The Moodz detector lives !

[BILLY]

Moodz was led astray by Doug (the mouthpiece of the QED) and was told by Doug that he was going to have the best detector ever made. Poor Moodz I am sure is rethinking his mentor Doug. So to answer your question, Moodz has nothing that is not already out there much cheaper. Sorry if the truth hurts.

[crane]

...so I played with the timings a bit and the ferrite rejection is good ( in terms of audio output ... ) and the target sensitivity is still good also.

https://www.youtube.com/watch?v=UHBV2hZrivo

In this video the ferrite is right on the coil ... hence the massive changes in X signal ... however the R signal is still very sensitive.

Moodz says "despite several volts of change on the sampled waveform we are getting very little audio change". He appears to be saying he is actually sampling the several volts of change, or has he delayed the sample until after the ferrite's (x) signal has almost settled?

It would be nice if he could clear this up.

[moodz]

... well its exactly as I said ... Ferrite does not impact target sensitivity ... or ferrite rejection is good. Take your pick :-)

[6666]

I have grown tired of secret squirrel detector projects so very shortly I will gather the documents and publish my design for a kickass PI detector.

Are there any documents that can be published yet ? all watching with great interest.

[crane]

... well its exactly as I said ... Ferrite does not impact target sensitivity ... or ferrite rejection is good. Take your pick :-)

So you don't wish to say how you reject the ferrite?

A conventional PI ignores the ferrite X signal by delaying the early sample until after the spike has fully decayed. You though say you are sampling the ferrite X signal but are somehow ignoring it??

I'm sure others here are also confused.

Do you have a video showing the viscous (VRM) ground signal?

[Davor]

Sorry to disappoint you, but X signal is not present in delay. Only fast decaying R components, of which the strongest one is the coil's own R decay.

[crane]

Sorry to disappoint you, but X signal is not present in delay. Only fast decaying R components, of which the strongest one is the coil's own R decay.

And I'm sorry to disappoint you.

Pulse induction's main claim to fame is that it doesn't have to deal with the very large X component that plagues vlf detectors and most other designs that transmit continuous wave forms. The X component is present during transmission or during any period where the tx field is changing, which obviously means you will be sampling X if you sample any part of your "coil's own R decay" and this includes any part of the spike that hasn't decayed to zero.

[Davor]

... which results in responses far below noise floor at any reasonable sampling point. But in essence and strictly speaking you are right - while there is any significant voltage at Tx coil, the X component follows verbatim. I'm also right because the Tx coil voltage after flyback consists only of exponentially decaying voltage due to the Tx coil L/R (R being the damping resistor) time constant, and any measurable target response will not show this decay at all because it is much faster than any meaningful targets time constant ... excepot water or salts.

[crane]

... which results in responses far below noise floor at any reasonable sampling point. But in essence and strictly speaking you are right - while there is any significant voltage at Tx coil, the X component follows verbatim. I'm also right because the Tx coil voltage after flyback consists only of exponentially decaying voltage due to the Tx coil L/R (R being the damping resistor) time constant, and any measurable target response will not show this decay at all because it is much faster than any meaningful targets time constant ... excepot water or salts.

I'm not sure how you can say this after seeing the huge ferrite response in the video, which is exaggerated because the coil is under damped. It would though still be present up until the spike has fully decayed if the coil was critically damped, and note that a late sample taken after the spike has settled can't subtract any residual X because it is no longer present.

As to "and any measurable target response will not show this decay at all because it is much faster than any meaningful targets time constant ...
except water or salts". The following picture clearly shows target (and ground) sampling beginning while X is very high, at around 5 usecs.

http://www.geotech1.com/forums/attachment.php?attachmentid=32917&d=1431559666

Contamination of the early signal by the X response is one of the main reasons why we can't sample earlier and the "rejection" in the video isn't complete because we can still hear what most prospectors, who spend most of their time investigating murmurs, would class as a very definite "dig me" signal from the ferrite.
You say X isn't a problem but it is important enough to be the subject of quite a few patents and obviously must be addressed if we wish to sample reasonably early in hot ground.

[Mechanic]

Hmmm, one thing I think that needs to be cleared up with PI terminology is sample delay. I think traditionally the sample delay is measured from the tx off point however the tx is actually finished when the current has reached 0A or a steady state at the end of the flyback pulse. And voltage really has nothing to do with it, it's all about current.
IMHO sample delay should be measured from the point where the current reaches 0A. This is the point where target eddy currents begin to decay, not the end of the tx period.
Now I realise that measuring from the point where the current reaches 0 is difficult as there is no real defined point where this happens, but if you look at it like this:
so we have a tx period with the peak current reaching say 1A and that 1A then takes say 1us to reach 0A
And we have another tx period where the current reaches say 10A and that takes say 10us to decay to 0v
For a small target, if we were to sample at say 5us by traditional terminology, in the first example we are sampling the target 4us into the receive period(assuming receive period is after the current has reached 0A)
In the second example if we were to sample at 5us by traditional terminology we are sampling while the tx turn off ramp is still happening, things are not going to be happy.
Now, if we were to sample at 5us from the point where the current reaches 0A, I would say that we should get the same response from the(small) target provided that the tx off slope is the same gradient.

That's my 2 bobs worth!

Cheers Mick

[Old cart]

Eureka moment?

