What's inside a dial-in pacemaker monitor system?
Datasheets:
LMV824 http://www.ti.com/lit/gpn/lmv824
MAX4330: http://datasheets.maximintegrated.com/en/ds/MAX4330-MAX4334.pdf
CNY64 Optocoupler: http://www.vishay.com/docs/83540/cny64.pdf
ADSP-2185N DSP http://www.analog.com/static/imported-files/data_sheets/ADSP-218XN_SERIES.pdf
CX88168 Smart Modem with Voice CODEC http://pdf.datasheetcatalog.com/datasheet/Conexant/mXrqvxu.pdf
MAX994 10bit ADC http://datasheets.maximintegrated.com/en/ds/MAX985-MAX994.pdf
Meder Reed Sensor: http://www.meder.com/fileadmin/meder/pdf/en/Products/Reed_Sensors/Reed_Sensor_MK06_E.pdf
Probe shock video showing the piezoelectric effect: https://www.youtube.com/watch?v=KFCRB4d991E
Forum: http://www.eevblog.com/forum/blog/eevblog-656-pacemaker-monitor-teardown/'>http://www.eevblog.com/forum/blog/eevblog-656-pacemaker-monitor-teardown/
EEVblog Main Web Site: http://www.eevblog.com
The 2nd EEVblog Channel: http://www.youtube.com/EEVblog2
EEVblog Amazon Store (Dave gets a cut):
http://astore.amazon.com/eevblogstore-20
Donations:
http://www.eevblog.com/donations/
Projects:
http://www.eevblog.com/projects/
Electronics Info Wiki:
http://www.eevblog.com/wiki/
Datasheets:
LMV824 http://www.ti.com/lit/gpn/lmv824
MAX4330: http://datasheets.maximintegrated.com/en/ds/MAX4330-MAX4334.pdf
CNY64 Optocoupler: http://www.vishay.com/docs/83540/cny64.pdf
ADSP-2185N DSP http://www.analog.com/static/imported-files/data_sheets/ADSP-218XN_SERIES.pdf
CX88168 Smart Modem with Voice CODEC http://pdf.datasheetcatalog.com/datasheet/Conexant/mXrqvxu.pdf
MAX994 10bit ADC http://datasheets.maximintegrated.com/en/ds/MAX985-MAX994.pdf
Meder Reed Sensor: http://www.meder.com/fileadmin/meder/pdf/en/Products/Reed_Sensors/Reed_Sensor_MK06_E.pdf
Probe shock video showing the piezoelectric effect: https://www.youtube.com/watch?v=KFCRB4d991E
Forum: http://www.eevblog.com/forum/blog/eevblog-656-pacemaker-monitor-teardown/'>http://www.eevblog.com/forum/blog/eevblog-656-pacemaker-monitor-teardown/
EEVblog Main Web Site: http://www.eevblog.com
The 2nd EEVblog Channel: http://www.youtube.com/EEVblog2
EEVblog Amazon Store (Dave gets a cut):
http://astore.amazon.com/eevblogstore-20
Donations:
http://www.eevblog.com/donations/
Projects:
http://www.eevblog.com/projects/
Electronics Info Wiki:
http://www.eevblog.com/wiki/
Hi Welcome to Teardown! Tuesday This one was sent into the mailbag segment by Matt so thank you very much! Matt And it's a remote pacemaker monitoring device and we saw a little bit of it on the mailbag, but it's basically a custom device designed for us and Jude Medical Center it's the house core plus Transmitter Modern Thirty-one ATT and I couldn't readily find any info on it or anything like that. so I think that's why I Do think it was that custom manufacturer that says manufactured for them anyway and what it does is it presumably allows you to call up the hospital, call up the doctor, or whatever you know there's a phone number in there. there's not, but there should be and you dial up. There's a phone port on the back and you can talk to someone and then you can put on these wristbands.
Here you can use this inductive pad to communicate or sit in something to your pacemaker. so you stick this on your chest and you can talk to someone while you're doing. They could probably give you instructions, you know, relax, breathe in, breathe out. I Don't know if anyone has a pacemaker and actually uses one of these things.
