Seawater question

Thread Starter

benha

Joined Jan 4, 2011
71
Hi,

I'm working on a project that's going to be submerged in seawater. I'm trying to minimize the cables coming in and out of it. The microcontroller in it will occasionally need to be reprogrammed (USB). Likewise, the batteries have balance connector and a charge/discharge connector.

For the balance connector and the USB connection, I'm wondering if using something equivalent to a fully sealed pogo pin would work so I just make physical contact with a mating connector on the outside, or if the connectivity of the seawater would create challenges.

In other words, usb would be via four little gold nubs that are a couple millimeters apart and you'd just push a corresponding connector against it to make contact. Likewise for the balance connector. Power obviously would need an actual cable

I see solutions like this on dive watches and such, but I don't know if they're doing any magic to make it work.

Thanks!
-Ben
 
Last edited:

strantor

Joined Oct 3, 2010
6,782
Is this going to be permanently/long term submerged or just occasionally and then rinsed (like a dive watch would be)? How far will it be submerged (how long will the power cable be)?
 

Thread Starter

benha

Joined Jan 4, 2011
71
More like a dive watch. Depths of up to 100-150'. The discharge power is all internal. The recharge at the surface is just 2-3' long.
 

strantor

Joined Oct 3, 2010
6,782
Those are the real deal. We used connectors like those on that page, when I worked in subsea/ROV industry. They are not cheap. The best price I ever managed to find on connectors like that was from a company called Amron International and I think it was like $30 for cable-end plugs and $100 for bulkhead connectors, which is about half the price other places charged.

@benha does your project budget support this? If so then I second the suggestion. If not I have other ideas
 

Thread Starter

benha

Joined Jan 4, 2011
71
Not really. It's not as much the cost - though that is a factor - but the size. This isn't a wristwatch, but it's a small package. (inches square).
 

nsaspook

Joined Aug 27, 2009
13,081
Those are the real deal. We used connectors like those on that page, when I worked in subsea/ROV industry. They are not cheap. The best price I ever managed to find on connectors like that was from a company called Amron International and I think it was like $30 for cable-end plugs and $100 for bulkhead connectors, which is about half the price other places charged.

@benha does your project budget support this? If so then I second the suggestion. If not I have other ideas
Unfortunately if you want real seawater rated mate connectors they are costly because it's a harsh environment for any electrical signal.
https://www.bulgin.com/us/industrie...z5pVacKs9MXwoIn_5-iLDiwY5xHhtucBoCrt4QAvD_BwE

I maintained sensor systems deployed on ships. The tiniest leak was catastrophic quickly if the connector terminals where hot. Some systems were pressurized to mitigate these issues.
 

strantor

Joined Oct 3, 2010
6,782
What you can do, if your device is a sealed enclosure, is fill the enclosure with oil (mineral oil or hydraulic oil with no detergents or addittives) and make a threaded hole for a (pipe plug, or hydraulic fitting). Behind the plug is your USB connector or whatever. When you plug your USB cable in, you're stabbing it into an oil bath, so a little messy, but it's a cheap low tech solution that meets your space constraints. You have to make sure to settle out all the air bubbles and top it completely off with oil before you cap it. The idea is that, water won't come in because there isn't room for it because of the oil, and the enclosure won't crush because the pressure inside is (close enough to) equal the pressure on the outside. What can be a problem (I think, only at depths deeper than you're going) is components inside giving way to oil pressure. Electrolytic caps, then crystals. For this reason we only ever used oil compensated boxes to house terminal strips and transformers, but we were working at thousands of feet.
 

Thread Starter

benha

Joined Jan 4, 2011
71
Yeah. I was hoping to avoid the oil bath solution, but it's a good idea. I can machine covered ports and use o-rings too. I just want it to be super simple and have as little room for error as possible. That's why I wanted to get an idea if the relatively low voltages of something like USB or a single battery cell (all under 5v) would be problematic if they were terminated a few mm away from each other in seawater without insulation. Basically I'm curious if the dive watch charge/data connections are somehow isolated unless a condition is met, or if they're just bare terminals...
 

