ESD protection of ISO devices

Hi Bernard, isolation device don't solve ESD issue and ESD still is in place and can be worstened by isolated section too.
You need place three level ESD, on input of both the isolation barrier to prevent some discharge on that part broke sensitive part due to capacitive coupling.
Between isolated part you also must prevent barrier voltage exceed maximum admitted to preserve SiO2 isolation barrier, if voltage exceed that I also can fear isolation too get broken.
Even if instrument is fully isolated and battery operated ESD must be in place.

Bernhard,

Can you give a little more detail on the exact ESD test (levels and connections) that was conducted. It will help in understanding the issue better. In the past, we have seen that customers  do violate the device capabilities sometimes in which case we suggest an alternative device.

Regards,

Sarangan Valavan

Hellooo--

thank you both for replying so quickly :)

The ESD tests were performed according to EN 61000-4-2 on 60601-1-2 (6kV/8kV). Everything was perfect, of course.

Here a short description of the device: Its an EEG device, i.e. the purpose is to amplify, digitize and frame EEG signals of a patient and transmit it as digital data stream via USB to a computer. The device of course has ESD protection on both sides, i.e. at the USB interface and the patient interface, i.e. the EEG input jacks. The problem does not appear there. What dies is the ISO7641, which is used for the galvanic isolation along with a DCDC converter using a transformer, which is also compliant with the medical devices norms.

Now, here is what I think what happens: If there is nothing which prevents the patient or clinician to be charged up to, say, 50'000V, then this voltage will appear at the ISO7641 and destroy it. Also the insulation of the transformer could be damaged. We do require ESD measures and best practices stated in the manual, but as we see, in some cases these appear not to be in place.

So, I wonder what else could be done. This problem must apply for all USB operated medical devices, so either all of them die under the same conditions, or there is an elegant solution. Do you know of any?

Thank you,

Bernhard

Hi Bernard, Medical device need have 8KV isolation test, so 76 series isolator are not compliant to this, see series 78xx reported as medical grade application not recommended for part you used instead.
http://www.ti.com/lsds/ti/isolation/digital-isolators-products.page#p1950=8000;8000
If patient and medical staff don't respect antistatic prevention wearing as from manual on clinical surge this can be a trouble to life security too, so check if a gas limiter can apply as extreme prevention to save isolation barrier.

Hello Roberto,
thank you very much! The 7641 was the best I could get back then, and with the kind of testing (contact / non-contact, etc), it passed the compliance checks and tests.
I am of course very happy that a "stronger" device is about to come out. The 7841 seems to be a pin-by-pin replacement (the datasheet does not load here- if available, could you eMail it to me?). When is this product expected to be active? We could use 600 for the lot that we are about to start..
Bernhard

Hi Bernard, sorry I cannot help you about availability, I am just a project free lance, please contact your local FAE or just wait someone from TI answer to you.
I just imposed medical application normative parameter to search and these where the only belonging to. Again just more isolation NEVER can protect from ESD between both sides of interface and <ISO> device isolation barrier forever die when finger model gun is applied between the two isolated part of device.

Hi Bernard,

ISO7841 is expected to release end of this month - Jan 2015. You can contact Kim Devlin Allen (kda@ti.com) if you need samples earlier. ISO7842, which is part of the same family, is already released and available. ISO78xx devices are expected to perform better than ISO7641, because of a stronger isolation barrier.

One way to prevent stress on the isolation barrier for ESD events is to inlcude a high voltage capacitor (5pF to 10pF) between the two grounds of the isolator. (AC leakage considerations may prevent using a bigger valued capacitor).

Rgds,

Anant

Anant Kamath said:

One way to prevent stress on the isolation barrier for ESD events is to inlcude a high voltage capacitor (5pF to 10pF) between the two grounds of the isolator. (AC leakage considerations may prevent using a bigger valued capacitor).

 Hi Anant, this is not enough IMHO to prevent isolation barrier break, I proposed gas discharger as extreme prevention to limit voltage of, adding a 5pF capacitor also tend accumulate a lot of charges help destroy barrier, an RC and discharger ESD network has to be provided between the two grounds.

helloo--

thank you so much! I really appreciate your help! I wonder how other medical equipment manufacturers solve that problem- it applies to all equipment with USB connection.. a gas discharger would zap the patient, and so would probably a capacitor. But maybe that is allowed - we are re-checking the ISO 60601-1-xx... 

We are also checking if we would have to re-do all the ESD and EMC tests if we switched to the new 7841. Are there any certificates available already?

Edit / Update: we are checking options how to reduce the stress on the isolation barrier. Will keep this updated. So, first finding. gas discharger is not allowed :( Maybe a combination of resistors and a Y1 capacitor would do.

Any input how other medical devices manufacturers solve this problem, if at all, are highly appreciated.

Best,

Bernhard