SIP/SOP requirements for USB port used for charging

EMengineer

Involved In Discussions
Hi everyone,

I have read many useful topics related to this question, but because of the large amount of information and opinions I got confused.
We have internally powered (3.7 VDC) body worn device with CF and BF applied parts and which is charged via USB. Data is send through BT thus the data pins are not used, thus when used on patient there is no connection to other devices.
The device is programmed that it could not be used while charging, thus the risk of mains on SIP/SOP is reduced. Is it possible to cover this with risk by not allowing charging while in use + warning + instructions (operator would be medical personel)?
Although there could be misuse - charging the device on the patient even if the device is not functioning. For example, the operator would charge the device enough to allow one measurement, thus it is not removed from the patient.
My question is about insulation between USB and applied part in such misuse case. Do I understood correctly that USB port shall galvanically be isolated from AP by 2 MOPP's for working voltage (5 VDC) and 1 MOPP for maximum mains voltage (according Fig J.4). In this case the largest a.c and c.d values 1 MOPP for 250 Vrms are - 4 mm c.d and 2,5 mm a.c (when 2 MOPP for 5 VDC are 3,4 mm c.d and 1,6mm a.c). Also dielectric strength test of 1500V should be applied. I hope I did get this part right.
But is it possible to use earlier mentioned risk mitigation measures reduce the potential for such misuse to an acceptable level? Or in any case the galvanic isolation should be used?

j4.png
 

Peter Selvey

Leader
Super Moderator
It's not required in the standard if the intended use, instructions and design are such that the device should be removed from the patient during charging. But it could be considered under general risk management.

It depends on the type of contact with the patient. Key point of concern would be electrodes or other low impedance patient contact, especially long duration, unconscious. In these cases, it can makes a lot of sense to have isolation between those parts and the USB supply as a risk control measure.

Since it is a risk control measure outside of the standard, it could be a little relaxed as well for the criteria (test voltage, creepage etc.). The 1 MOPP requirements in the standard are based on the insulation being continuously stressed with mains voltage for ~10 years, so 1.5kV/4mm is a huge overkill for this kind of situation. You might find an off the shelf USB isolator that has for example 1kV isolation, which is fine in my opinion.
 

EMengineer

Involved In Discussions
Dear Peter, thank you for your answer ir really became clearer. There is one more question about separation of AP. Do I understood right that the only solid insulation is 1MOPP between patient circuit and Enclosure based on mains voltage and 2MOPP based on working voltage between internal circuit and enclosure. There are only the limiting resistors and buffers between AP and internal circuit. In this case the single fault condition during leakage current measurement would be short circuit of limiting resistor? Do you need additional ac/cd from battery operated internal circuit ant AP?
 

Peter Selvey

Leader
Super Moderator
That sounds like something we were discussing in another thread. The standard is written around insulation between circuits but does not fit for parts that are intended to limit currents arising from within a circuit.

In my opinion, it's OK to ignore the standard for ac/cd, dielectric strength within circuits as long as the voltages and currents are inherently low, which is usually the case for circuits that are used for monitoring and measurement such as ECG, EEG. The potential harm from these currents is is relatively low, compared to say contact with 230V mains.

However, if the circuit is high powered (higher voltage, higher currents) then use of properly approved safety components (capacitors, resistors) along with ac/cd is reasonable. Examples are high voltage simulators, electrosurgery. The limits in 601-1 might still be overkill, but some limits should still be applied, maybe based on IEC 60664-1 or some particulars which usually have more reasonable limits.
 
Top Bottom