Please let us know a bit. Yeah, it should be rather interesting. It's got a phone line interface and because it is a medical device, it will have different certification requirements in terms of like the power adapter and stuff like that. So this pair adapter here is not just a regular power adapter, it's well, it might be, but it's been certified for medical equipment.
So presumably it's got like better isolation and better type-approval and testing and stuff like that. Presumably that's what that cardiac symbol there is so that uses are five volts at five watts there. So anyway, should be interesting. Let's crack this thing open.
And by the way, if you haven't seen some pacemakers here, they are. This is what gets implanted into you. and well, I'm not sure how old these particular ones are. They're probably a more modern these days, but these were sent in to the mailbag by a man who a long time ago sorry, man who haven't got around to tearing these things apart.
They are like ultrasonically welded around the outside you can see, so they require, you know, and probably put it inside or something like that. So I require quite a bit of teardown. I'm not sure if there's still got a battery in there and can be powered up and not even sure if they're probably not even compatible with this thing. All the different brands and brands would have all their own that protocols and things like that, but anyway, um, these are the pacemakers.
and yes, these ones have been removed from somebody. They have been thoroughly cleaned and sterilized, so it's okay. But yeah. I've still got these for a future teardown, but that is what basically sits under your chest.
and then you come along with this inductive pad like this, and presumably there's a matching resonant coil in there. and you can actually communicate to the device and get some sort of diagnostic. or maybe even reprogram it or something like that. Not quite sure what to expect inside here. There won't be a lot like there won't be huge amount of processing. There's no big display on the thing, so there's no point out pairing up unless we wanted to see actually what we got out of the inductive coil to see if it actually, yeah, transmits anything, but may not. Unless you're making connection and stuff like that, it may not actually do anything at all. So anyway, and there's four screws here we go, so there might be a huge amount.
There will be some processing, of course, and some sort of comms for the phone line. And here we go. That's some light pipes. There we go.
There's a little light pipe for a run, diffusing and getting the LEDs out. That's a nice touch. They've gone to a bit of trouble there and then we go and we've got a speaker. They've got some nice, uh, acoustic up? Well, yeah, it's a little.
They've gone to a bit of trouble to make the acoustics a bit better inside this thing so you can look. The idea is that you dial up and you know you call the hospital, call the doctor or whatever and they talk to you and you can use this hands-free You don't have to, you know, sit with you and with the phone up against your head and stuff like that so it look nice bit of strain relief in there. Check that out there we go. I've got a little nice little rubber grommet and they've curved the cable around like that on both the left and the right sensor wrist strap, so that's that's.
quite neat. and they've done exactly the same thing at the top here with the Ah, there we go. That is how you do strain relief properly. That is really good.
Doesn't get much better than that or some huge thumbs up to the designer there. And here's the volume slider on the side. It's just like an exposed carbon track one like that. I Don't particularly like those because they can get gunk in them, but this is fair.
You know, pretty much a sealed unit so you know it's not like this. vent, holes and fans, and everything else actually collecting dust in this thing. So yeah, that's adequate for the task. Here's the main: PCB I'll get in there and we'll check out the individual chips, but it looks like we've got two main devices here.
We've got some firmware there and you know, some line interface our stuff over there. Then we have a closer look at the board. I'm some transformer isolation here going around to the pads. Of course you can actually see it this top.
LED board out here you can see the isolation look right going around there. it's covered by that sticker down there. But look at the huge ground plane isolation running under that transformer there. And if that transformer isolation wasn't enough built-in braces, of course, when you got something like this, you're gonna put a bigger series resistor in there. Check it out. 330 ka pop and they aren't mucking around there. Look at the isolation on that so nothing's going to go wrong and arc over that sort of thing. You have huge big 330 K power resistors in series with each one of those sensor lines and that is how you designed medical devices.
and presumably you'd have to do that sort of stuff to pass our compliance for isolation because you know this is used by the user in their home. And of course you have these two metal straps connected to each arm. Look at that. That has to be the most de Duras electrical body connection scenario for electrocution that you could possibly have.
