Help with large (900liter / 243 gallon) aluminum diesel tank design

Get help and share Ideas
leftcoast32
Posts: 10
Joined: Mon Nov 08, 2021 4:20 am
2

Help with large (900liter / 243 gallon) aluminum diesel tank design

#1

Post by leftcoast32 »

Hi all,

I'm hoping for some help in repairing a leaking aluminum diesel tank. I'm on a 54 foot GRP sailboat on a circumnavigation and boats being boats, the tank started leaking in the middle of nowhere and needs to be repaired or replaced. The boat is currently in French Polynesia, so while I'm lucky enough to get access to a decent aluminum boat building shop, they are asking that I provide design/specs for the replacement tank. This project would be much easier if I was in Europe or North America, but I'm not.

The tank has poultice corrosion underneath it (see photo), where it was placed on carpet. All was fine for years until we had a small saltwater leak and the carpet got soaked. In it's current state, I cannot access the bottom to repair it and there are 6 baffles internally, so pouring a sealant inside is not a feasible solution.

The existing single tank on the boat is made of 5086-H111, 900liters, in the engine room and was installed prior to the deck being glassed in. See attached drawing of the EXISTING TANK. The tank is 200cm long, 42cm deep and around 110cm high. The low point is on the left side of the drawing, near the diesel shutoff valve, labelled CSA 05 019. The engine room opening is only 85cm, so to remove the leaking tank, I will need to cut the tank.

The tank is original and around 14 years old. The bottom is sloping as the diesel is gravity fed via a stopcock on the lowest point of the tank and being an ocean-crossing boat, the tank needs to fit perfectly without any movement.

OPTION 1 - Repair

Because I am not 100% confident that the local shop can build new tanks that fit perfectly and have the correct geometry to ensure lack of movement and proper diesel flow, I was considering cutting it out in 3 pieces, indicated by the two cuts on the attached drawing. I would have the local shop try to replace the failing bottom panel and then install new side panels where the cuts were made. We would then find a way to secure the 3 repaired tanks.

There appear to be some small pinhole leaks in the vertical welds, so I was considering roughing it up and applying Belzona 1161 to all existing welds.

My desire to repair it is really only to retain the existing tank, which I know fits. Of course, due to the 85cm engine room opening, the one tank will be replaced by 3 interconnected tanks. The repair cost will likely not be much cheaper than replacing them.

OPTION 2 - Replace

The second option would be to cut out the old tanks and have 3 completely new tanks fabricated. The local shop can only source 5083-H111, which I think is fine.

Given that the original tank is 14 years old, should I just bite the bullet and go through the pain of having them build 3 new tanks? My fear is that I fix the original tank and it leaks again in an even more remote location. Again, it won't cost much more, my bigger concern is getting it to fit perfectly.

Common elements and questions

Whether I choose repair or replacement, I will still need to connect the tanks and make a few changes and being that this is the first tank I've designed and well, it needs to work perfectly since we're going to even more remote places soon, I'd appreciate comments and suggestions.

1) For the inter-tank connections, they will weld male 1.5 inch threaded connections to each tank:
(i) Tank 1, on the right most and where diesel is initially poured into the tank system, will have an outlet on the bottom left side of its 1151mm high face, the low point.
(ii) Tank 2, in the center, will have an inlet on bottom right side of its 1151mm high face, the high point and an outlet on the bottom left side of its 1151mm high face, the low point of this tank.
(iii) Tank 3, on the left, will have an inlet on the bottom right side of its 1151mm high face, the high point. The outlet, which feeds the Racor filters, will be the existing one indicated by CSA 05 019.

These male threaded connections will have these 316 stainless shutoff valves:

https://www.kent-marine.com/en/product/ ... 1-1-2.html

with 90 degree outlet elbows, ending in a nipple and connected via fuel hose and hose clamps (see photo with items circled in red). On top of each respective tank, we'll have 16mm T nipples welded onto each tank and have fuel hose connecting all the tanks for venting purposes.

Underneath the tanks, I plan to use 1/4inch Starboard HDPE, bedded to the aluminum using polyurethane sealant to keep the tanks above any stagnant water.

The engine room walls around the tank are covered with sound deadening insulation and was resting up against the tank. I plan to leave a gap of at least 1/4 inch to prevent future poultice corrosion issues. As mentioned, the tank is in the engine room, which, while it has a suction fan, is quite warm due to the presence of the auxiliary Volvo motor and the Onan generator. Originally the boat builder covered the engine room side 1151mm x 2000mm face of the tank with insulating foam. Upon removal of that foam, I could clearly see some poultice corrosion underneath.

My plan is to NOT insulate that face of the tank, but should I be concerned about possible heat issues since the tank is exposed to the engine room? We can motor for days on end and I'm concerned perhaps the tank/diesel will get too hot?

