Battery powered device - electrical protection requirement

[sreibs]

Hello,

It is my first post on this forum and first of all, thank you for these very precious information !

I tried to read as many threads as I can on this topic, but I am a little bit confused. We are designing a battery powered medical device. While in use by the patient, it is totally wireless, and worn on the patient or hand-handled. The device can be charged via USB, but it is quite impossible that the patient use it at the same time: the device is automatically off while charging, and the usb connector is located in a way that if it is plugged, the patient can not wear the device correctly. Plus, we provide a charger with 2MOPP within (if it can help).

The battery is a LiPo (which come certified with the correct IEC...) and the maximum voltage within the device is 5V. I have an SPO2 sensor and mini-camera (for otoscope purpose). The SPO2 is integrated in the main enclosure of the device, and the camera is attached with a 10cm cable (which cannot be unconnected). Again, while used by the patient, the device is fully wireless.

The labs said that I need to provide 2 MOPP for SPO2 and camera. Which involve to have a 1000Vrms insulation, clearance/creepage 3.4m/1.5mm. Which is huge especially on the camera side. The point, actually, is that I don't understand why I must do this, as I don't understand where the current can flow if the device is battery powered. Moreover, there is no intentional electrical contact (whereas ECG probe for example).

Can you help me clarify this please ?

 

[Benjamin Weber]

Why do you need 2 MOPP against the 5V? - Of yourse a current can flow in a battery ooperated device, that's actually what the battery is there for ;-) You have to make sure, that the patient cannnot connect the (-) and (+) of the battery at the same time, eg. with two fingers. In this case a patient leakage current of 5mA would flow which exceeds tha maximum allowed values. Sometimes there are mounting screws or other conductive parts which are accessible to the patient and connected to battery (-) at the same time, e.g. via the mounting hole on the PCB and a common ground. In this case you have to assure, that the patient is separated from the battery (+) with 2 MOPP.

For a working voltage of 5Vdc the corresponding dielectric strength test voltage is 1000V for 2 MOPP, that is correct. But maybe it is possible to apply two independent 1 MOPP with 500V each only? E.g. if you have areas, where you need to protect from the battery (+) or (-) only?

Regarding the creepage and clearance for 2 MOPP. I think the required creepage of 3,4mm is to large. Creepages can be interpolated. In my oppinion it is possibel to interpolate between 0mm for 0V (which is not in table 12) and 3,4mm for 17V (first line of table 12. For 5V you would end up with 1,0mm. BUT: The creepage cannot be less than the clearance, which is 1,6mm (no interpolation possible).

Again it might be possible to seperate the required creepage/clearance into two single 1 MOPP with 0,8mm each?

 

[sreibs]

Hi Benjamin. Thank you for this clear and quick answer.

Of course the current can flow between + and -. The lab tells me that their concern is about current that may flow from battery (+) and earth. Which I don't understand how it can be possible...

If I understand correctly I should make sure that there is no way that the patient can touch (+) and (-) at the same time. The design of the product makes (-) not accessible anywhere (no screw or external conductive part connected to (-). Well, on the SPO2 sensor for example, the (-) and the supplied voltage (which is 3.3V on this part of the circuit), are very close, so we can say that the patient may touch both (within a 2mm² surface though). In this case, should I separate the (-) and (+) from the touch part with a distance of 0.8mm to implement the 2MOPP ?

 

[Benjamin Weber]

In battery only powered device without any other connections and no conductive accessible parts the chance of currents flowing from (+) to earth is actually negligible. The only possibilities I see here are capacitive coupling to earth and then through the earthed operator/patient (for AC with higher frequencies nony, not for DC). The other way would be along creepage distances.

Regarding the SpO2 sensor: You don't need to separate (+) and (-) here in general. You have to make sure, that the patient cannot come into contact with either both (1 MOPP for each pole) or only one pole (2 MOPP). This is done by the required minimum creepages, clearances or solid insulation. If the sensor is actually fully enclosed (as far as I know SpO2 sensors, this is usually the case), maybe there is only solid insulation and no creepages/clearances.

 

[sreibs]

Thank you Benjamin. I am still confused on the reason to have 2MOPP on the SPO2 sensor. Here is a picture of what I mean:

Obviously, the SPO2 sensor component is powered by a positive volage (3v3) and a ground (- of the battery). The component has a plastic body and a cover glass on top of it. In this case, even if the finger reach the PCB (worst case ever which is actually impossible), I don't understand where the current can flow (through the patient body I mean). So I am not sure to understand why I need to have 2MOPP here.

 

[sreibs]

Hi there. Any idea ?

 

[SebastienBoisvert]

I'm curious if you were able to solve this problem. I plan to do something like you, and I thought that the isolation barrier made by the SpO2 enclosure was sufficient for that.

For the USB input, I plan to use a digital isolator since I have an electrode on the device, and it could be touched. Both on your device, with the enclosure, I think you are ok without it.

 

[jdoran]

My Twopence worth...
Given the geometry you have outlined, and the test house comments, I believe they are looking for 2 MOPP because they consider the camera and SPO2 to be separate (type BF) applied parts. This is to limit the potential for auxiliary currents between each of the applied parts (leakage currents are not possible because the charger cannot be connected while the device is in use).

Given the small size of the device, and only 10cm length on the camera, and that the camera is not detachable, I would consider the SPO2 and camera as the same TYPE BF APPLIED PART, (i.e. the whole device, excluding the charger, is a single type BF applied part) and this will greatly reduce the testing burden.

 

[sreibs]

Thank you jdoran for your reply. I keep you post

 

[sreibs]

@jordan: Actually camera cable length is more 30cm than 10cm. However it is not detachable.

If I understand correctly the principle. They are looking for 2MOPP between the applied part in case a dangerous voltage comes in contact with one part while the patient is in contact with the other part ? In this case, I think we can reasonably consider that if there is a dangerous voltage within 30cm radius of the patient, the medical device is not the source of danger and cannot do anything about it ? Am I correct ?

Thank you very much