In the beginning… Part 2

Our discussion resumed the next day over morning coffee:

I began: “You know Trish, I’ve been thinking…”

“Uh..Oh..” she injected, with an impish grin.

“No seriously,” I replied ” The loss of the latest hydrolab has me thinking about all the kit you’ve used over the years, and how, in one way or another, it was never quite the right fit for cave research. Like the old Aanderaa RCM’s you borrowed from Southampton for your PhD.”

Trish (working her PHD in 2001)  with RCM7 & OTT

Trish with RCM7 & a modified OTT Nautilus with a custom housing.

“Absolute beasts.” she agrees, “Almost twenty kilos negative! Even with lift bags to get them into the caves, we often couldn’t install them because the cave walls were so friable that they would just break under the weight. And in our caves, the pipe flow gradually slowed down and reversed direction twice a day. That’s just totally different from open ocean flow that has more than enough strength to push the fin around. So it’s not surprising that they didn’t spin on their gimbals properly.  The worst though was the cut-off at about 1 cm per second, so the data was pretty choppy in the quieter caves. But that’s what you get with something designed for deployment in the open ocean.”

“Right” I continued, “And if I recall, the batteries were expensive custom cells, and there was the potential for stuff to fowl the rotors. So then you went up scale to the yellow ball. What was that again?”

“It was an InterOcean S4“, she described, “A lovely electromagnetic flow sensor. But it was still the size of a  beach ball, and worth a few grand. I bit my nails every time I deployed the thing, for fear someone might just walk away with it.  I must have been pretty persuasive for Southampton to let me borrow that thing for fieldwork in Mexico.”

S4 ready to deploy in the cave.

S4 ready to deploy in the cave.

“Wasn’t there another really expensive one?” I inquired.

“Ummhmm. For a while we tried the Falmouth Scientific ACM. I had really high hopes for their acoustic doppler. But we had problems with the signal. Some of those systems are really tight; so it could have been reflections from the cave walls.”

“So did anything really work in the underwater caves?”

“Yes, remember the little OTT Nautilus?” she asked.

I respond, “The stream unit we converted for underwater work. The yellow duck billed thing?”

“Yep, the dive housing that Andres from Germany made for us turned it into a really fantastic piece of kit. Dang!” she adds with a frown. “I really miss that thing.”

“Uhhh, I’ve been laboring under the assumption that it was just hiding under some pile of maps in your office… What happened?”

She explained, “It disappeared in a piece of lost luggage on the last Northwestern student trip I led to Mexico.  I did receive an insurance settlement, but since Ott had almost gifted their demo unit to me, the settlement was nowhere near enough to replace it. I’ve been keeping an eye on the forums, eBay, etc. But they don’t come up often, and even used they are still not exactly what you would call cheap. And we still have to recreate that housing, since the display end was never designed to be submersible. So it’s not going to be that easy to replace. And it had no facility for logging; remember writing all the data on our underwater slates when we calibrated each cave site?”

I pause for a moment, and then ask “So give me a wish list.” I suggested. “If there was a perfect piece of equipment out there for your work, what would it be?”

It took a lift bag just to move the RCM units.

It took a lift bag just to move the RCM units.

“Well…”she began, and I could see a beat up hydrogeologist fieldwork hat materializing on her head: “It goes without saying that the unit has to measure how fast the water is flowing, and in the cave systems, those flows can vary quite a bit: from fractions of a centimeter per second all the way up to 20 cm/s at the coastal discharge sites. And it should be small enough to deploy when you only have a meter or so between the ceiling an the floor, plus light enough that I am not hauling boat anchors with me on the dive. So something the size of a baseball would be perfect if I was trying to set out a network of them.”

“A compact, light weight, water flow measuring device. Any other sensors?” I ask.

“Oh yes, I’d want CTD to put the flow rates in context. And ideally I would want the unit to log for an entire year, so that I can compare the dry season and the wet season in the same units data set.”

“Ok” I confirm, “So a completely submersible data logger, with sensors that measure water flow plus conductivity, temperature and depth for an entire year. Anything else?”