That is a good point. Current is not too tough to measure. Even without a current probe you could measure the voltage drop across the ground supply wire and just use ohms law. The scope probe ground, unless it is differential, has to be on the detector ground preferably at the main supply cap.
Some thoughts:

1.What would happen if we sensed charge current and then use a circuit to terminate the transmit pulse when the current approached 0?
2. What if we measure the resulting pulse width?
3. What if we combined this with a constant current source?
4. What if we measured pulse width using #1 with only ground present and then subtracted the same with the target added?

Could any knowledge of the target be gained by doing any or all of these things?

[green]

Hmmm, one thing I think that needs to be cleared up with PI terminology is sample delay. I think traditionally the sample delay is measured from the tx off point however the tx is actually finished when the current has reached 0A or a steady state at the end of the flyback pulse. And voltage really has nothing to do with it, it's all about current.
IMHO sample delay should be measured from the point where the current reaches 0A. This is the point where target eddy currents begin to decay, not the end of the tx period.
Now I realise that measuring from the point where the current reaches 0 is difficult as there is no real defined point where this happens, but if you look at it like this:
so we have a tx period with the peak current reaching say 1A and that 1A then takes say 1us to reach 0A
And we have another tx period where the current reaches say 10A and that takes say 10us to decay to 0v
For a small target, if we were to sample at say 5us by traditional terminology, in the first example we are sampling the target 4us into the receive period(assuming receive period is after the current has reached 0A)
In the second example if we were to sample at 5us by traditional terminology we are sampling while the tx turn off ramp is still happening, things are not going to be happy.
Now, if we were to sample at 5us from the point where the current reaches 0A, I would say that we should get the same response from the(small) target provided that the tx off slope is the same gradient.

That's my 2 bobs worth!

Cheers Mick

I see it different. Example, 300uH coil, Rd=1000 ohms, snubber or avalanche volts 450v. Decay rate (I/T=E/L) at 450 volts=450/.0003 equals 1.5 amps/usec. Once current drops to( I=E/R) 450/1000(.45 amps) current decays exponentially with a TC=L/R .0003/1000(.3usec). Add .37usec for 1A peak or 6.4usec for 10A peak to the exponential decay. Takes about 20 TC for the coil current thru the 1000 ohm resistor to reach 1uvolt (6usec). When I try it in LT spice, a critically damped coil exponential decay is faster, maybe 15% less time. My question is how close to zero do we need to be since an exponential decay never reaches zero?

[soverey]

I see the discussion deviated somewhat, although the deviation from the topic is interesting ..., the question regarding the main topic still remains; and that is: is the the promised detector project or kit going to be published or offered for purchase or is it just like I called it a demonstration of technology?

I mean there is on one side Carl, Tec, Orbit and a few more that published either a commercial detector project or their own project ..and on the other side are few more that partially published a project either to please their own ego and show others how good they are or just to advertise it in order to sell it.

Generally speaking: if you going to post an unfinished detector project just make it clear: is it for sale, is it just to show people how good you are as an electronic engineer, or if is work in progress then answer other members questions on the progress of the project.

I see a lot of people retaining from giving too much information on more advanced designs ...for sure there are a lot of persons that could publicly develop/ publish a top notch detector project.. but of course here is where ego and greed comes in.... and that's why we only build 20 years old designs and detectors cost up to 10.000 us dollars.. because of not sharing

[WM6]

... and that's why we only build 20 years old designs and detectors cost up to 10.000 us dollars.. because of not sharing

Don't forget factor of time.
It is not easy for single full time working person to substitute different (Research, develop, test etc.) teams in MD developing and producing companies to be able to prepare and put good ideas on market.
And sometimes alone inventors at some points face with saying of "rooster that crows too early" (not to value this moodz project).

[Qiaozhi]

I mean there is on one side Carl, Tec, Orbit and a few more that published either a commercial detector project or their own project ..and on the other side are few more that partially published a project either to please their own ego and show others how good they are or just to advertise it in order to sell it.

Remember that everyone here is a volunteer, and no-one gets paid for their efforts, so you have to take things as they come. Many members have shared complete projects for free. One of these is (of course) the MiniPulse plus (MPP) project. An original MiniPulse from Eric Foster was back-engineered and then modified to improve its response to smaller targets such as gold nuggets. Other members, such as KRinAZ, have made further modifications to increase the sensitivity to even smaller gold nuggets, with some extensive testing in the field. We're also lucky to have SilverDog, who is able to supply the necessary PCBs and/or kits for a very reasonable price.

Generally speaking: if you going to post an unfinished detector project just make it clear: is it for sale, is it just to show people how good you are as an electronic engineer, or if is work in progress then answer other members questions on the progress of the project.

I do agree though that it can be frustrating when a project starts off open-source, and then the designer gets tempted by the dark side; especially when they've leeched some valuable technical information from other Geotech members before making a sharp dash for the exit. Quite often though, the designer is not really sure what he wants from the project itself, prior to getting the dollar signs flashing in their eyes.

I see a lot of people retaining from giving too much information on more advanced designs ...for sure there are a lot of persons that could publicly develop/ publish a top notch detector project.. but of course here is where ego and greed comes in.... and that's why we only build 20 years old designs and detectors cost up to 10.000 us dollars.. because of not sharing

It's everyone's right not to share if they so wish, but they should make it clear from the start. Over the years we've had many projects proudly displayed on this forum, only for them to disappear from sight ..... and never to be heard of again. Quite often this is the result of an over-zealous imagination that culminated in nothing, and then embarrassment in having to admit they were wrong.