Sending a trigger pulse to a BNC connector, safety issues

Hi.

(1) I'm looking for a way to send a simple 5V TTL triggering signal to a medical device we're using as part of a study.
(2) The signal to the device comes in through a BNC connector.
One of the options would be using an Arduino Uno, connected to the computer sending the pulse via USB.
I was thinking about using a very simple program that waits for input, which then sets (say) pin 2 to HIGH for a brief period.
That pin and the GND pin would be connected via alligator clips on a cable ending in a BNC connector (or a clip-in BNC connection), which would then go to the device.

(1) You want to send a 5V TTL trigger signal to a medical device.
(2) The signal comes in (Where?) through a BNC connector (from where?)

Can you draw a diagram of what you want to connect from what to what please, even a picture.

A picture, a circuit diagram and a sketch is worth a thousand words.

Use REPLY rather than QUICK REPLY and it has an attachment facility, so you can post your files/images as an attachment.

Thanks Tom.... :slight_smile:

You should use some means of galvanic isolation between the device and the patient - such as an optocoupler. By this means the patient is completely isolated from the PC and arduino. Power for the 'output' side of the optocoupler to come from your "device" that is receiving the trigger pulse.

TomGeorge:
Use REPLY rather than QUICK REPLY and it has an attachment facility, so you can post your files/images as an attachment.

Oh, please don't (use attachments); because of the forum foul-up, that makes it very difficult for some people to read them.


XY problem. Please specify what the "medical device" is, what program you are using on the PC, what the purpose is?

Why would you want to use an Arduino? What processing does it need to do? If you only want the 5V control signal, a USB-TTL convertor module on its own can give you those signals, or a USB-to-Parallel converter can give you at least ten TTL control signals.

jackrae:
You should use some means of galvanic isolation between the device and the patient

He said a "medical device" using a TTL signal from a BNC connector.

How about a link And attachments? Many hosting websites are blocked from access for security issues.

Hi Tom, jackrae and Paul,

Thanks very much for the replies! (Edit: Ah, good idea CrossRoads, the system just said your comment came in. I've also added an attachment.)

Tom: I uploaded an awful drawing of the setup to Imgur, which I hope doesn't just make things worse; I also included a picture of the input port to the device, and the USB-to-TTL cable I was thinking of trying as an alternative to the Arduino. USB to BNC question - Album on Imgur

jackrae: The medical device itself (which gives electric shocks on purpose, for background) is all isolated. The concern is more the connection that would let us send a signal from the desktop's USB port, through "some kind of adapter" (which itself might not be safe, even if the shock device is), to the device's TTL trigger input. I think the current coming out of the electrodes connected to the subjects fingers is acceptably secure. Fingers of the same hand, to confirm, no current crossing the chest / heart!

Paul: this is for a psychological fear conditioning experiment. Stimulus-presentation software (Neurobs Presentation) is used to run a task on the desktop computer, subjects sit at that and use the keyboard to respond. That software also has to initiate the trigger to the shock-device so subjects get their shock at the right moments.

The trigger the device needs is a pulse from +3 V to +15 V for at least 5 microseconds, and it responds to the upwards flank. The device is set up to give the desired shock when it detects the trigger.

So if I were to use an Arduino, all it would have to do on each loop would be to read the serial input, check whether some code has arrived, and if so set a pin to HIGH for at least say 15 microseconds. That pin would be connected to the signal-wire of a BNC cable, and the GND pin would be connected to its ground. (I'm afraid I don't have the code for that yet, but I'm fairly confident about that part. I could try to program it already though if it would help. I know I can send codes to the USB port using the stimulus presentation software, I've done that before.)

I also ordered the FTDI USB-to-TTL adapter in any case, but that would still need to be connected to a cut-off BNC cable via jumper cables, which I'm sure will evoke concern. I guess all I can do there is wrap their connection in electrical tape? With that solution it's less a specific Arduino question and more a general "what could conceivably go wrong en route from the USB to TTL connections" question...

Thanks again!

Best,

Tim

USB_to_TTL_question.zip (559 KB)

I don't know anything about medical requirements, but in consumer or commercial applications anything below 50V is considered safe. Anything above 50V has to be insulated/protected so the user can't touch the voltage connections.

I've got my fingers on 5V - 15V circuits every day. You can touch the Arduino while it's running and you cannot feel any electricity.

If you touch the contacts of a 9V battery to your tongue, you can feel something so there may be stricter requirements for medical use where something is going under the skin or inside the mouth, etc.

As I understand it, nothing you are building will touch the patient. That should give you an extra measure of safety.


Of course, there's the possibility of your power supply shorting-out and power line voltage coming-out where it's not supposed to be, but that's unlikely if you buy a power supply approved by CE or whatever agency covers your country. Or, you can make it battery operated. And, I'm sure all of the power outlets in a medial facility are ground fault protected (which means if there is current flowing through a person, or anything else, to ground it shuts-down).

Where there are inductors you can get high voltage spikes. You've probably seen battery-operated "pranks" where a little battery operated device gives someone a shock.