So no wonder you have to have the series resistors, the transformer isolation, the ground playing isolation, everything else when you hooked up like this. I wonder if these even have a series resistor in them like you? I Get on the anti studied wrist straps and we can check for that. Here we go: No, that's a dead short so there's no series resistor in there like you'll get with those are ESD strap. So typically we have a one Meg resistor often embedded into those things.
so you have direct electrical connection from both the wrists on each arm into directing the circuit board. That's why you got the big R series resistors, the big-ass transformer, and the nice ground plane isolation beauty. If we have a closer look at the sensor pad electronics here, let's go in and see what we can see. We've got an LM V 8 to 4 that's just a precision chord our pant.
We've got a max 4, double, 3 for another precision cord Op-amp 3 9, 3 comparators, HCT 174 and yeah, a few other comparators and well, not much else. So that's about it for the sensor pad connection. So it's basically just amplifying the differential signal probably I'm amplifying the signal from the sensor pads and then not coupling that back over the transformer there. and I Couldn't find any ready data on that transformer.
But here we go. We got an Opto coupler here. It's a CNY 64 series from Vishay and sure enough, that meets medical devices our compliance requirements. In this case, our video 7 500 and IEC aren't 6600 one medical standards.
So yes, specifically designed for medical stuff. Now it's interesting to note how they actually did this front panel and what effort the designers went to look. They got this vertical header. This is the user interface.
They got the header on the back of that and there. so the pins on the board that sits in there like that LEDs shining up vertically and then they've the right angle light pipe going across like that. That's a you know, that's a lot of effort to go do to build that second board and populate that whole different assembly design and get custom-made the light pipe and everything when you really could have just you know stuck the LEDs out the front and been a bit how you're doing with it. but this is like a really professional. They've gone to a lot of effort to do that and more attention to detail here that they haven't just you know, slept on the microphone. They've actually wired over art separately, use their own, used its own connector, and they've mounted it in a little rubber arm shock absorbing holder - awesome. No real surprises for finding analog devices DSP in here. a DSP R21 85 series 16-bit DSP A team.
It's pretty beefy for something like this. I Kind of expected to find. You know a reasonable amount of that processing in here. Not huge, but of course they have to do some math, sensor conditioning, and stuff like that.
so you know a DSP is really good for that sort of stuff. So there you go. There's the external firmware up there for it, and that's about all she wrote. It's probably not processing the voice from the microphone or or anything like that from the actual core.
The reason: You can tell that is because the microphones all the way over here. We've got this other mysterious device over here which is almost certainly coupled directly into that. so you can tell these sort of things just by you know, physical location on the board. So this is obviously the you know, the telecoms Art processor which handles the voice and the connection and data and well, it's probably just getting.
Now you know serial data from oh you know, serial parallel data from the DSP over here, over to it and then transmits the data back. That's about all she wrote, so see if we can get some data on that. And yep, I was right, are no surprises whatsoever. This is a Kinect --nt CX 88 168 Basically a smart modem in a chip with voice codec capabilities.
hence the microphone input directly on there and it can do like answering machine functions. So it's a full V90 V.34 modem so can do you know you're old-school 56 RK bits per second modem interface and with voice as well. So you know where to be driving the speaker directly, microphone speaker and accepts data from the DSP processor up the top. and well, that's all she wrote.
You wouldn't roll your own solution there. You can buy these off-the-shelf designed specifically for this task. Uh-huh I wasn't hundred-percent right on that. This main our Smart modem over here yes is the main chipset, but it doesn't have the voice codec built in.
It has a companion chip which is optional which and they need in this case of course. And there you go, the 24 37. That is the voice codec part of it. This one's actually a pretty powerful beast.
It's got a microcontroller building with one megabit of RAM and two Meg bit of program flash memory as well. So quite a powerful beast. And there's our speaker connector and Driver over there. and once again with the isolation.
look at this. you can see the isolation start over here from the telephone interface here going across. So we've got some suppression happening there, some AC suppression, but there are high voltage withstanding readings. And then you've got the isolation with your opto coupler coming across here and that thing is completely isolated. And that's let's take a look at that meter electronic relay up there. And of course, when you're on a winner, you're gonna stick to all the same chipset. So once again, connect stand. That's the 24 6-3 line side device they call it.