“Cheap as chips!” she replies, reminding me of our time in the UK. “That would seriously reduce my chances of ‘accidentally’ strangling my students the next time they return home with a flooded unit” We laughed a bit. “But seriously, take a couple of zeros off the price of those commercial units, and I could deploy them all over the place; characterize an entire system! The nice toys produce really high quality measurements, but only for one single location, or one single dive.  And you can only extrapolate so far with data from one or two monitoring sites, no matter how accurate the numbers are”

“Right” I say, as I pretend to check off a list. “So less than..say..$500 each? What about other issues… that might need to be addressed?”

“A year is a long time under water. So you are bound have corrosion on anything made of metal, so generally speaking the less metal used the better. And you get organics drifting by in some systems, so anything that could foul up with a leaf, or a bit of seaweed should be avoided. And there is also the basic chemistry of the place to deal with. There is plenty of calcite raft deposition going on. So the perfect design would keep on working after it became encrusted with accumulated bits of crud.”

So I clarify: “No metal, and no moving parts?”

Another O-ring failure.

O-ring failures lead to incredibly fast corrosion anywhere that two different metals meet.

Trish doing field repairs.

Trish doing field repairs on an RCM9.

She replies: “The moving parts bit is not absolutely required, as the RCM7’s did run for a while before they stopped spinning. But it would still be a good idea.  And that reminds me of something else: you don’t want something that uses any kind of custom battery. I get enough grief flying with research equipment as it is, without also running afoul of TSA regulations on some weird  Litho-Phospho-Unobtainium batteries . And while we are on the subject of things that matter, Easy to repair in the field should be on the list. Eventually, everything floods, which usually means waiting for custom part X to ship from half way around the globe before you are running again. It would be nice to just go down to the hardware store to fix something for once.”

“Sure”, I respond, as I pretend my invisible list just got long enough to run off table. “Standard batteries, and fixable with off the shelf materials. Anything else the queen of hydrogeology might want under the tree this Christmas?” I am being a bit silly at this point, but only because she was being serious. But in my head I was starting to hear echos of all the three legged stool comments that tech support folks “ahem” so often make.

“Yeah, I am sure if you give me more time I could come up with half a dozen other criterion. But even that list of things does not exist in one single device, or I would already be using it….what are you thinking?”

Trish wiring up one of our home made drip sensors for her post-doc research (2004)

Trish connecting one of our home made drip sensors.

“Well,” I cleared my throat, “You know how Mike and Steve have been telling me to look into the open source micro-controllers, like the Arduino’s you use with the students in your instrumentation course?  And I’ve been reading quite a bit about the makers movement, and 3d printers, and, well… I just though perhaps I could give it a go. You know, bodge something together, like we did with the drip sensors in Vancouver, but this time I try to do the whole thing, including the electronics.”

“Mmmmmm.” she replied,” And this would be Ed’s new project number…?”  with eyebrows raised to remind me of the many other projects I seem to have lying around the house…in various states of completion. And to be fair I guess that last part did sound a bit hand-wavy, even for me. But now I had a well defined challenge to sink my teeth into, and I knew that, if such a device could be made, it would open up some serious research opportunities. So I thought about it. In fact, I was rather obsessed with the idea for several days. I scoured the web, reading everything I could find about flow meters. How they were built, and how they worked, about Arduinos, data loggers, and anything else I could find that was relevant to building something that could even do half of the things on her list.

<—Click here to continue reading—>

3 thoughts on “In the beginning… Part 2

  1. Mike

    It has to be sealed in an air tight container to protect it from the water right? The reason you get leaks is the gaskets fail and water gets in right? I have no idea if this would work but you can run most electronics completely submerged in mineral oil. Lots of over clockers do this for their entire pc systems. I was wondering if you could fill your containers with mineral oil to remove the air. The pressure difference across the gasket would be much less and if you did have a failure the oil should repel the water. i.e. the electronics might actually be salvageable. I have no idea if this would be of any use to you but it’s just a thought.

    1. edmallon Post author

      I don’t know if anyone has tried it with a commercial Hydrolab, but that’s a great suggestion.

      Often when the electronic design is settled, people pot the entire board in epoxy. However, the way my tiny boards are stacked, combined with my use of memory cards, prevents me from going that far yet. However the mineral oil idea would be worth trying while I am still changing everything around..thanks Mike!

    2. edmallon Post author

      Mike, Just a little followup to your earlier comment. Several of the DIY ROV builders have experimented with potting in wax. If I can figure out how to maintain access to the SD card, this might be a good option for the Cave pearls.

Comments are closed.