MPP Ground Balance

[6666]

Is the pulse controller something you can share with us ?

[CAS]

It's just being controlled by a Chipkit Uno. Pretty basic controller but great for experimenting with. Once I am happy with the results I will post the interface circuit and code.

[Davor]

Would there be any advantage in an adjustable GB delay? I was thinking of trying a method of one TX period having x GB delay, then the next TX period having y GB delay, next period back to x delay and so on. Maybe this alternating GB delay will lessen the 'hole' issue?

Yes. If you keep all the duration ratios intact it should work. Additionally you'd need to take care of phase reversal of the response beyond the hole.

The delay tuning is there so that you can fix inconsistencies like EF sample contribution, imperfections in Tx etc, but the ratio should work for a span of main sample durations, which opens the toggle operation.

[6666]

It's just being controlled by a Chipkit Uno. Pretty basic controller but great for experimenting with. Once I am happy with the results I will post the interface circuit and code.

Thats great mate, hope it works for you.

[6666]

Davor with your timing should the main sample width = main sample delay ?

[Davor]

No. Not the way I do it. In an analogue solution I can change initial delay a little in + or - direction and thus achieve GB tuning.
However, since you do it with a micro, you can't employ my method verbatim, so you better keep main sample width = main sample delay. A delay that is more suitable for GB tuning with a micro is the delay between main and GB samples.
Basically both approaches work to the same end, only I found initial delay to be much easier to realise in an analogue solution.
Keeping main sample width = main sample delay, followed with a GB sample in a same time ratio, should provide close match to a GB solution. The remaining effects are a) EF sample, and b) charging time.
a) The EF sample influence on GB can be diminished, but not eliminated, by delaying it further, but that worsens system SNR.
b) All GB solutions relay on the linearity of ground response in log-log scale, however, it is so only up to the time equivalent to the charging period. So if charging period is 100us the ground response will remain linear in log-log scale for 100us after flyback, but it will skew after that.

Both effects, together with probable soil features, will ask for slight deviation from the mathematically straight GB solution, and there you have it. You need some means of tuning.

[6666]

Thanks Davor for the reply, the info is very helpfull.

[CAS]

If you connect a pot or rotary encoder to an analogue input then you can manually tweak any settings you want. Davors system certainly worked for me. All I really needed to do was to refine the electronics side of things. Unfortunately all my testing and experimenting has been put on the backburner while I play with my new detector. Once the novelty wears off I will get back to playing with the circuits.

[6666]

Hi Cas, what is your new detector ?

Davor, I've been running a lot of sims with your spice file with the pulse from the 555 matching the pulse from my MPP,
and using different values of the delay and time base pots, and recording the timing.

I've made a graphic based on your micro description from post #21 , which has a variable delay between the main sample and ground balance sample
while keeping the pulse width ratio's the same as the cd4011, does it look correct ?

[CAS]

What I started doing in the end was taking the EF sample before the TX pulse. That way the Ef Sample is always as far from the end of the TX as possible.

My new detector is a GPZ 7000. Despite all the negative comments from some people, I have to say I love it. Only thing I can fault with it so far is its a little heavy.

[6666]

WOW a 7000 you are the only person I know that has one, I really hope that you find something good.

[CAS]

Yeah so do I. I have found plenty of tiny bits of lead down about 60cm - a bit frustrating doing all that digging but at least I know if I walk over gold, I will get it lol.

[mickstv]

Yeah so do I. I have found plenty of tiny bits of lead down about 60cm - a bit frustrating doing all that digging but at least I know if I walk over gold, I will get it lol.

Hi CAS, so what sort of size were these tiny bits of lead that you found down at 60cm (23 inches) ?

[CAS]

They were between about 4 - 8mm across and varied. Some were small round lead shotgun pellets, others were just fragments of lead. Bit of a pain as I was getting pretty excited on a few as the soil looked undisturbed. Also I only have a short handled pick and it took me ages to dig some of them.

Did find a cool boot nail down about 30cm that looks pretty old.

[Davor]

Davor, I've been running a lot of sims with your spice file with the pulse from the 555 matching the pulse from my MPP,
and using different values of the delay and time base pots, and recording the timing.

I've made a graphic based on your micro description from post #21 , which has a variable delay between the main sample and ground balance sample
while keeping the pulse width ratio's the same as the cd4011, does it look correct ?

The timing is not correct.
If you use a delay between main sample and GB sample, the duration of GB sample in the example picture should be 3x main sample. If main sample to main delay ratio is 1:1, the GB sample to (main delay+main sample+insert delay) should also be 1:1. Or better say 3:3, but in your picture it is 3:2. To make it correct, you need GB=60us, and EF=40us.

My analogue solution does not have any delay between main and GB samples, and hence GB duration is 2x main duration. This scheme is the most compact of all, but apparently the target attenuation achieved this way is the biggest. So pushing GB sample further in time may be beneficial, of course with keeping the duration ratios.