MYPI direct sampling front end.

[moodz]

Design goals.

0. No pulse amplifiers .. I hate em.
1. No balancing pots .. autobalance
2. No exotic amplifiers.
3. Hum / noise resistant

The tx part is straightforward.

Typical 100 uS pulse.

The sample is the key .... the sample switch mosfet is turned ON at all times except during the sample period when it is OFF. ie the RX is muted except during the sample period.

The sample event should occur 10 or so uS after the end of the TX pulse or the front end peak detector will cop the flyback voltage.

Sample time 50 uS or whatever ....

Instead of a S&H the first stage is a peak detector ( there is a difference )

The last stage autobalances the output to 0 volts.

Output exhibits less than 10mv of noise.

My prototype as above detects targets over a meter. ( probably could do with sensitivity control )

Because o/p is autobalanced to 0 volts it is well suited to input to ADC or meter, VCO sound gen etc etc

[moodz]

doh ... Q1 the IRFP460 gate drive should read TX_PULSE not RX_PULSE ... change in next schematic.

moodz

[okantex]

Hi Moodz,

what is ırf 1405 and rx pulse for ?
why do you need to drive such unit there,can you explain a bit.
regards

[moodz]

hi okantex ... the 1405 basically replaces the clamping diodes in a 'classic' PI design. The difference is that the RX signal chooses when to unclamp the signal whereas in the traditional design the diodes will clamp when the voltage they see exceeds 600mv or whatever the particular diodes forward conduction voltage is.

So in normal operation the 1405 is turned ON ( ie clamping ) but about 5 uS to 10 uS or whatever you would like AFTER the Tx pulsed has ended and the flyback has mostly decayed the RX PULSE turns the 1405 OFF thus going to the unclamped state and allowing the frontend to see the unclamped signal.

I guess you could call it active clamping.

In summary the 1405 ( Q2 ) does three things.

1. Provides active clamping to protect the frontend from flyback voltages.
2. Provides the sample switching window.
3. Prevents HF components in sample by limiting rise time.

cheerz ... moodz

[okantex]

Hi Moodzi

I could not understand this property "3. Prevents HF components in sample by limiting rise time."

also ,I could not understand that both mosfets loooks similar at work ( both are N channel),

why do not you use multiplexer to switch signal,
and both diods are in same manner placed that ,I see that 1405 is also driver like 460.

regards
okantex

[moodz]

hi ok ... you see how the curve of the decaying spike is nice like a ski slope ... ie no bumps .... if you take a fast rising square edge sample pulse you introduce high frequency components .. as I am only interested in the peak value not the width of the pulse I dont need sharp edges which by their nature introduce higher frequency components.

The first transistor switches ON at Tx pulse to conduct the coil current it must be able to handle the current and flyback voltage. That is why it also is a fairly robust device.

The second transistor Switches OFF at RX time ... protects the front end reciever by grounding the flyback voltage via the 1K resistor by remaining switched ON till RX pulse. A conventional mux chip would not be able to handle the flyback voltage. ( that is why clamping diodes are used )

hope this explains ... moodz.

[moodz]

sampler results at output of sampling switch.

NO TARGET

Copper target at 20 cm ( peak drops )

Steel Target at 20 cm ( peak rises )


moodz

[moodz]

updated schematic