Hybrid Ballast Tank/Piston System

[Monahan Steam Models]

New vent valve for the main ballast tank.




Simple sprung vent slide valve. Normally I'd like to see the sealing surface be on the pressure side which would increase sealing under pressure.



Hard to see from the photo but there is a 3/16" OD Viton ball inserted into the retracted brass vent seal cup above.

Hope to test this new vent valve in the next couple days.

Nick

[Monahan Steam Models]

I'd expect a valve size of 3mm perfectly adequate to vent the tank. try it once you have the cylinder inside a hull, that may make a difference.

Andy,

You are absolutely correct that the 3mm vent at the top of the tank is perfectly adequate to vent the tank. In fact it takes 7 seconds to completely flood the main tank by uncorking the 3mm vent fitting directly at the top of the ballast tank.

The poor venting flow is due to three issues. The biggest issue of the three being the remote mounted vent valve substantially restricting the venting air flow. The flow gets worse as the tank reaches full. Takes approximately 40 seconds to completely flood the tank with the restrictive valve. Way too long.

Second issue is where the remote valve is mounted. As you’ve mentioned earlier, it is best to vent at the top of the tank. (This is the first time I’ve experimented with putting the vent valve in a remote location). Any water covering the vent valve outlet has negative effects on venting. Even with adding the vertical tube to the valve to raise the outlet above water level that I tried and found to work, there is still the risk of that tube getting enough water into it after multiple dives to have negative venting effects.

The last issue was the length of 3mm ID tubing that connected the vent fitting at the top of the tank with the remotely mounted vent valve. Flow tests showed that just adding a 12” length of tubing made the flood time for from 7 seconds to 18 seconds. Using a 12” length of tube with an ID of 3/16” had no restrictive effect on flow.

Currently, I’m making a new vent valve assembly that will be mounted on the top of the tank and have a pushrod connecting it to the ballast servo pushrod exiting the forward end cap. Was originally wanting to avoid this as it adds slightly more complexity to removing the forward end cap.

[Subculture]

I'd expect a valve size of 3mm perfectly adequate to vent the tank. try it once you have the cylinder inside a hull, that may make a difference.

[Monahan Steam Models]

This is impressive. Well done! Now put a hull around it!

Thank you! Believe me, I am dying to move on to putting it in the hull !!! I can’t begin to express how over it I am spending all this time on the WTC part of the project but it needs to be tested, finished and rock solid before I can confidently move on the hull build.

[JHapprich]

This is impressive. Well done! Now put a hull around it!

[Bob Gato]

I knew it couldn't have been the Monahan engineering!

[Monahan Steam Models]

Is the Mav-2 giving you 1/8 venting internally? many 1/2" water valves don't really open internally the .19635 sq" that gives you 1/2" worth of flow.

Consider a flow check-a timed leak down of a set volume thru an unrestricted 1/8" tube vs the same volume thru the mav-2 depressed

Bob,

At low pressure the valve flow volume is terrible. Good call on checking the flow rates. Will need to modify the design to accept a higher flowing low pressure vent.

Nick

[Monahan Steam Models]

That thing just looks so stable. It's gonna work great! But what challenge will you have trying to maintain PD? lol

Thanks Steve,

I was surprised myself to see how stable it was too. Having the underwater camera gives a new perspective that I did not have before. Pretty cool.

As far as maintaining PD goes, that probably wouldn’t have been much of a challenge anyway with the onboard Depth Cruiser on the bow planes and the AD2 pitch controller on the aft planes with out all the complicated junk I packed into this thing! Lol !

The project was a personal learning experiment for me. As interesting as I have found it along the way, I doubt that I will build another system like it again. Too complex with too many things that could potentially go wrong. That being said, the only real advantage that I see it having is the statically adjustable trim which might come in handy for the different types of gas and electric torpedo loads and that is really the reason why I decided to go down this path. Even then, I won’t know how much those will effect the trims until I finish the type VIIC project.

I’m already looking forward to a future build of a Bronco Type XXIII that will use a 2.5” MSD which is set up with a snort system. That’s a rock solid system in my opinion. Will use a DC as well as a AD2 on that project and that will be as complex as it will get. Should be a fun boat to skipper.

Nick

[SubDude]

That thing just looks so stable. It's gonna work great! But what challenge will you have trying to maintain PD? lol

[Monahan Steam Models]

Good night.

[Monahan Steam Models]

Quick update.

I spent a couple hours today working on the WTC project. Machined a simple part that allowed me to move the Tee fitting that was mounted on the back of the vent valve. The Tee fitting is now directly connected to the vent fitting at the top of the main tank. Same connections to the Tee fitting as before. Vent valve connection, emergency gas blow connection and low pressure blower connection.

A low pressure release one way check valve was added to the low pressure blower intake snorkel that only allows flow to occur when the pumps are drawing air in. The valve prevents any flow being pushed through the plumbing in the opposite direction.

Did some tests this evening and so far so good. I wanted to take some under water footage yesterday with the GoPro 9 I picked up recently for the Neptune sub but the digital micro servo for the ballast vent gave up the ghost yesterday from all the testing and mainly because I set the end point stop further then the vent valve could travel. Live and learn the hard way. It was replaced today and the end points set appropriately.

I did a number of quick dives and surfaces using the system in full manual mode to make sure everything was working as it should.

Decided to pull the GoPro out of its box and try it out finally. I had already charged it and downloaded the app a couple weeks ago when it showed up but I had not yet tried to actually take video with it.

I was excited because the WTC was working and hopefully the camera. So into the tub the camera went. I wasn’t even sure if I set it up correctly or if it was recording but it did record and here’s the very first video underwater from it and the WTC. Sorry it is not edited at all and the digital micro servos chattering under water are very loud so turn down your speakers a bit if you watch the video.

https://m.youtube.com/watch?v=HASNcURMxro&feature=youtu.be

In the video I vent the main ballast tank then make a small adjustment to the forward piston tank. Then a quick adjustment to the aft piston tank. The static dive trim is set for about periscope depth. At the end of the video the main ballast tank is blown.

