TESORO GOLDEN SABRE

[dbanner]

I think in a concentric, the tx should lag the RX by around 20 (tgsl). The opposite for the DD ( lead by 20).
But I do remember 200 being measured on an original tesoro concentric. There is a thread somewhere.

[dbanner]

Now here is a question.
Start off with two in phase sinus waveforms. Phase shift one of them to the right. Now the one shifted to the right can now be described as leading. Now continue shifting it to the right, at what point can it be said to be lagging. After 180 degrees? How would you know if it were leading or lagging merely by a snapshot of the oscilloscope screen? How would you know where to take the reference. From the two closest adjacent peaks? Two cursors merely give an angle, one relative to another. But leading, lagging?
Someone explain. Could it be for the sake of nomenclature, a waveform should never be described as leading or lagging more than 179.99..... degrees. At 180 degrees it would be ambiguous?
I shall try to find the thread where Qiaozhi measured the phase of a Tesoro concentric. I hope I remembered correctly, if not, I apologise.

[Qiaozhi]

It depends on which side of the null that the loops are balanced.
On one side a non-ferrous target may cause an increase in amplitude at a 20 degree offset, whereas on the other side it may cause a decrease. In the latter case you would have to sample at a point which is effectively 180 + 20 degrees (i.e. 200 degrees) otherwise discrimination would work opposite to what you want.

[dbanner]

It depends on which side of the null that the loops are balanced.
On one side a non-ferrous target may cause an increase in amplitude at a 20 degree offset, whereas on the other side it may cause a decrease. In the latter case you would have to sample at a point which is effectively 180 + 20 degrees (i.e. 200 degrees) otherwise discrimination would work opposite to what you want.

All well and good, however...
Sometime ago, was it not you who measured and described the phase balance of a stock tesoro concentric coil at 200 degrees? I seem to recall a relevant thread on this matter. If not, l apologise.

[dbanner]

You did not quite understand me correctly --- there cannot be a phase difference of 200 degrees, maximum 180. Show me an oscillogram with such a phase)))

SWL, now I am understanding your point on this matter. I think you are right. Practically speaking, phase should be measured from the two closest adjacent peaks of the two sinus waveforms of identical frequency, in which case there would never be more than 180 degrees phase shift. In addition, the phase difference could the be described as minus( lagging) or plus( leading).
So if I say, for example, the TX has a minus 20 degrees phase shift relative to the RX, then it's obvious to everyone that the TX is lagging the RX by 20 degrees.
I should intuitively measure the two closest adjacent peaks, in which case I will always get less than 180. Minus and plus ( lagging or leading) is simply who is relative to who.

[waltr]

In comparing two sinus waves. I measure from the Rising ZERO Crossing of the 'reference' wave forward in time to the rising Zero crossing to the other.
Phase difference in my mind can go from 0 to 360 (which is back to zero) degrees.

[dbanner]

In comparing two sinus waves. I measure from the Rising ZERO Crossing of the 'reference' wave forward in time to the rising Zero crossing to the other.
Phase difference in my mind can go from 0 to 360 (which is back to zero) degrees.

Yes, zero crossing is the right way, but not the only way, especially if the sinusoids aren't perfect.
But for the code guys, it's probably the easiest way to extract phase angle by formulaic means.

Surely a waveform can be shifted by up to 360 degrees along the x Axis in the time Domain, this is for a single waveform being shifted computationally or by means of a circuit, and compared to it's original position, in other words, you assume a zero crossing of T=0. On the scope, you are comparing the same signal but at two different points in the circuit. There is no ambiguity about which one is leading, and it will end up having a negative phase angle.( It's zero crossing will be behind the zero crossing of its phase shifted counterpart, hence leading)
But my assertion is when comparing two separate sinusoids (same frequency) but of different origins on a scope screen, one cannot say for sure who is leading vs who is lagging. It's all relative. T=0 is merely an abstraction. The two closest adjacent zero crossing points will always be less than 180 degrees apart. That's my point. But if you were writing code for DSP, then I imagine you can compute all the way to 360. Or even 720 and beyond. That's software.

[dbanner]

Above, zero crossing at 2π is implied for adjacent waveform comparison.

[dbanner]

It depends on which side of the null that the loops are balanced.
On one side a non-ferrous target may cause an increase in amplitude at a 20 degree offset, whereas on the other side it may cause a decrease. In the latter case you would have to sample at a point which is effectively 180 + 20 degrees (i.e. 200 degrees) otherwise discrimination would work opposite to what you want.

How can a coil like this work? And yet it works perfectly, even though it is balanced where tx is leading by app. 200 degrees. Won't the phase detectors sample at the wrong intervals? Wouldn't they be off by some 180 degrees?

[waltr]

How can a coil like this work? And yet it works perfectly, even though it is balanced where tx is leading by app. 200 degrees. Won't the phase detectors sample at the wrong intervals? Wouldn't they be off by some 180 degrees?

I posted this above:
You need to look at the SAMPLING switch timing also. Then know that the next stages, differential circuits (I said integrators above in error), are AC coupled so only give a response to Changes.

[SWL]

So where is the correct phase to watch?

[SWL]

And one more question. Today I nulling the concentric sensor and noticed the distortion of the lower part of the RX sinusoid on pin 1 LF353. What is the reason for this - a bad power supply minus or a step on the TX? Or poor quality op amp?

[waltr]

So where is the correct phase to watch?

IF Yellow trace is the TX signal, this is the "reference" for measurements.
Then 1.png is the correct measurement since we know that due to the circuit the RX signal will LAG and appear later in time.
Therefore the phase shift is:
Period = 1/11.987kHz = 83.5us
Shift = 45.2/83.5 *360 = 194.8 degrees

[waltr]

And one more question. Today I nulling the concentric sensor and noticed the distortion of the lower part of the RX sinusoid on pin 1 LF353. What is the reason for this - a bad power supply minus or a step on the TX? Or poor quality op amp?

That is odd and I really do not know why it looks like that.
Take a look at the input, on the RX coil, to the preamp and see if it is there.

Also ensure NO metal is near the coil.
Could be a bad opamp, swap with a new one to see. Do ensure the LF353 is NOT a fake.

[SWL]

Exactly! I completely forgot to look at the input signal. I'll take a look in the evening. There was definitely no metal nearby