So that's got a codec in there and handles all the modem line side interface stuff. So you can see that on the block diagram here of how all these pieces go together. Although this device, although the main device, the smart modem looks like a different one they're using. but the line side device and the voice codec is designed as part of this.
You know, one big package and that's what you're looking for. These manufacturers specifically provide this sort of stuff so that it makes you know custom designs like this. Really quite easy. just off-the-shelf app.
Note stuff. Pretty much. We got ourselves an unpopulated cereal or other sort of comms interface here. Look at the big isolation slot they got under the transformer there, that's not there and they haven't populated this.
There's a JTAG presumably that's the programming header. Of course it is the programming header for the DSP And yeah, we could probably try and get some. You know, if you really wanted to hack and have a little reverse engineer a play around with this, you could try and find a serial interface or just get in there and and play around with the thing and then over here. of course we've got power supply a little like PCB mount a heatsink here.
Nothing much happening there. No huge power requirements on this thing, so it's probably adequate. Nice. Schaffner Common mode at Chokey And if you're wondering what, that little meter device down there it looks like a little coaxial relay, but it's not.
It's only two terminal, but it is actually a reed relay. It's a magnetic switch where instead of having a and coil in there to actually activate it, you extract available an external magnetic field. So presumably it cuts off our power to the sensor coil. It's just a so you know just the proximity to it there.
But the way it actually works is in the case. Here There we go. They got a little cutout and that sits when you fold that in. Oh sorry, I'm gonna mess of wires here.
But when you when you fold that in that actually sits on top of there. and where's the magnet? Well here it is on the case Tada. So when this case closes, shut like this. Bingo.
It goes down and disconnects the power. In fact, that may not be. Come to think of it, the sensor may be actually disconnecting the power to the entire device. There's really nothing worth writing home about. There we go, other than MLM V8 to force as well. If you're gonna use them in your Bill of Materials, well use them bloody well everywhere. And if you're wondering where the ADC is for the sensor arm bands, here we go. The Mac's are 12:48 for channel 10 bit, I Am serial ADC and that looks like some sort of you later I'm not going to look up the number or some sort of regulator or voltage reference perhaps.
but I can't find a DAC on here at all for drive in the the inductive, you know, pad up there. So really, um I guess they're probably just driving it. you know, digitally like you know, square wave. There's no sort of, you know, like analog modulation or anything else are going on.
presumably. All right, let's just do a quick power up and see what we get here. Woohoo! I've got some LEDs I Can hear the speaker are sort of punching in and out there. We got some LEDs flashing.
There's no oh, there we go. green flashing green. whatever that means I got no idea. but yeah, I'm not gonna hook it up to a line and try and try and call the number or anything like that solid green.
that's gonna be good, right? Went from flashing to solid green. So there you go. Let's see if we get anything at all out of this inductive pad. Maybe not, because as I said, it might have to actually connect first before it actually generates anything on this at all.
All right. I've put the sensor pads on my wrists and just to see if it, you know it does anything but no zippity-doo-dah Hmm. Well, I didn't really expect it to do anything though. Okay, and I'm just gonna use my scope probe are shorted out here because that's we can do.
it's not. it's well, at DC its shorted out. but at higher frequencies this actually works as a one turn pickup loop and you can use it as a crude pickup. If this doesn't work, then I can go to you know, more turns and get a bit more creative and things like that.
But anyway, it's on and it's not connected. and I got there 5 millivolts per division and there's nothing much happening there at all. I'm afraid that's just boring, so no, there's nothing significant. See a few blips in there I Believe that's from my flurry lights here.
So yeah, you can see a bit of crud happening there. but if I turn my lights off, does that go away? No, it's there. So, but you know, look at common mode noise. pick up whatever I see so you just saw the big big poll says so that's obviously not coming out of the loop and I can verify that by actually taking it off.
there's nowhere near the thing I can actually even D power the device so completely there we go and we still get that crap so it's nothing to do with that. It's just, you know, pick up from the lab here. Oh, by the way, if you're wondering what that weird low frequency waveform was there before I've shown this before, look, watch this. Look at that just by banging your probe on the table. That is the piezoelectric effect. Anyway, if you want to see more of that, I'll link in a video here and you can check that out. So there you go. I mean I could try and knock hoax this thing into.
you know, transmitting something in here, perhaps? But yeah, I don't think it's hugely worth it I could open up this thing I Think there's you know there's probably just going to be a coil of wire inside the thing. Nothing fancy at all. It could be completely potted. That wouldn't surprise me, so it might be a pain in the ass.
