MPP beach rev-e problem

[Qiaozhi]

The purpose of a spider coil is to reduce the interwinding capacitance, which is important if you're trying to get the main sample delay down to a minimum.
This is not important in the MPP because it lacks a ground balance function. If you run the MPP with aggressively low sample delays in the surf you will get falsing.
You probably won't be able to run the MPP below about 10us without falsing, and you can easily achieve this with a scamble-wound coil using teflon wire.

I would suggest starting with a TX pulse width of 100us and a repetition rate of 1000pps. Then experiment with different settings if you want to achieve better results.
The final values will depend on your particular coil, the minimum sample delay, and the battery consumption. Also remember to keep an eye on the operating temperature of the MOSFET.

[cristiano.sar]

The build document says to set the TX pulse width to 400us. This is OK for initial testing because the pulse repetition rate is only 125pps.
Remember that the original MiniPulse repetion rate was much lower at 80pps.

However, there is nothing to stop you adjusting the settings to suit your particular coil. Although this project is called MPP Beach it could easily be adjusted for gold nugget detection.

In the Technical description it says:
Tests have been made with both mono and DD coils:
Initially a simple 400uH 9” diameter mono coil, wound with 0.56mm enamelled wire, was
connected. The TX pulse width was set to 100us, with a pulse period of 1ms (1000pps). Due to this
being a “slow” coil, the earliest sample delay possible was 30us. The sample pulse width was set to
60us. A Victoria penny was detectable at 12” in an electrically quiet environment.
The DD coil was a Garrett Infinium LS, which an elliptical coil of 14” x 10”. The initial (more
aggressive) settings defined in step 3 were used. i.e. a pulse width of 400us, with a pulse period of
8ms (125pps). Since the Garrett coil is a much “faster” coil, the preamp was coming out of
saturation at 8.4us. In this case the minimum sample delay was set to 10us, with a sample pulse
width of 65us. A Victorian penny was then detectable at 13 to 14”.

Thanks for the clarification.
The coil I'm using is a spider with 409uH.
What settings you recommend-me for use in the beach, searching for gold rings?

Thanks.

[Qiaozhi]

The build document says to set the TX pulse width to 400us. This is OK for initial testing because the pulse repetition rate is only 125pps.
Remember that the original MiniPulse repetion rate was much lower at 80pps.

However, there is nothing to stop you adjusting the settings to suit your particular coil. Although this project is called MPP Beach it could easily be adjusted for gold nugget detection.

In the Technical description it says:
Tests have been made with both mono and DD coils:
Initially a simple 400uH 9” diameter mono coil, wound with 0.56mm enamelled wire, was
connected. The TX pulse width was set to 100us, with a pulse period of 1ms (1000pps). Due to this
being a “slow” coil, the earliest sample delay possible was 30us. The sample pulse width was set to
60us. A Victoria penny was detectable at 12” in an electrically quiet environment.
The DD coil was a Garrett Infinium LS, which an elliptical coil of 14” x 10”. The initial (more
aggressive) settings defined in step 3 were used. i.e. a pulse width of 400us, with a pulse period of
8ms (125pps). Since the Garrett coil is a much “faster” coil, the preamp was coming out of
saturation at 8.4us. In this case the minimum sample delay was set to 10us, with a sample pulse
width of 65us. A Victorian penny was then detectable at 13 to 14”.

[cristiano.sar]

I am not familiar with this project. But your tx pulse looks extremely wide at 50us/div. If so this would cause heating and other problems.

The Step 3 of the project manual says "Monitor the TX oscillator output at TP2 with an oscilloscope, and adjust R5 to give a TX pulse
width of 400us." so I think it is correct.

Thanks.

[Altra]

I am not familiar with this project. But your tx pulse looks extremely wide at 50us/div. If so this would cause heating and other problems.

[cristiano.sar]

I think it's all correct. Anyway I replaced U7 (the pulse has a better appearance now). But in previous question I made a confusion about the delay (I thought this was the space between TP2 and TP6, but it really is a pulse)
Here are the pictures of TP6, TP8 and TP9:



Thanks.

[Qiaozhi]

thanks, I changed the capacitor, for another with a little more precision but the result is the same. Adjusting R37 works (affects the delay pulse.)

With C13 as 220pF and R37 (100k) set to maximum resistance, the delay should be 220E-12 * 100E3 = 22us.
Perhaps U7 is faulty.

[cristiano.sar]

Check the value of C13. It should be 220pF.
Also, check whether adjusting R37 has any effect on the delay pulse at TP6.

thanks, I changed the capacitor, for another with a little more precision but the result is the same. Adjusting R37 works (affects the delay pulse.)

[Qiaozhi]

Check the value of C13. It should be 220pF.
Also, check whether adjusting R37 has any effect on the delay pulse at TP6.

[cristiano.sar]

I think you may be confused about the sample timing.
TP6 is the sample delay, and TP9 is the main sample.

It seems my TP6 is wrong as apparently there is no delay (the sample pulse is starting immediately after the TX pulse collapses without delay):
Scope settings are:
Time/Div: 50uS
V/Div: 5V
Trigger: Channel 1

What could be wrong? Thanks

[cristiano.sar]

Here is a small video testing my Minipulse: https://www.youtube.com/shorts/eY49lD1dBcw
I think I still need some adjustments but for me the performance is good with a 20cm coil I also use in the Felezjoo.
Do this kind of spider coils need shielding? Thanks!

[Qiaozhi]

I just finished Minipulse but I have not followed the step by step during the assembly. It works, but I'm lacking some distance/depth. Can I follow the testing/tunning steps even after all the components in place? Thanks.

Yes.

[cristiano.sar]

I just finished Minipulse but I have not followed the step by step during the assembly. It works, but I'm lacking some distance/depth. Can I follow the testing/tunning steps even after all the components in place? Thanks.

[Qiaozhi]

The problem was solved and it works very well now. Also I used 4066 instead of j113. thank you very much

That's good.

[convola]

The problem was solved and it works very well now. Also I used 4066 instead of j113. thank you very much