avalanching mosfets

[c_batchelar]

hi all
so im woundering about avalanching mosfets and what actualy happen when they avalanch.
now ive tried doctor google a few time and i get either a break down of the quantim physics behind it, how to correctly read the mind feild of miss leading info that are in spec sheets or a beat around the bush articals that basicly can be summarised as avalanch = bad. so can any one explane it in plan english maybe with a oscilloscope reading for the people that just dont quite get it

[Qiaozhi]

hi all
so im woundering about avalanching mosfets and what actualy happen when they avalanch.
now ive tried doctor google a few time and i get either a break down of the quantim physics behind it, how to correctly read the mind feild of miss leading info that are in spec sheets or a beat around the bush articals that basicly can be summarised as avalanch = bad. so can any one explane it in plan english maybe with a oscilloscope reading for the people that just dont quite get it

When the voltage across the MOSFET exceeds the stated avalanche voltage, it simply breaks down and allows current to flow between its drain and source. The current continues to flow until the applied voltage drops below the avalanche level. It's more or less the same thing that happens with a spark gap. Basically, the MOSFET can only withstand a certain level of reverse voltage across its drain and source. Once you go above that, the junction breaks down. In some MOSFETs, that are not avalanche rated, this can destroy the device.

[homefire]

Some what like a diode reaching the Braking point in Reverse Biased. Bad Ju Ju.

[bbsailor]

hi all
so im woundering about avalanching mosfets and what actualy happen when they avalanch.
now ive tried doctor google a few time and i get either a break down of the quantim physics behind it, how to correctly read the mind feild of miss leading info that are in spec sheets or a beat around the bush articals that basicly can be summarised as avalanch = bad. so can any one explane it in plan english maybe with a oscilloscope reading for the people that just dont quite get it

When the Transmit (TX) pulse drives a MOSFET that is connected to the coil the charging current raises gradually by the time constant of the coil and TX circuit. This includes the DCR of the coil, the MOSFET on-resistance and any series resistance/resistor in the coil circuit. Lets assume that the total resistance is 6 ohms and the coil inductance is 300 micro Henries (uH), then the charging coil time constant is 50 micro-seconds (uS) (300 divided by 6). The TX pulse will need to be 50 uS long to raise the current to 63 percent of its maximum current governed by the supply voltage divided by the coil circuit resistance. In this case lets assume a 12 V supply then the maximum current will be 12 divided by 6 or 2 amps. The current will raise to 63 percent of 2 amps or 1.26 amps. If we let the TX pulse go another 50 uS, for 100 uS total, then the current will raise to about 85 percent of maximum. Add another 50 uS to the pulse, for a total of 150 uS, and the current now raises to about 92 percent of maximum.

When the TX pulse turns off, a flyback pulse occurs where the voltage will spike and looks like a knife with a sharp point only as long as the peak of the spike is below the MOSFET avalanche voltage. If however, the peak voltage spike exceeds the MOSFET avalanche voltage the sharp pointed knife now looks broken off with a flat top rather than the pointed top. As we increase the TX pulse width we put more energy in the coil and as a consequence we generate more energy in the flyback spike and its voltage gets higher. Always choose a MOSFET avalanche voltage that is higher than the maximum flyback spike at the widest pulse width your PI circuit produces.

We use a damping resistor across the coil to quickly quench the energy in this flyback spike so we can quickly change the coil from being a TX coil to now becoming the receive (RX) coil. The time between this transition from TX to RX is called delay. During this transition/delay time the energy induced into the target is starting to decay. The shorter this delay time the more potential target energy there is to detect. Smaller targets need shorter delays to find them. This is why PI machines have some adjustable controls to optimize for various targets such as coins, gold rings and small gold nuggets.

Many techniques are used to reduce the delay time. One key area is to reduce coil and TX circuit capacitance so that there is less oscillating flyback energy that needs to be damped. Other techniques use faster pulse rates at lower energy (shorter TX pulse widths) and integrate many RX samples to boost the receive circuit by reducing noise pickup.

If the MOSFET is in the avalanche mode, the excess voltage energy is being absorbed by the MOSFET and it is getting hotter. Also, the length of the MOSFET avalanche flat top, delays the time slightly when the damping resistor can get the flyback voltage down to zero when the RX circuit can begin detecting potential targets.

As you can see, there are factors that are interrelated.

Joseph J. Rogowski

[c_batchelar]

thanks all, that diffinitly puts that question to bed for me, tho i would have problems expaling it in my own words if asked to....
and i can see the bad ju ju being like 100m dynamic rope and 105m cliff, yeah you can get it to strech if you time a running bounce right but it allways going to hurt...
Mr Rogowski special thanks for the subject specific answer

[Sean_Goddard]

You ALWAYS get a good answer from BBSailor