High Voltage inside Ultrasonic Probes - Do SIP/SOP ports require isolation?

[sankui]

Dear all,

Regarding the ultrasonic equipment, as you may know that the ultrasonic probe driving control circuit will generate high voltage pulse (about 100V) to excite the piezoelectric crystal inside the probe. Because there is no separation between the excitation pulse voltage (100V) and other SELV circuit (e.g. 24V), 100V voltage may appear on the SIP/SOP ports in some situation (e.g. non-safety relevant components failed).

I'm worried about is whether the above mention allowed? Whether SIP/SOP ports need for isolation?

[Peter Selvey]

Hi Sankui,

You have many interesting questions - is your company designing many different types of medical devices?

For ultrasound HV supplies:

Typically there are three aspects which make these safe:

First, in "normal condition" although the peak voltage may be high the rms voltage is usually around 5V, and the frequencies in the MHz which are harmless. You can touch this voltage without any fear.

Second, most ultrasounds have a sensor which detects if the probe is connected. If the probe is not connected, the +HV and -HV supplies (dc/dc converters) are turned off, or disconnected by a relay.

Third, the HV supply should be current limited to the minimum necessary for the function. Because both of the above methods of protection are electronic based, the reliability is limited. In practice it should be OK, but to be on the safe side it makes sense to limit the current as far as possible. Literature indicates for dc currents, you need around 150mA for ~0.001 probability of fibrillation. Most ultrasounds work with around 20mA, at which the worst expected effect is a bit of pain.

Hope this helps.

Additional note:
The above is intended for the connector for the ultrasound probe itself. However, your question may be the SIP/SOPs at the rear of the device. This is almost for sure safe but the reason is complicated. Again it relies on a few factors:

1) The HV supply being physically well removed from any components, cabling that directly connects to a SIP/SOP
2) The HV supply having a current limit that is below the sink current of the lower voltage power supply rails (+5V, +/-12V etc)
3) The HV supply having a common ground (0V) to the SIP/SOP circuits

The theory is that if the HV supply were to short to SIP/SOP circuits, in most cases the current limit would operate and pull the voltage down to less than the lower voltage power supply rails. This can be verified by actual tests shorting the HV to the nearby power supply rails.

There are rare cases where the HV shorts to a nearby point that has high impedance or to a lower power supply rail with insufficient sink capacity, for example:

+100V 20mA supply shorting to a point with 4kohm to 0V (results in 80V)
+100V 200mA supply shorting to a -5V rail with only 50mA sink capacity (-5V regulator blows and >60V appears)

These have cause problems in designs where the HV circuit is situated physically close to the SIP/SOP circuits (e.g. same PCB, adjacent circuits, even same connector). But, in most ultrasound designs the HV parts are well removed (e.g. different PCBs). There is no functional reason to have HV circuits adjacent to SIP/SOP circuits except for the ultrasound probe connector which is covered above.

[sankui]

Hi peter,

Yes, my company manufactures several types of medical equipment, including ventilator, humidifier, patient monitor, ultrasonic equipment (developing).

Your input is very detailed, you must be a medical expert. Do you work for test agency and/or IEC working group?

Regarding the sink current, as far as I know, many low voltage regulators have small sink capability (normally less than 20mA). This can also know from the IC datasheet. If so, the situation will become worse.