Field testing modified MPP

[Davor]

This looks slow
Optos are notorious for sluggish response, and discharging IRF740 (or any other power mosfet) through 1k is most certainly slow. A Schmitt of some kind should do much better and cooler.

[CAS]

The 6N136 is a high speed opto with 2mhz bandwidth that has high to low turn on time of .3us. http://www.vishay.com/docs/83604/6n135.pdf

I tested the response times on the cro with the mods vs standard MPP and there is no difference in TX turn on/off times.

[Achillion]

This looks slow
Optos are notorious for sluggish response, and discharging IRF740 (or any other power mosfet) through 1k is most certainly slow. A Schmitt of some kind should do much better and cooler.

I came across the IX3180 which is a dedicated high speed Mosfet driver. Would make a perfect match with the uC.

[CAS]

The IX3180 has a 20ns turn on and turn off time so it's a little slow in this situation.

[Achillion]

The IX3180 has a 20ns turn on and turn off time so it's a little slow in this situation.

Not sure about that, 20ns is 0.02us which is negligible for a PI front end. The advantage of a dedicated driver is that it will pull the gate to either 0V or Vg and therewith force the gate charge. This will improve the response time of the Mosfet. Anyway I will order one of these and give it a try. They're only recently introduced and not easy available. Believe Mouser is one of the suppliers (order a 50 cent component and pay 50 dollar shipping). CAS please continue with providing details. Some scope images would be great. I'm in the same process (not as far progressed as you) and I read this thread with interest.

Cheers mate.

[CAS]

Sorry I had a brain snap. You are right of course. I had seen that IC before in a trade mag but dismissed it as I couldnt find a supplier. Will have to have another look for it but I refuse to pay Mousers crazy shipping charges.

[CAS]

This looks slow
Optos are notorious for sluggish response, and discharging IRF740 (or any other power mosfet) through 1k is most certainly slow. A Schmitt of some kind should do much better and cooler.

One other thing I want to point out is I was not trying to improve the TX turn on/of time of the MPP. The 555 timer has a 200ns typical rise and fall time so the opto was close enough to simulate it. The field test was all about testing timings and to see how the MPP worked in the real world. There are lots of things that could be done to make it faster/better but you may as well design a completely new circuit if you want to do that. The MPP is a great learning platform where basic ideas can be implemented and tested out in the field before going to the expense of having a new design built that may or may not work well.
I now know what GB/EF timings work in mineralised soil. I also now know that the physical construction of my detector was quite well balanced and easy to use. All of these things can not be determined in the workshop.

[mickstv]

HI Cas, the frontend gain might be too high in the MPP for AU soil, could you post up some oscilloscope screen shots, at the input of the first preamp and on the output of the second preamp.


Cheers
Mick

[CAS]

HI Cas, the frontend gain might be too high in the MPP for AU soil, could you post up some oscilloscope screen shots, at the input of the first preamp and on the output of the second preamp.


Cheers
Mick

I have already pulled it apart again to remove my current limit mod and to change the opto connection. Now waiting to try the IC that Achillion mentioned previously. Once I get it back together I will post some screen shots.

I had thought perhaps the gain was too high as it would go from very little signal to full on very quickly.

[6666]

Hi Cas, thanks for the info, well done on the code, been trying to source a uno32 locally here in Melbourne but no go so far, so off to the net
did you compare the 2000 timings to Davor's did they gave simular results ?

[CAS]

Hi Cas, thanks for the info, well done on the code, been trying to source a uno32 locally here in Melbourne but no go so far, so off to the net
did you compare the 2000 timings to Davor's did they gave simular results ?

Element 14 have them in stock according to website.

Davors timings were probably a little easier to use as its all done mathematically where as the SD2000 timings were manually tweaked. Similar results but there were slightly different sounds at MPP speaker. Not having a cro out at diggings meant I couldnt see what was going on. Maybe next time I will have to take portable workshop lol.

[CAS]

I came across the IX3180 which is a dedicated high speed Mosfet driver. Would make a perfect match with the uC.

I just had a quick look and perhaps a FOD3184 may be ok and is readily available.

[6666]

Not having a cro out at diggings meant I couldnt see what was going on.

You DID bring back a bucket full of dirt ?

[CAS]

I certainly did. Lots of ironstone and quartz. Going to pan it first though just in case.

[mickstv]

I have already pulled it apart again to remove my current limit mod and to change the opto connection. Now waiting to try the IC that Achillion mentioned previously. Once I get it back together I will post some screen shots.

I had thought perhaps the gain was too high as it would go from very little signal to full on very quickly.

Thank CAS. Basically the first sample needs to be taken after the flyback reaches flat line. I'll post a pic shortly of what I'm talking about.

I certainly did. Lots of ironstone and quartz. Going to pan it first though just in case.

You should also find plenty of slightly magnetic rocks in your sample.