Thanks in advance for your input and sorry for the long post!
Attachments
ORIGINAL TANK.jpg
ORIGINAL TANK.jpg (120.55 KiB) Viewed 9050 times
Existing Tank.jpeg
Existing Tank.jpeg (187.52 KiB) Viewed 9050 times
20210707_113115.jpg
20210707_113115.jpg (233.29 KiB) Viewed 9050 times
catalogue-marine-2020-page-58 copy[1].jpg
catalogue-marine-2020-page-58 copy[1].jpg (87.63 KiB) Viewed 9050 times
kmorin
Donator 08, 09, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24
Posts: 1735
Joined: Mon Aug 18, 2008 1:37 am
15
Location: Kenai, Alaska

Re: Help with large (900liter / 243 gallon) aluminum diesel tank design

#2

Post by kmorin »

leftcoast32,

my first two comments relate to the alloy and size of the replaced tank/repaired tank.

First, 5083, like 5086 do have strain hardening properties that require a bending nose bar, on a press brake, to be 4T- the radius should be four times the thickness of the material or you risk strain hardened zone that will be A) potentially cracked just sitting still! & B) more highly susceptible to corrosion than the surrounding material. (There are online tables to confirm this bending tooling requirement.)

So, unless the shop you're considering to do your work will be a) cutting all edges to provided outside corner welds- NOT BENDINg?? or b) using the correct bending dies? I'd be more inclined to ask for 5052 H32 due to its higher malleability in bending w/o the same (identical) hardening potential problems. But if 5083 is all you can get? then making sure any bending is done correctly (right dia. radius nose bar or multi-shot bends?) will be critical to the new tank's life.

Next is fitting this tank, any tank to a space. Even the most rudimentary ship's carpenter- should be able to put any smaller tank into any void by simply shimming the sides/ends/top and bottom to support the tank's surfaces to the interior surfaces of the structural void where the tank will install. So making the tank fit to the last fraction of an inch is not critically important- what is important is the retainers/bedding/framing/and chocks that are used to get it to be solidly part of the ship's structure and not move- & still drain all surfaces- and hopefully have some bilge air flow around the surfaces of the tank.

More when I think more about your post and reread a few times.

Cheers,
Kevin Morin
Kenai, AK
kmorin
kmorin
Donator 08, 09, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24
Posts: 1735
Joined: Mon Aug 18, 2008 1:37 am
15
Location: Kenai, Alaska

Re: Help with large (900liter / 243 gallon) aluminum diesel tank design

#3

Post by kmorin »

leftcoast32,
next comment is about bedding and surfaces; vertical, top and bottom.

Any aluminum sheet surface can corrode IF there is a means to create and retain a thin film of water- vertical/bottom/top...... the key is the thin gap between any other surface- that allows capillary action to retain moisture; that can then shift pf. The worst is a poultice type of circumstance, like carpet, but two pc of aluminum tightly spaced/touching will do the same thing if they can't dry out.

Rather than solid nylon/sheet plastic bedding I'd want to run the surface on the table saw to create ridges/valleys/troughs where water can escape and not form sealed flat puddles against the tank's walls- anywhere. I often recommend vinyl carpet runner- as it comes ridged and isn't reactive to aluminum.. while Starboard isn't reactive, it would be vastly improved if the surface stopped capillary action by having grooves cut in the tank side support surface.

Re-reading several times (I'm old and slow) I have come to the conclusion that you can move the tank out of its niche- and cut both (all) sides but you can't move it out of the engine compartment -w/o cutting it up? So you can get access to the tank on all four surfaces to cut it into thirds?

If that is accurate, and IF the inside of the tank is not corroded because it has no "water bottoms" sump (means to clean out bottom from water in the fuel) then you could plan to reuse the three sections by cutting the length of the middle section down/short enough to allow the four new ends to be added and still have some air space- between the vertical tanks ends. Say you're adding four 3/16" plates or just shy of an inch of plate to the cut ends? Then if the middle section were cut down by 1-1/4" to 1-1/2" in length; you should still be able to put the tanks back where they were.

By adding threaded or bolt-on tabs top and bottom to joint the three tanks together, athwartships, you could be sure the three wouldn't shift side to side as the boat rolled. Not sure what is above and below in terms of structure? but another set of bolt or thru-bolt tabs could help anchor the tanks into the hull.

As long as the tanks would be removed and opened up end to end- you can inspect for internal corrosion resulting from water bottoms that may have been left in the original tank? Also you might be able to etch the tanks inside and out? and add a sump to the low point with a small drain where you could regularly check/drain the very lowest point to see if you're collecting water?

more to follow.

