TI E2E Community
Precision Data Converters
Precision Data Converters Forum
defibrillator protection circuit
I am designing defibrillator protection circuit for my ADS1298 based ECG logger design.
I have seen the references for this protection circuit but not sure which one to use.
At one place, it is recomonded to use series resistance,Neon lamp and TVS for defib protection
and at other places only series resistor and TVS is used for defib protection.
In my case, becasue of space constraint , Neon lamp can not be accomodated.
Can I use only series resistor and TVS for defib protection?
Do you see any problem here.
Hello Sachin,If a defibrillation voltage (which can be thousands of volts) may be applied while the electrodes are affixed to the subject, I recommend using the neon gas lamps along with the current-limiting resistors and diodes to protect the circuitry. I do not believe a TVS diode can turn on fast enough to shunt the voltage.
Precision Analog Applications
Hello Pete Semig,
Thanks a lot for the response.
Can you suggest a part number for neon lamp, i found Multicomp neon lamp relatively bigger to put on 12 lead ECG on smaller PCB.
Unfortunately I do not have any neon lamp part recommendations for you. It would be very helpful to the E2E community, though, if you could post your recommendations given your design criteria.
Hello Peter / Sachin , actually i too have the same question , but when i see the datasheet of TVS diodes , i can see that they have response time of 1 pico second (Please see specifications for Littlfuse SMCJ series TVS diodes). So i am not sure if "Not Turning on Fast Enough" is the reason TVS diodes can not be used for defib protection -
Hence pointers to why TVS diodes can not be used for defib protection will be very useful to me as well?
I've been looking at the same protection issue and have the following to offer - TVS devices have higher leakage and higher capacitance, which can affect your ECG measurement. Gas Discharge/Spark Gap devices appear to have a higher transient current handling ability as well - there are designed for lightning strikes, which is closer to a defib discharge than ESD, which is where TVS' are typically used. Very small GDT devices are offered by Meritek.
Good luck, Doug
It seems I am sitting with a similar problem to you guys.
What I just need to know - is it necessary to have this current limiting resistor in front of your GDT? (The R1 in http://www.ti.com/lit/an/sprab36b/sprab36b.pdf)
If no, then how much voltage should the current limiting resistor R2 be rated for assuming 5 kV on the defibrillator circuit described in EN 60601-1? Again referring to http://www.ti.com/lit/an/sprab36b/sprab36b.pdf.
Yes, it's necessary to drop the voltage (potentially 5kV at the electrode, maybe 100V at the gdt --> 4.9kV) across a load. ECG leads are made (sometimes) with resistors in them to take this pulse (typically 1k, but they can be higher as well from what I've seen), using a material that can handle a very fast voltage/ very fast pulse (carbon comp?). The app note you point to notes a >1W power resistor, if you want to put it on your board. You'll need to check the pulse-handling specs for the resistor. I'm planning on having the resistors in the leads, so I'm coming right to the GDT.
At the moment I'm wondering how to avoid the AVDD+0.3 - AVSS-0.3V input voltage range spec on the ADS1298 front end...
Good luck, Doug
Seems like everyone designing a ADS1298 with defib protection will come here at some point ;)
I just wanted to add that the protection resistor also acts to comply with the 60601-1 clause on how much of the defib pulse energy may be sinked by the device... Yet another design constraint ;)Apart from this, I would be very interested in any solutions you guys have reached.
Cheers! // Rune
Having designed ECG monitors for many years I will try to describe Defibrillator Protection.
Devices that electrically connects to the human body must pass Defibrillation Discharge Compliance testing to assure it does not shunt energy. In addition; monitoring devices such as 12 lead ECG recorders and hospital monitors have to withstand the defibrillator discharge and continue to operate within 5 seconds. However, ambulator monitoring devices such as holters and event recorders are not required to recover after the defibrillator discharge.
The standard tests for compliance is a 5000V discharge at 360 joules across a 100 Ohm resistor that simulates the body resistance. The ECG device being tested is connected in parallel across the 100 ohm resistor, therefore to assure less than 10% shunt current the resistors used in the ECG electrodes should be no less than 1K ohm. This assures that even if one resistor shorts, there is still 1Kohm and less than 10% shunt current. The 5KV Defibrillator discharge voltage will damage the ECG device and must then be clamped down to a voltage level that does not exceed the device power supply rails. There are several ways to accomplish this but they all take 2 steps. The first step is to clamp the defibrillator discharge voltage down to a managable level, using neon lamps or high power TVS diodes such as SMBJ-14 or SMCJ-14. The next step is to add another series resistor and dual schottky diodes after the HV clamping component. One schottky diodes is connected to Vss and the other to Vdd so they will conduct if the input voltage exceeds power or goes below ground. The value of the series resistor must be chosen to limit the current through the secondary schottky diodes. The TVS diodes and dual schottky diodes should be very low leakage, less than 1uA.
