K1 project page INCLUDING publically-available files

Scuba_Vixen

I'm running it along side the KISS valve, from a seperate 1st stage (IP set at 135#). I like the manual control from the KISS setup alot, but with a camera and scooter for example, I can run out of hands sometimes, and having electronic control eases those situations.

My injection window comes every 6 to 8 seconds, and the injection duration timing is proportional to how far off setpoint the PO2 is. (if it's not off, it doesn't fire) The shortest time is 1/4sec and longest is 1.5sec. These timings are subject to programming change as "in water" experience warrants.



Darlene

jradomski

at your max injection rate, you can only inject about 4lpm.. I would raise the max inject time to 2.5 to 3.0 at a minimum.. Ideally at 135psi I would like to have the ability to have a 50% duty cycle available for ascents... That would yield a capability of around 10lpm +/- a bit...

__________________
Joe Radomski
CCR Trimix Instructor Trainer
ANDI Instructor Trainer Director #10

All posts are personal opinions and DO NOT reflect any affiliated agency unless specifically stated.

Scuba_Vixen

Thanks for the sugestion on the max firing time, Joe.

Being able to make an ascent without (or at least very minimal) manual addition (Raised PO2 for deco, not included) could be a very nice thing.



Darlene

Genesis

Cool.....

BTW I do not appear to have problems with porosity in the cases I had made.... one just got a nice soaking this afternoon as a "tag along" on some dives, and its bone dry inside...... nothing like being bashed around on a boat clipped to a set of doubles (with zero regard for what hits it, falls on it, etc) as a test paradigm

__________________
"A venturesome minority will always be eager to get off on their own, and no obstacles should be placed in their path; let them take risks for Godsake, let them get lost, sunburnt, stranded, drowned, eaten by bears, buried alive under avalanches - that is the right and privilege of any free American."
http://www.denninger.net
http://www.diversunion.org/liability.htm - Fix the Diving Cert racket

Genesis

Thought you guys might like a bit of an update....

I've resolved the problem I was having with the noise on the depth sensor channel - discovered that the instrument amp I was using really did not like any capacitance on the output at all, and would start oscillating! Aieee... that's not mentioned in the datasheet.... anyway, the run is direct enough to the ADC that removing the filtering on the output didn't have any other bad effects, so.... SNIP!

I've also redesigned the CPU board to use the nice single-chip driver that Darlene found, and re-did the head board in a 4-layer. That cut the size significantly - the new ones just showed up today.

Here's a pic of the CPU board before mod to change the transistor solenoid driver to the IC one (the new board, built up, is in the handset), along with both the R3 and R3.1 head boards. The new one is the unpopulated board on the lower right - I should be able to get it stuffed either this evening or tomorrow, and then connect it up and see how it works. I don't think I can make it any smaller than this without going to surface-mount....

The scrubber design is also coming along - with a bit of luck in the next few weeks I should have something that actually looks - and perhaps might even work - like a rebreather!

BTW these boards WILL work with the public-access code. Not all features will be supported unless you hack it in yourself (e.g. shorted or open solenoid detection on the monolithic driver chip) but nothing has been moved in terms of the pre-existing I/O signals. At worst some constants would have to be adjusted...

__________________
"A venturesome minority will always be eager to get off on their own, and no obstacles should be placed in their path; let them take risks for Godsake, let them get lost, sunburnt, stranded, drowned, eaten by bears, buried alive under avalanches - that is the right and privilege of any free American."
http://www.denninger.net
http://www.diversunion.org/liability.htm - Fix the Diving Cert racket

Genesis

Stuffed.... (except for one connector that I'm out of - barf - here we go to Mouser again!)

__________________
"A venturesome minority will always be eager to get off on their own, and no obstacles should be placed in their path; let them take risks for Godsake, let them get lost, sunburnt, stranded, drowned, eaten by bears, buried alive under avalanches - that is the right and privilege of any free American."
http://www.denninger.net
http://www.diversunion.org/liability.htm - Fix the Diving Cert racket

Scuba_Vixen

Looks nice Karl,
Can't get thru-hole much more densly packed than that.

Glad you liked that Hi-Side Driver chip,



Darlene

Genesis

It took quite a while to lay that out, playing with a number of different component layouts to get to where that ended up....

In the end I decided to hard-solder the interconnect, power and solenoid cables on this turn of the board. My reason for doing so is fire risk - anything that potentially could arc is big trouble inside the loop, even at very low power levels. Of course the solenoid draws considerable current and is the largest component of this risk, but not the only one.

Plus, hard-soldering those cables took a good chunk out of the size requirement, as even .100 KK Molex connectors are quite large in terms of board real estate. But - I want to keep plug-ins for the sensors, as those are obviously something that will need replacement from time to time. The one "missing component" spot (see if you can find it!) is for a current-limiting resistor for the depth sensor - not technically needed, but its insurance against a potential problem if the sensor gets shorted for some reason. Not sure exactly what has to go in there to prevent linearity problems as of yet... testing will be required on that one.

