Signal Decay of Small Nuggets.

[Anonymous]

I recently did some tests on a couple of small nuggets, that may be of interest. In fact it was once one nugget, before one end broke off, so the composition should be the same. As far as I remember, the nugget originated in W. Australia.
The larger piece weighs 0.9gm, and is 10 x 6 x 1.8mm. The other is 0.4gm and 5 x 5 x 1.8mm. On my very high frequency PI, which can sample at a true 1uS after TX switch off, the total decay time of the larger piece is 35uS, and the smaller is 20uS. These give time constants of approx 7uS and 4uS respectively. If sampling took place at the same time as the TC, the signal is already only 37% of its maximum value. To improve the signal to 60% of its maximum value, you would need to sample at 3.5uS and 2uS respectively. This would be for an instantaneous sample, rather than a sample window as is normally the case.
This illustrates that there is still a way to go, before PI's can achieve the high frequency IB's sensitivity to tiny targets.
Eric.

 

[Anonymous]

Thanks Eric for posting this very useful information. This is just the sort of info I am looking for at the moment.
A friend and I are currently designing/building a PI detector. We are using an Atmel ATMega32 MPU for timing and an RX frontend fairly close to the one used in the Hammerhead design (NE5534 set to 50x gain).
At present, the signal from the coil decays to 0v in around 8 or 10uS depending on the coil we use (Minelab mono and Coiltek DD respectively). Do you have any tips for reducing this?
We are currently using an IRF840 for TX with a pulse length of 80uS. Would replacing this with a device with lower output capacitance help significantly?
Anyway, thanks in advance for any information.
_Raz_