Auxiliary current limit with coil cell powered device

[nbcater]

Good morning,

My name is Nathan and have been lurking on this forum for a while absorbing lots of information from some very knowledgeable folks but have got to a stage in my requirement pulling process where I need some advice.

To give some background I am working on a very low power BF device which is powered from a 2032 coin cell, it wakes up when an NFC coil gets close to it, gets used for a few minutes and then goes back to sleep. It’s such lower power that there is no charging capability so there is never a connection to mains.

The requirement that is causing me problems is that of auxiliary current. This device is modular, and has spring loaded pins that are occasionally exposed to the user/operator, and one of these pins is connected through a fuse, and parallel current limiting resistors to the positive terminal of the internal battery. In Table 3, and under 8.7.4.8 the limits for auxiliary current are 10uA, and 50uA in normal and SFC respectively. At mains voltages that’s not an insignificant amount of power, but when I only have 3V, that is not enough current for our application.

Link to Table 3:

https://imgur.com/a/eWSffrt

• I was wondering if anyone had come across an application like this before and what their solution was.
• Alternatively, is there a clause somewhere that can be interpreted in a way that makes this test irrelevant for such a low voltage device.

For reasons I have not gone into the solution you maybe thinking at this point of removing, or hiding the pins is not a path we can go down. : (

This post is open for discussion and will be active in replying if members want some more information. This issue was not something I have come across on this forum so my hope is that it will be valuable for others too.

Best Regards,
Nathan

 

[Peter Selvey]

What is the function of the pins?

The low dc limits are there as even small currents over long periods (hours) can cause tissue necrosis. Even so, for this particular limite, the standard has a one size fits all so the limits apply regardless of the actual time. I think because it is generally no problem for designers.

However these limits only apply to the "applied parts" which are parts that need to contact the patient for function. It sounds like the pins are not applied parts but this should be clarified, see the definition 3.8. If they are not applied parts, there is still some risk management for parts that could contact the patient (clause 4.6) but that's more flexible and you could take into account the duration and other factors associated with the contact with the pins in the analysis.

 

[nbcater]

What is the function of the pins?

The low dc limits are there as even small currents over long periods (hours) can cause tissue necrosis. Even so, for this particular limite, the standard has a one size fits all so the limits apply regardless of the actual time. I think because it is generally no problem for designers.

However these limits only apply to the "applied parts" which are parts that need to contact the patient for function. It sounds like the pins are not applied parts but this should be clarified, see the definition 3.8. If they are not applied parts, there is still some risk management for parts that could contact the patient (clause 4.6) but that's more flexible and you could take into account the duration and other factors associated with the contact with the pins in the analysis.

Thanks for the quick response Peter,

Was unaware of necrosis over longer periods so thanks for pointing that out to me : )

You are correct in that these pins do not contact the patient, however there is still a possible connection (SFC) through some conductive material (uninsulated) which is wrapped around parts of the patient. Although this material is coated in a thin layer of silicon as a post process, if there is a break in the coating, the worse case is a connection to these pins through ~10R resistance.

Interesting point about risk mitigation, we could go down that route but ideally we would have several milliamps flowing though this conductive material, and for a maximum of 5 minutes.

Although we could try and justify in our specific application, surely there is a limit to that too? And when we're talking about magnitudes over the original limit how effective are we going to be?

 

[Peter Selvey]

Some factors to consider:

1) the actual risk of auxiliary currents is based on current density rather than current. The limits of 10uA assume a fairly small area in the order of 1mm². Currents spread over large areas are not a major concern.

2) small contact areas usually have high skin resistance >> 1k assumed by the standard. The only way to get decent currents to flow at low voltages like 3Vdc is to have large areas or bypass the skin (e.g. via needles, catheters etc). Combining with (1) might mean small holes are no risk due to high resistance, while large holes are no risk due to the current spreading out.

3) you need two points of contact for current to flow. That might be unlikely even though the coating is not "safety" insulation.

4) Even long term tissue necrosis is not high severity harm, not like stopping the heart

5) Also the time is short, 5min, so the harm is likely to be negligible

So it could be reasonable under risk management. However, it also makes sense to limit the current as much as possible without getting in the way of the function (inherently safe). Also should justify the use of thin coating on clinical needs or practicality.

As always, if you need to work with a test lab, they may be reluctant to accept a risk based rationale, so it's best to bring this up early on.

 

[nbcater]

Some factors to consider:

1) the actual risk of auxiliary currents is based on current density rather than current. The limits of 10uA assume a fairly small area in the order of 1mm². Currents spread over large areas are not a major concern.

2) small contact areas usually have high skin resistance >> 1k assumed by the standard. The only way to get decent currents to flow at low voltages like 3Vdc is to have large areas or bypass the skin (e.g. via needles, catheters etc). Combining with (1) might mean small holes are no risk due to high resistance, while large holes are no risk due to the current spreading out.

3) you need two points of contact for current to flow. That might be unlikely even though the coating is not "safety" insulation.

4) Even long term tissue necrosis is not high severity harm, not like stopping the heart

5) Also the time is short, 5min, so the harm is likely to be negligible

So it could be reasonable under risk management. However, it also makes sense to limit the current as much as possible without getting in the way of the function (inherently safe). Also should justify the use of thin coating on clinical needs or practicality.

As always, if you need to work with a test lab, they may be reluctant to accept a risk based rationale, so it's best to bring this up early on.

1) The minimum distance we are looking at is 4-5cm so looking at much larger areas, so the density might be acceptable.

2) 1K does seem like a very generalised value for skin resistance, are there sources available that offer a more accurate value based on distance, location, contact area etc?

3) Good point there, we might be able to get around it by arguing this.

4) I want to avoid going down the route of arguing this area, especially when increasing the current by >1000x.

5) Same point as above, although could be useful as additional evidence?

It's a problem that is possible to solve with the 10uA limitation, it just requires a change of approach to our design, and likely some more complex circuitry. The higher the current limit we can argue then the easier it is for me to design for, so it might be a balance of some thing discussed here.