Once I've verified the operation of the board I'm going to pot it up to the connectors (hanging in a mold for epoxy) so the entire thing will be sealed other than the plugs.

The amusing part of all this is that the 4-layer boards were significantly more expensive than two layer (about $40 each for 4), but if I was to order up a bunch of them (e.g. 50) they'd get quite reasonable - for 50 the blanks are about $8 each, and that includes silkscreening (which I didn't do on these as they're prototypes). That means I could sell them at a reasonable price yet still significantly cheaper than someone could design up their own on a "onesie" basis.

Assuming the boards verify - both the head and controller - I'm seriously considering ordering up a batch of blanks and making them available, along with a bill of materials...... They should work fine with the GPL'd code (same hardware interconnects) but of course are really built for the "new generation" stuff I'm working with now...

__________________
"A venturesome minority will always be eager to get off on their own, and no obstacles should be placed in their path; let them take risks for Godsake, let them get lost, sunburnt, stranded, drowned, eaten by bears, buried alive under avalanches - that is the right and privilege of any free American."
http://www.denninger.net
http://www.diversunion.org/liability.htm - Fix the Diving Cert racket

Genesis

Looks like the boards - both head and controller - are good.

Here's a detail on the buttons for the front of the handsets. The only metal part in there is the spring (the magnet will be expoxied in the hole drilled in the shank and covered with epoxy so it cannot corrode) which should prevent any dissimilar metal issues that have arisen with some other designs.

Pardon the flashing still on 'em..... I haven't cleaned 'em up yet.....

In addition the entire button/boss/magnet assembly is replacable very quickly if necessary for some reason....

I love that little lathe I picked up.... it made turning these out of Delrin rod a piece of cake.




The proper springs and magnets should be here in a couple of days from McMaster at which point I can check the "throw" of the buttons and cut down their length if they're longer than necessary..... (which I suspect they are, but I left the "stems" long just in case.)

__________________
"A venturesome minority will always be eager to get off on their own, and no obstacles should be placed in their path; let them take risks for Godsake, let them get lost, sunburnt, stranded, drowned, eaten by bears, buried alive under avalanches - that is the right and privilege of any free American."
http://www.denninger.net
http://www.diversunion.org/liability.htm - Fix the Diving Cert racket

Genesis

Spindle for the center of the cannister.... 1" PVC turned down slightly to fit smartly into a 1-1/4" nylon hose barb (which will screw into the lid); two O-ring grooves cut 25 thousandths; nice, firm, even fit and a great bubbletight seal. The nylon hose barb is available at West Marine.



In use the can with the spindle sticking up (going to the bottom of the sorb) will have the head lowered over it engaging the spindle in the center inlet tube and then be latched down....

I'd love to use delrin or some other better material for the dip tube but have had a devil of a time finding something appropriate at McMaster..... so for now, PVC it is.

BTW I've figured out the T-pieces and lungs. The MSR bags when removed will allow the backing nut for a Poesideon "Hazmat" drysuit dump valve to pass (barely, but it does go.) Thus I can get the OPV and manual adds in there.

The cap, if chucked in a lathe and the inside hit with a cutoff tool down to 30 thou or so, gives a reasonable parting line for use with an exacto, followed up by cleanup with a Dremel.

The result is that a 1-1/2" SCH 80 piece of pipe will fit down the bore with a small amount of available room. Due to the slightly loose fit I am using Sch 80 so I can cut an O-ring groove deep enough without compromising the strength of the fitting materially for a seal ring which will then seal on the plastic outer ring. This O-ring also prevents the T-piece from being withdrawn when the cap is on. Directly on top of that is a much smaller O-ring which acts as a crush seal between the screw-on cap, the pipe and the mating piece.

In a pressurization test I exploded the MSR bag before the seal gave way... that ought to be good enough

To provide the water trap function I'm epoxying a piece of clear lexan into the T before inserting the pipe which goes to the bottom of the socket. The pipe piece (with the cut-in for the O-ring) is then fitted and a second piece of plastic is epoxied in place inside it, mating with the first. This forces any water travelling down the T-piece into the counterlung before it can continue onward. The exhale lung (the only one that should be able to get water in it, as the one-way valve on the mouthpiece should prevent anything from going back that way) will be the one with the drysuit OPV which will serve double-duty as a water dump for clearing partial floods.

__________________
"A venturesome minority will always be eager to get off on their own, and no obstacles should be placed in their path; let them take risks for Godsake, let them get lost, sunburnt, stranded, drowned, eaten by bears, buried alive under avalanches - that is the right and privilege of any free American."
http://www.denninger.net
http://www.diversunion.org/liability.htm - Fix the Diving Cert racket