Still need to test and document the automatic functions. Haven’t even tried those yet as these in the water tests have been only the past two days. Trying to crawl before I walk. Still getting to know this contraption.

Nick

[Monahan Steam Models]

QUOTE=Subculture;n150204]How about an extra pipe to increase the cross section of the pipe, or use a larger hose?[/QUOTE]

Andy,

I like your thinking and I can see that would definitely increase the flow rate of the escaping air from the ballast tank to the vent valve. I suspect part of the problem lies more in the connections I made at the Tee fitting at the vent valve.

Allow me to explain. If the outlets from the low pressure blower as well as the outlet from the emergency gas blow were routed Instead separately from the vent valve to directly to the top of the main ballast tank top fitting, then it would not matter if their plumbing became flooded as the water would be cleared or pushed out during blowing cycles.

If the connection between the venting and blowing plumbing was split at the top ballast tank fitting this would hypothetically reduce chances of water getting into the vent plumbing circuit due to the leak by of the Diaphragm pumps.

Still other matters would need to be addressed like guaranteeing water will not enter the vent valve outlet tube and create a tension blockage.

Nick

[Subculture]

How about an extra pipe to increase the cross section of the pipe, or use a larger hose?

[Monahan Steam Models]

Is the Mav-2 giving you 1/8 venting internally? many 1/2" water valves don't really open internally the .19635 sq" that gives you 1/2" worth of flow.

Consider a flow check-a timed leak down of a set volume thru an unrestricted 1/8" tube vs the same volume thru the mav-2 depressed

I agree a flow test should be done. I did notice the Tee fitting and the outlet barb have slightly smaller than 1/8" ID's.

[Monahan Steam Models]

Tom, Andy, I agree with both of you.

I did some more tests this morning to see what was actually occurring. Made a couple small changes then did some more tests and observed. That's when I saw what was happening. It was so dumb.... I'll try my best to explain below. Long story short though, problem found, solution to problem tested and the venting works now every time. Still need to decide on how to go about adding a permanent fix.



And now for the long winded explanation:



First here's the changes I made this morning.


Above is the first change that was made. After witnessing short slugs of water inside the clear tubing blocking the venting air from escaping I decided to try stepping up the tube's ID to increase flow by replacing a section with a 3/16' ID piece of hard aluminum tube.




Next I swapped where the tube from the vent connects to the barbed Tee fitting on the back of the vent valve to eliminate the possible U-bend sink drain trap effect I mentioned earlier. I re-attached a short vertical length of clear tube to the brass barb on the vent valve to allow the vented air to be above water level while surfaced. These changes did make slight improvements of the venting but they were not the problem, nor did they solve the intermittent and variable filling of the ballast tank after the first dive.


A quick explanation of the connections to the TEE fitting attached to the back of the vent valve. Center connection is to the vent fitting at the top of the ballast tank. Small tube connection at bottom of TEE is the input from the emergency gas blow valve outlet. Upper tube connection is the input from the low pressure blowers outlet.




Okay now for the dumb...

Here's what I observed happening and what I did to stop it from happening. After doing this the main ballast tank vented perfectly each time. I'll try my best to explain.

While observing the WTC submerged with a completely full ballast tank and with the vent valve closed, I noticed water slowly entering the clear tube attached at the top of the ballast tank and traveling through it to the TEE connection. From there it traveled into the low pressure blower line. During the design process I had anticipated water getting into these lines. I debated whether or not to add a check valve to the blower outlet tube but decided it was not not necessary for two reasons. One any water entering would be blown out during the blowing of the main tank. Secondly I assumed the diaphragm pumps would not allow any flow when they were not being used..... Wrong bonehead.

Back to the water getting into the tubing. Yes it does get blown out during the blowing process of the tank but not all of it. A small amount of residual moisture is left inside of the vent tubing mostly on the inner walls after blowing. However, the more water that floods into the tubing the more residual moisture that will be present after blowing. This is a part of the problem. The next time the tank is vented the residual moisture is pushed through the tubing, collecting into small slugs of water while traveling towards the vent valve.

When the tank is vented from the completely empty state, most of this residual water is pushed out through the vertical vent tube outlet attached to the valve but not all of it. As the airflow being vented from the main tank becomes a less and a weaker flow as the tank fills, the small amount of residual moisture still present creates enough of a blockage in the system that air trying to escape is not enough to push the last of the moisture from the vent valve.


Here's what I did to stop this from happening.

Starting fresh with a dry system, I pinched closed the snorkel valve tube leading to the low pressure blower intake. Then the main tank was flooded completely. Once the tank was flooded, no more water would enter the tubing at the top of the ballast tank vent fitting even if the vent valve was left open and no water would flow into the vertical vent tube outlet attached to the vent valve. The system was then blown as I stopped pinching the snorkel tubing. I repeated this multiple times and the system dove consistently each time with not moisture in the tubing.

Possible longer term fixes could involve adding a check valve on the snorkel intake. Or possibly adding a float valve to the snorkel intake. A sledge hammer approach would be to not mess with the snorkel tubing at all and add another small low pressure blower to the vent valve outlet tube to pull the air out of the ballast tank but that is more complicated and not necessary.

Anyway this is what I observed happening with the system this morning and what I tried that worked. I'm definitely open to hearing suggestions from others.

I agree the best practice is to have a simple free flowing vent valve at the top of the tank but at this stage that would require a substantial redesign and now that I know what was causing the problem and so far what I quickly and crudely did to make the system function as planned.

Nick