLC)'s, etc. As such, I have had little need for remembering the details of higher math involved or magnetic theory, let alone some of the terminology that has been discussed lately. However, I do understand what the pictures are displaying and they do help clear the cobwebs out of my mind as I try to remember the theory part.
I hope you have run the scope tests I mentioned because I feel they would add to the overall view of what is happening under different conditions commonly encountered.
Finally, I would love to see the same tests using a gold nugget instead of a silver ring as a test target. Have you ran this test? Again, I feel it would be helpful to display the results of a lower conductive target as it compares to a more conductive target like the ring.
Anyway, those are my thoughts. Once again, I want to thank you and all others who are actively participating in the latest round of discussions. These discussions have been very informative and enlightening.
Reg
A
Anonymous
Guest
Thomas
I would describe the results of your experiment in slightly different terms.
Eddy currents flow in the target both at turn on and turn off of the coil. At turn on, the target current flows in one direction, and at turn off, it flows in the opposite direction.
In channel 3 (receive voltage) we can see that the time constant of the target is roughly 50 usec.
In the case with the long on time, the coil current reaches the desired level after 60 usec and is then held constant for more than 4 target time constants. During this time the turn on eddy current decays. You can see the signal from this current decaying in the final graph in the period from 60 to 250 usec. It has effectively decayed to zero by the time turn off occurs. During the time that the coil current is constant, the magnetic field of the coil is approximately constant, but the magnetic field through the target is not. The eddy current in the target generates a magnetic field that is opposed to the field of the coil. The net field through the target does not reach its maximum until the eddy current has decayed to zero. After the coil is turned off you see a peak receive voltage of about 3 mV.
In the case with the short on time, the coil is only on for about one target time constant. So the turn on eddy current is still flowing in the target, and the magnetic field through the target has not reached its maximum possible level yet. (You could say the target is not fully charged yet.) So when the coil is turned off the receive voltage peaks at about 1.8 mV rather than the 3 mV that is achieved when the on time is long enough.
For anyone who does not want to think about the magnetic field, think about it this way. After turnoff we want to measure the current flowing in one direction in the target. But during the first couple of time constants after turn on there is a current flowing the wrong way in the target. If the coil is turned off too early this wrong way current gets subtracted from the current we want to measure. After 3 to 5 time constants the wrong way current has decayed enough to not affect the result.
Robert
I would describe the results of your experiment in slightly different terms.
Eddy currents flow in the target both at turn on and turn off of the coil. At turn on, the target current flows in one direction, and at turn off, it flows in the opposite direction.
In channel 3 (receive voltage) we can see that the time constant of the target is roughly 50 usec.
In the case with the long on time, the coil current reaches the desired level after 60 usec and is then held constant for more than 4 target time constants. During this time the turn on eddy current decays. You can see the signal from this current decaying in the final graph in the period from 60 to 250 usec. It has effectively decayed to zero by the time turn off occurs. During the time that the coil current is constant, the magnetic field of the coil is approximately constant, but the magnetic field through the target is not. The eddy current in the target generates a magnetic field that is opposed to the field of the coil. The net field through the target does not reach its maximum until the eddy current has decayed to zero. After the coil is turned off you see a peak receive voltage of about 3 mV.
In the case with the short on time, the coil is only on for about one target time constant. So the turn on eddy current is still flowing in the target, and the magnetic field through the target has not reached its maximum possible level yet. (You could say the target is not fully charged yet.) So when the coil is turned off the receive voltage peaks at about 1.8 mV rather than the 3 mV that is achieved when the on time is long enough.
For anyone who does not want to think about the magnetic field, think about it this way. After turnoff we want to measure the current flowing in one direction in the target. But during the first couple of time constants after turn on there is a current flowing the wrong way in the target. If the coil is turned off too early this wrong way current gets subtracted from the current we want to measure. After 3 to 5 time constants the wrong way current has decayed enough to not affect the result.
Robert
A
Anonymous
Guest
Dave,
I have sucessfully soldered small ali bits together using just normal but sturdy very hot soldering iron with 60/40 lead resin cored solder,
you need to scrape clean & cover the joint area in a small pool of household lubricating oil to prevent oxidisation, then keep scraping the iron tip thro the oil against the ali joint feeding solder at same time.
makes a bit of smoke and fumes but does work!
Rob
I have sucessfully soldered small ali bits together using just normal but sturdy very hot soldering iron with 60/40 lead resin cored solder,
you need to scrape clean & cover the joint area in a small pool of household lubricating oil to prevent oxidisation, then keep scraping the iron tip thro the oil against the ali joint feeding solder at same time.
makes a bit of smoke and fumes but does work!
Rob
A
Anonymous
Guest
Robert,
Thanks for your comments. Please see my new post
Thanks for your comments. Please see my new post
A
Anonymous
Guest
Hi Reg,
You wrote:
>> I have a question as to whether you have the same two sets of pictures of the transmit, receive and current signals with no target of any kind present (air test), and also, where the coils are resting on the ground with no target present, except the ground mineralization. <<
No target, no signal ;-) Note that the two coils are balanced, so no voltage will be induced in the RX coil when there is no target present (channel 3). Channels 1+2 show just reference signals which are rather independend of external influences, especially when only small objects are used.
Concerning ground mineralization, these signals are very weak compared to the signals of the metal objects used in this experiment. I studied this, too, but the setup is a bit different. I will post info on this later.
About the nugget test, my experiment is only to demonstrate some PI basics. Maybe we can do some tests with low conductive objects later.
Thomas
You wrote:
>> I have a question as to whether you have the same two sets of pictures of the transmit, receive and current signals with no target of any kind present (air test), and also, where the coils are resting on the ground with no target present, except the ground mineralization. <<
No target, no signal ;-) Note that the two coils are balanced, so no voltage will be induced in the RX coil when there is no target present (channel 3). Channels 1+2 show just reference signals which are rather independend of external influences, especially when only small objects are used.
Concerning ground mineralization, these signals are very weak compared to the signals of the metal objects used in this experiment. I studied this, too, but the setup is a bit different. I will post info on this later.
About the nugget test, my experiment is only to demonstrate some PI basics. Maybe we can do some tests with low conductive objects later.
Thomas
A
Anonymous
Guest
Hi Thomas,
Oops, I either missed or failed to understand the info about the secondary compensation loop to zero the signal in the primary receive loop. Hopefully, this will teach me to read more thoroughly.
Thanks for the fast response.
Reg
Oops, I either missed or failed to understand the info about the secondary compensation loop to zero the signal in the primary receive loop. Hopefully, this will teach me to read more thoroughly.
Thanks for the fast response.
Reg
A
Anonymous
Guest
The simplest way to terminate aluminum wires is with a spring loaded cage clamp. Crimping methods usually degrade over time cause of the differential coefficients of expansion between aluminum and copper - after a number of hot/cold cycles they work loose. Crimps with spring loaded compliance hold up very well and are generally gas tight and low resistance.