I will end by sharing a general rule for ECG monitor inputs; Resistance, Inductance, and Capacitance is never a good thing in ECG monitoring inputs. The effects of this is filtering that causes phase delay at the amplifier. The phase delay differences due to the mismatches in components degrades the CMRR. The body as a lot of electrical fields on the skin so a Poor CMRR directly equates to a Poor ECG signal! Remember, the object of the ECG monitor is to reproduce "exactly" what is on the human body! Avoid anything that causes filtering on the ECG inputs! If you have questions or comments please add a response!
Bill Burris, Design Engineer, Sotera Wireless, San Diego CA.
Thank you for this very clear explanation. I have two follow-up questions:
1) How do you then manage to protect a stimulator? I wish to design a stimulator to supply <20mA constant-current pulses (using low-impedance electrodes/cables). Current-limiting (power) resistors would require a very high supply voltage. Do you know of any other means of 'voltage gated impedance' that reacts fast enough? (eg. How is this done in a compact pacemaker?)
2) Are there any integrated solutions to this? It seems like these techniques has not changed in a couple of decades?
Again, thanks a lot for your post :)
// Rune Paamand, stud. M.Sc. Medical Engineering, Copenhagen
1) I'm not sure how pacemakers protect against Defibrillator Discharge but I would assume a simular approach; The stimulator would use diodes that limit HV but do not effect the signal, with series resistors to limit the current and drop voltage. Yes the voltage is very high, Ohmite OX series 1K ohm 1W resistors with SMBJ-14 TVS diodes will withstand the voltage and power without effecting any signals up to +/-14V
2) I have not seen any integrated solutions for this on the market.
Bill Burris, Design Engineer, Sotera Wireless, San Diego CA
1) It's just that 20mA through 2x1kOhm + load impedance (<500ohm) will require a very high supply. I hoped for a 'non-linear'/less resistive way of doing it.
2) Ok. Maybe a future product for TI :)
Again thanks!//RunePs. I like the ViSi Mobile platform. Vital signs monitoring is an area with huge needs for improvement - and your system (including 'open' API's) seems like a leap in the right direction :)
In response to your last paragraph we have a more difficult issue with ECG monitoring as we are looking at fetal ECG monitoring. The fetal signal is approximately 1/100th of the the maternal signal. Any distortion or attenuation of the fetal ECG signal may mask the detection completely. So the question is how much attenuation will the defib circuitry cause to the fetal signal?
Also on a different topic what is the possibility of wireless ECG transmission from a MRI environment? Thanks.
Attenuation occurs when the HV protection resistors on the ECG inputs create a voltage divider with the differential amplifier input impedance. Attenuation from defibrillation protection should not be a problem when using high input impedance amplifiers. For example if the protection resistors are 10Kohm and the input impedance is 100Mohm the ratio is 10K/100M or 1/10K. So there would only be 1/10,000th reduction in signal amplitude. Defibrillator protection also involves using TVS diodes or neon lamps for clamping. These components leak very little current but they do have capacitance to be aware of. The HV input resistors and the capacitance of the TVS diodes create a filter pole. Generally the capacitance is very low creating a high frequency pole and this is still not an issue when monitoring low frequency ECG. For example if the input resistors are 10Kohm and the TVS diodes are 500pF they would create a pole = 2Pi x R x C = 6.28 x 10K x 500pF = 0.000031415 S = 31,830Hz. This pole is high enough to not cause any attentuation of the ECG signal.
As far as MRI monitoring; The wires in the ECG leads will conduct dangerous levels of current into the patient and have to be removed before the MRI procedure. There has been research into ECG monitoring during MRI procedures with some success in the pacemaker and AICD industry. But at this time I do not know of any external ECG monitors that can be used during an MRI scan. So even if the WiFi radio worked, the monitor cannot be used during MRI procedures.
I can share your statement above about the attenuation but I just saw a typo in the bandwidth calculation. It should be 31.8 kHz.
It should be remarked that the cable capacitances also need to be added to the TVS component capacitance. Usually a 10 kOhm series resistance with a total capacitance to shield of 1 nF (bandwith of approx. 16 kHz) allows the ECG amplifier to still detect and reject pacemaker pulses as required by ANSI/AAMI EC13 subchapter 3.1.4.
There are some commercial solutions for ECG and other monitoring in MRI environments, Invivo Corp. (now Philips) has been around for a while and is the market leader for MRI monitoring. The ECG wires work with distributed resistance (coiled conductors as described in US 6032063 patent for example). Some applications may also allow braided leadwires with carbon fiber conductors.
Bernd Richter GmbH
All content and materials on this site are provided "as is". TI and its respective suppliers and providers of content make no representations about the suitability of these materials for any purpose and disclaim all warranties and conditions with regard to these materials, including but not limited to all implied warranties and conditions of merchantability, fitness for a particular purpose, title and non-infringement of any third party intellectual property right. TI and its respective suppliers and providers of content make no representations about the suitability of these materials for any purpose and disclaim all warranties and conditions with respect to these materials. No license, either express or implied, by estoppel or otherwise, is granted by TI. Use of the information on this site may require a license from a third party, or a license from TI.
TI is a global semiconductor design and manufacturing company. Innovate with 100,000+ analog ICs andembedded processors, along with software, tools and the industry’s largest sales/support staff.