I Might try and dig into it, but don't expect to find much at all. Well, well, well, what do you know? what sort of voodoo is happening inside here? I Expected to find a, you know, an inductive coil, but no. Looks like we've got something else that looks like the shield of the cable in there connected to this. So this is actually a three wire interface.
Look, You can see the green wire up here is connected through to the shield. so we've got the black and the red has to be in here somewhere doing what though? And if I lift the skirt up on that Tada, No surprises There we go. There's our inductive coil. Not a huge number of turns I'm not sure how many turns in there.
you know, a few dozen large gauge enamel coated wire, but yet nothing fancy. So I'm not sure exactly you know or what that and that pattern is doing there. but anyway, it's You know it is quite deliberate. so yeah, sorry.
I couldn't get anything out this puppy at all. but yeah, like I could probably spend enough time and it coax it into maybe generating something. but it's just. you know, it's not hugely worth the effort unless you had a specific need.
So yes, sorry. And we did some basic checks and nothing came out of it so oh well. Can't always win. So there you go I Hope you enjoyed that tear down.
Tuesday These medical wired devices are usually are quite interesting and nothing you know. it's pretty much what I expect. It's got a DSP in it. it's got a coil of wire to inductively couple over resonantly inductively coupled, probably over to your pacemaker and well.
and it's got line interface. but it's really well engineered. really well designed inside here, so they really knew what they were doing. and a classic example where you know as some company was hired, some engineering firm was hired to custom manufacture this I don't know how many you know they might have made a couple of thousand of them or something like that could have been more I don't know how big us and Jude Medical you know thing is and how many patients and stuff like that they've got.
but presumably like every year patient with a pacemaker maybe gets one of these so it could be could have made them in the tens of thousands or something. but really well engineered. Classic example of where you know the designer just uses all the off-the-shelf art chipsets for the line interface, for the codec, the modem, everything else. and really, you know there's kind of. There's custom of course, sensing circuitry for the pads and Driver and stuff like that and the rest is all in the firmware and how it talks as a system and some really nice touches inside this thing so that's not bad at all. So thank you very much Matt for sending that in to mailbag Monday that's a yeah I'm always an interesting not tear down when you do what medical devices like this so if you got any more info I couldn't find like a a manual or anything like that let alone specs for this thing. but if I do I'll link them in down below. As always, all of the data sheets for the chips that we have are seen in here will be linked in down below as well.
So if you want to watch, check out the data sheets then please do. And as always, if you like teardown Tuesday please give it a big thumbs up because that helps a lot on YouTube rankings and all that sort of jazz. And if you want to discuss it, jump on over to the Eevee blog forum also linked down below. And don't forget to subscribe.
If you haven't subscribed, what are you doing? Catch you next time you.
Suspect the electrode input amplifier section would be some sort of active band pass filter. I recall when I was making my EMG amplifier the frequency range of interest was 50Hz to 200Hz, anything outside this you would want attenuated
I think the cardiac symbol means it's defibrillation proof
those medtronic pacemakers are made in my hometown in maastricht in the netherlands
I would love to see you talk about a modern motherboard 🙂
Aren't the "Voodoo" fingers on the white plate the reflector, since it's connected to the shield of the coax. A directional antenna with the coil side held to the patient/pacemaker?
Those wrist bands look like they came from the Timex Classic Digital watch model number T785879J
Put in the shelf for now. It might come in handy for the last of us after the zombie apocalypse.
When i seen Conexant chip i knew it's a modem, because i got same chip (but diffrent model) in my old Zoltrix 56k modem ;3 I didn't know about voice codec chip, i looked at it and it has it too.