Hi Doug,

Thanks for the reply and explanations! I'm reasonably sure it would be fine, at least no more dangerous than using a computer normally, or playing with an Arduino as you say, but since it's all very, very strict here I'm trying to nail things down as much as possible (as well as for my own understanding). For instance, OK, if there's a power surge or voltage spike somehow, could there be more than 5V on the Arduino or its connections? If so, have we precluded a power surge or spike either happening or reaching that far? Is there an inductor hidden on the Arduino somewhere that could affect anything? Is that circuit breaker on our power strip sufficient or do we need a different kind of safety switch? That kind of thing.

If there's some unavoidable unlikely but conceivable doubt we do have another option, to get dedicated "official" hardware from NI or BioPac, which would take away any concerns about amateur wiring. But I'd like to know if that would just be (expensive) overkill, or whether it would actually make a difference in the kind of extreme-case hazardous situation we have to consider.

Best,

Tim

I'm assuming your shocker has an opto isolator on the input ( it should ).
The ardquino should be in a metal box. A BNC through hole connector
installed.
The case should share the electrical ground with the medical unit.
A circuit breaker on the power strip is not a people safety device.
All AC power to the Arduino, PC, and medical device should go
through a common GFI.
Dwight

I believe you have already lost the potential law suit by identifying the "test subject" as a patient. You posting here will be a convenient trail for the FDA attorneys. Obviously you are not in a research institution or a licensed medical facility. They all have consulting engineers available to to aid in experiment development.

Local electrical code has little bearing on what you are trying to do. Who is directing your research? I am concerned because of the strict requirements for manufacture and use of any medical device on a "patient". Have you received written permission from the manufacturer of the device you are trying to trigger? I bet not. Their lawyers would never allow unapproved use of their FDA approved equipment.

Once we had a customer ask us about manufacturing some medical equipment. We cannot, because it requires a FDA approved electronic assembly service. More strict than military or aircraft related.

Paul

Any one else find it ironic that you intend to give a patient / subject / victim an electric shock and yet you are worried about him receiving an electric shock!

Hi Dwight,

Awesome, thank you very much! This sounds like something we could present as a possible solution.

If a follow-up question for clarification is OK, about the shared electrical ground - there's a connector with a circle-in-a-downward-pointing-triangle symbol on the shocker that I think is its electric ground (as seen on the bottom picture here: USB to BNC question - Album on Imgur). Is the idea that that should have a wire connected to it that's soldered or otherwise fixed to the metal box?

Paul, thank you for the warnings. We're definitely not going to do anything without the hospital's engineers. One of the reasons I'm trying to get more of a grip on things myself is to help talk about options with them; another reason is to understand as much as possible about whatever the lab set-up will be, if only to have more chance of recognizing anything going wrong. But the technical department will absolutely be making the final decisions on how to proceed.

Grumpy_Mike, yeah, there is some irony there :slight_smile: They just have to get just the right shock!

Best,

Tim

They just have to get just the right shock!

Any electrode connected to a person has to have a 1M series resistor. That will be enough to keep it non fatal from any excess voltage you can get from a mains breakdown.

Hi, Tim.
Looks like you are on the right track. Be sure to read the manuals for the equipment you are connecting to.

Relating to BNC connectors. there are 50 ohm and 72 ohm connectors. Difference is the size of the inside pin. The equipment manual should identify the type. Your connecting cable should match the impedance for the equipment. Otherwise reflections will distort your trigger pulse.

BNC will handle hundreds of volts without problem. If you are going to be putting the connectors on the interconnect cable, be prepared for an interesting session, and follow the length specifications on prepping the cable.

At one time there were BNC type connectors that were twisted onto prepared coax cable, but they were all 75 ohm, as I recall. Used in old LAN that I was in charge of. Don't use them.

Good luck.
Paul

Mike, there are definitely safety measures built in to the shocker. It's really more an issue about the rest of the set-up, but I think things are a lot clearer for me now!

Paul - thanks a lot! I'll keep that in mind, good to know, and will definitely read the manual (at some point it'll be time to test it on a person and that'll probably be me, so that can only help the motivation to read carefully :)).

Best,

Tim

Just pushing wires into Arduino pins isn't the most secure of connections. Neither are crocodile clips. Connector block is a much better alternative.
Get yourself some male header strips and solder the wires to the top of those.
Ideally, solder the ends of the BNC cable to them and use a cable clamp to prevent the connection from being strained.

And use yourself as the test 'victim' :slight_smile:

I don't know anything about medical requirements, but in consumer or commercial applications anything below 50V is considered safe.

I believe the Fire Marshall allows 35 V and below (AC or DC) without any special approval .

if that bnc connector is on the shocker.
Its there for a reason.

The manufacturer is the ONLY safe body to confirm how it is to be used.

TimTaa:
it's limited to generating a 5V potential no matter what comes out of the USB port?

Use a battery to power it.

And use an optoisolated usb cable to the pc.

This abstract may be helpful.

Opto USB are or were available for about 200 for the purpose,

If you use the fibre optic type check on the power requirements as i think but do not know that they require power at both ends, the arduino will not supply its own end.