Branson 8210 Ultrasonic Cleaner - purpose of diodes

[MLXXXp]

I've been asked to test and, if necessary, repair an old (built in early 1994) Branson Bransonic 8210 MTH Ultrasonic Cleaner.  I was unable to find much information on this unit so I decided to trace the circuit board and create a schematic to help with in-circuit testing of the components. My schematic is attached.

I'm curious about what the purpose of the back-to-back diodes are, in series with the drain of the MOSFETs, as well as the diodes between the drain and source. These are D9, D10, D11, D12, D13, D14.

D9 and D11 are BYV29-400 400V 9A Ultrafast
D10 and D12 are MBR1635 35V 16A Schottky
D13 and D14 are 1N4937 600V 1A Fast recovery

Other similar circuits I've seen don't include these diodes. Does anyone know what their purpose is?

[Paul Price]

Undoubtedly these diodes are used as maelstrom flux modulators.

When the top MOSFET is tuned on, current flows to the Drain though the Schottky diode. When a spike is produced when this MOSFET is turned-off, voltage spike currents will flow mostly through the ultrafast D11 diode and bypass the built-in diode of the MOSFET because this path involves two diode voltage drops due to D14 being in series with MOSFET's internal diode and so these diodes improves turn-on time and reduces MOSFET dissipation. Same game with the MOSFET below.

[MLXXXp]

Paul, thanks for the response. It makes sense.

I had already assumed that D11 was to absorb reverse voltage spikes quicker than the MOSFET's internal diode could, but I wondered why you wouldn't just put it directly across the drain and source, leave out D12 and D14, and let both D11 and the internal diode share the load.

I hadn't considered that any current allowed to flow through the MOSFET's internal diode could affect its turn on time. As far as power dissipation goes, the MOSFETs are on a pretty big heat sink, but the power diodes also each have a heat sink of their own, as can be seen in the included board photo.

[megajocke]

It's rather MOSFET body diode turn off that is the problem which prompts using these diodes. This is a very common feature in applications where the bridge current may be leading the voltage.

Especially for high voltage and/or old MOSFET types, the body diodes have hideous reverse recovery properties. If the body diode of one mosfet is conducting when the opposite side one turns on, the results may range from anything between high losses, and latch-up of parasitic structures followed by shoot-through and instantaneous destruction. Look for "maximum rate of rise of Vds" or some such parameter

Modern low voltage MOSFETs tend to be quite robust against this phenomenon.

[MLXXXp]

It's rather MOSFET body diode turn off that is the problem which prompts using these diodes.

The APT4020AN MOSFET's internal diode has a reverse recovery time of 720ns max.
The BYV29 diode's reverse recovery time is only 60ns max.

So improving recovery time, by minimising any current flow through the MOSFET's internal diode in either direction, is likely the goal.

Thanks for educating me.

[levesqj]

Did you find what was your problem? I have one identical and I have only half power (sonic & Degas) and when I turn on the Heater (Mine as the heater too) it reduce the power to the quatter.

Thanks,
Jonathan Lévesque

[MLXXXp]

Did you find what was your problem?

It turns out there wasn't a problem.

I tested all the components in circuit, as best as I could. Everything seemed OK, so I filled the tank with water and dish soap and tested it using strips of aluminium foil. The foil ended up perforated as expected.

The person who wanted me to test it had actually sold it to someone else who said it didn't work. I guess the person who bought it wasn't using it properly, but then figured it out, because I haven't heard any reports of problems since I returned it.