strantor

Joined Oct 3, 2010
6,782
Yeah. I was hoping to avoid the oil bath solution, but it's a good idea. I can machine covered ports and use o-rings too. I just want it to be super simple and have as little room for error as possible. That's why I wanted to get an idea if the relatively low voltages of something like USB or a single battery cell (all under 5v) would be problematic if they were terminated a few mm away from each other in seawater without insulation. Basically I'm curious if the dive watch charge/data connections are somehow isolated unless a condition is met, or if they're just bare terminals...
Seawater will definitely conduct electricity, most likely enough to cause problems. There's a way to find out, if your components are cheap... empirical data FTW.

another idea, just came to me so I haven't yet decided if it's stupid... any pin(s) that will have voltage on them could be ran internally through a Reed switch just inside the device. Place magnet on the outside in order to make the connection. High speed data might not like to pass through a Reed switch but (if it does, maybe it's not a problem) you could slow the data down. 9600 baud serial instead of light speed USB
 

Jim@HiTek

Joined Jul 30, 2017
59
How small? Could you use an inductively resonant xmitter/receiver? A small coil primary would be epoxy coated and you'd plug it into an epoxied hole with the coil secondary, held in place by strong micro magnets and friction fit. Sort of an epoxied pin and socket setup. The xmitter and recvr would run at 433 Mhz or so, the modulated/demodulated signals connected to uC's for decoding and saving. That wide bandwidth would allow lots of data. It would be a close but no contact RF system. With a constant carrier setup, that could provide power too.

Since it's under water, and sea water to boot, you might want to xmit the power that way too, like today's battery operated toothbrushes that have a power coil in the stand and/or like a modern wire free phone charging setup. Perhaps using a Li-Io battery in the sealed waterproof package would provide filtering and more short term power, unless everything is in the microamp range then just a cap could act as the battery. Doing things inductively would mean no contact with seawater at all. Entire package and umbilical could be sealed in epoxy.

Now that I think about it, that package might be a couple inches square so maybe too big for your plans? It would be kind of fun designing it: https://core.ac.uk/download/pdf/78897614.pdf
 
Last edited:

Thread Starter

benha

Joined Jan 4, 2011
71
This is really interesting. Hadn't thought about wireless. The power doesn't need to be transmitted that way. Can't really. I need to recharge a 4s 10AH battery in 60-90 minutes. But the comms stuff could work that way and it's one less penetration. Probably can't do the balance connector for the battery wirelessly without a mountain of work I don't want to do, but I could include a BMS inside the package so all I need for charging is a + and a -.

This is a great idea
 

strantor

Joined Oct 3, 2010
6,782
How small? Could you use an inductively resonant xmitter/receiver? A small coil primary would be epoxy coated and you'd plug it into an epoxied hole with the coil secondary, held in place by strong micro magnets and friction fit. Sort of an epoxied pin and socket setup. The xmitter and recvr would run at 433 Mhz or so, the modulated/demodulated signals connected to uC's for decoding and saving. That wide bandwidth would allow lots of data. It would be a close but no contact RF system. With a constant carrier setup, that could provide power too.

Since it's under water, and sea water to boot, you might want to xmit the power that way too, like today's battery operated toothbrushes that have a power coil in the stand and/or like a modern wire free phone charging setup. Perhaps using a Li-Io battery in the sealed waterproof package would provide filtering and more short term power, unless everything is in the microamp range then just a cap could act as the battery. Doing things inductively would mean no contact with seawater at all. Entire package and umbilical could be sealed in epoxy.

Now that I think about it, that package might be a couple inches square so maybe too big for your plans? It would be kind of fun designing it: https://core.ac.uk/download/pdf/78897614.pdf
Really cool suggestion!
 

Jim@HiTek

Joined Jul 30, 2017
59
Hmm, 10AH is what? 0.16667 amps per minute? A single layer coil wound on a 1/4" diameter ferrite stick all nicely coated with thin epoxy, with 1/2" exposure of the ferrite. Stick that exposure into a 1/2" deep epoxied hole with a coil surrounding it so basically a transformer coupled with the ferrite core. I didn't bother with the math, but I think a 1 KHz oscillator might be enough to transmit 200 mAmps.

Not sure about that since I didn't do the math but might be worth the bench time to test it out.

It's too bad you need such a large battery.
 

Deleted member 115935

Joined Dec 31, 1969
0
Have you considered, wireless
this seems to only need to be "talked to" above the water ?

If so , you can wireless charge,
and then you could use blue tooth to talk to unit

No need for any holes in the sealed unit,
 
Top