IMHO, the finger on the back of the wire loop is some kind of shielding, they're not using a solid copper plane or a copper mesh because they don't want to make a shorted copper loop right under an antenna, that will ruin your antenna.
EEVblog What's interesting is that in spite of all the "Certified for medical use" wank, these pacemakers can be read and updated without any authentication whatsoever. Extract sensitive medical information, make it give shocks or put it in test mode making the battery run out and fail when the patient needs it. I encourage you to look up Karen Sandler and hear her experiences getting one implanted and trying to figure out what it really does.
Open the pacemakers! Mike did on his channel and got a nice surprise when he cut one of the welds 🙂
I put the sensors on my wrists and lets see if it will , oh it's tingling, oh no ahhhh ahhhh i'm getting shocked, ahhhhh shit its defective, I think my heart is gonna explode, ahhhhhhh call ahhh for ahhhhh help ahhhhhhhhhhhhhhhhhh!!!
Medical devices are expensive because they spend so much on design and making a device that is made extremely well both for reliability and continuously performing precisely, and they have to be designed on that high level too to really be considered well made and worth the cost. Attention to all these details I think are completely expected, at least in my eyes it is. I'd guess based on the video that you'd tend to agree too.
Very nice. I've got a Medtronics version for my pacemaker. Saves a trip to the doctor every 3 months.
I've just started my first electronics class! Thanks to your videos Dave for making EE so interesting!
I'd be careful with newer pacemakers as, IIRC, many are powered by nuclear batteries. I had a few old ones I picked out of a bucket in an abandoned hospital. Yes, a bucket full of pacemakers.
Good video. You need to use a Dremel with a cutoff wheel to open up those pacemakers like a clam shell. There should be a lithium Iodine battery inside. That is another video. 🙂
Awesome video – as always 🙂
do the power supply please
Donate to St, Judes, it is the best Hospital in the world, No child turned away, No charge to any child or child's parents.
Cancer Hospital created by Marlo Thomas father to cure children with cancer.
I have a retired EOTD (End Of Train Device). It was involved in an accident and was recently retired. It has the battery removed for obvious reasons but it does power up and transmits data bursts on 451.5375 MHz. I would send it to you in Australia but the shipping cost may be too high as this thing has some weight to it.
I believe the strange pattern on the coil cover is a faraday shield.
ten episodes from now, do a teardown of some vintage tech that was/is absolutely hell to use.
I have a little bit of background on this type of electronics having been a contractor for a device maker. The coil reads, and on Guidant devices, could write pacing information to the implanted device. All implanted devices "spoke" their own protocol and none are interchangeable between different manufactures; even some lines were incompatible (latitude and prizm for example). When a device is interrogated and data on implanted device health and, in later models, pacing information and cardiac anomalies are read out. A cardiologist could read the ekg and modify parameters of the device and those would be written back. Devices typically fall in to two categories a pacing device and a defibrillator device. In rare instance you will see both.. One will put out a stream of pulses to keep the heart at a given rate and the other delivers a shock to the heart like you see on TV. As far as I am aware no charging of devices ever took place with a coil/wand. These "wands" were leapfrogged by the Guidant Advanced Patient Management devices which interrogated devices using RF circa 2004.
I have a Medtronic pacemaker and was given a similar remote monitoring device – some people require check-ins more often, but I only get checked out twice a year – one office visit, and one telephone. The Medtronic runs on 4 AA batteries and does not have the wrist straps. The inductor is basically a RFID type reader that will trigger the pacemaker to transmit data, the monitoring device then sends the data to a central Medtronic database that the doctors can access. The whole process of reading from the pacemaker and dialing into the Medtronic computer and transmitting the data takes about 5 minutes.
I once came across something slightly more exciting to tear down – a Physio-Control Lifepak 5 EKG/defibrillator in working condition with the defib paddles! It had been discarded in a dumpster, was full of water from recent rain, and had not been calibrated/certified in a LONG time according to the sticker on it – but the internal circuit boards were well sealed and it was not damaged. I have not obtained the correct EKG leads for the machine, but using the defib paddles confirmed that it does indeed pick up a heartbeat just fine. Truly some impressive construction inside that thing.