Cheers,
Kevin Morin
Kenai, AK
kmorin
kmorin
Donator 08, 09, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24
Posts: 1735
Joined: Mon Aug 18, 2008 1:37 am
15
Location: Kenai, Alaska

Re: Help with large (900liter / 243 gallon) aluminum diesel tank design

#4

Post by kmorin »

leftcoast32
as regards heat in the engine room tank space:

First if the tank is supply fuel for a diesel engine (?) doesn't the return stream of fuel used to cool the injector piping return to the tank warm or hot over time? So, if you're concerned w heat build up- doesn't seem like engine room heat would be primary source? I'd be more concerned with return fuel heat dissipation? rigging a water to fuel (liquid to liquid) or an air to fuel (gas to liquid) Heat Exchanger shouldn't be all that difficult?

Either tube and shell type or even radiator type of exchanger would cool the tank's contents below any critical temp where the fuel might 'boil off' and become an issue?

When you take out the old tank, and before you decide on new or refitting what's there (?) you'll be able to see pretty well into the existing tank. If there's interior corrosion (which wouldn't surprise me at all) then judging the extent of internal corrosion for repair- compared to new metal should be a much more informed decision?

Joining the tanks with 'commoning' hoses and valves at their bottom surfaces would allow the tanks to level side to side when drawing down - only consideration is the fill rate would have to be less than the common leveling connections. If, for example you have reduced port ball valves instead of full port ball valves between the tanks- you'd need to learn to adjust the fill rate so you don't overfill the tank connected to deck fill plate before the other two could equalize levels.

Cheers,
Kevin Morin
Kenai, AK
kmorin
User avatar
welder
Site Admin
Posts: 4667
Joined: Thu Jan 03, 2008 11:51 pm
16
Location: Whitesboro, Texas
Contact:

Re: Help with large (900liter / 243 gallon) aluminum diesel tank design

#5

Post by welder »

Why can't you cut an access hole in the deck to pull the tank out and install the new one?
Lester,
PacificV2325, Honda BF225
2386
User avatar
welder
Site Admin
Posts: 4667
Joined: Thu Jan 03, 2008 11:51 pm
16
Location: Whitesboro, Texas
Contact:

Re: Help with large (900liter / 243 gallon) aluminum diesel tank design

#6

Post by welder »

And could you put a fuel bladder on deck to use until you get to a better place for repairs?
Lester,
PacificV2325, Honda BF225
2386
leftcoast32
Posts: 10
Joined: Mon Nov 08, 2021 4:20 am
2

Re: Help with large (900liter / 243 gallon) aluminum diesel tank design

#7

Post by leftcoast32 »

First off, apologies for the late reply. My internet isn't always the best and somehow the reply notification emails went to spam.
kmorin wrote: Mon Nov 08, 2021 3:07 pm
First, 5083, like 5086 do have strain hardening properties that require a bending nose bar, on a press brake, to be 4T- the radius should be four times the thickness of the material or you risk strain hardened zone that will be A) potentially cracked just sitting still! & B) more highly susceptible to corrosion than the surrounding material. (There are online tables to confirm this bending tooling requirement.)

So, unless the shop you're considering to do your work will be a) cutting all edges to provided outside corner welds- NOT BENDINg?? or b) using the correct bending dies? I'd be more inclined to ask for 5052 H32 due to its higher malleability in bending w/o the same (identical) hardening potential problems. But if 5083 is all you can get? then making sure any bending is done correctly (right dia. radius nose bar or multi-shot bends?) will be critical to the new tank's life.
Thanks, this is very helpful. They did mention to me that they wanted to work with 5mm or thinner material, which indicates to me that they have the bending dies for that thickness aluminum.

Unfortunately, being in the middle of the South Pacific compounded by the supply chain crunch, getting other grades of aluminum will be challenging. Being that they've been in business for a few decades and built a few large 100+ passenger ships and stated that they prefer using 5083, I hope they know how to make those bends correctly. That said, I'll trust, but verify.
leftcoast32
Posts: 10
Joined: Mon Nov 08, 2021 4:20 am
2

Re: Help with large (900liter / 243 gallon) aluminum diesel tank design

#8

Post by leftcoast32 »

kmorin wrote: Mon Nov 08, 2021 5:15 pm
Any aluminum sheet surface can corrode IF there is a means to create and retain a thin film of water- vertical/bottom/top...... the key is the thin gap between any other surface- that allows capillary action to retain moisture; that can then shift pf. The worst is a poultice type of circumstance, like carpet, but two pc of aluminum tightly spaced/touching will do the same thing if they can't dry out.

Rather than solid nylon/sheet plastic bedding I'd want to run the surface on the table saw to create ridges/valleys/troughs where water can escape and not form sealed flat puddles against the tank's walls- anywhere. I often recommend vinyl carpet runner- as it comes ridged and isn't reactive to aluminum.. while Starboard isn't reactive, it would be vastly improved if the surface stopped capillary action by having grooves cut in the tank side support surface.
Wouldn't polyurethane sealant well bedded tween the tank and the Starboard prevent any capillary action? My intention is to bed the starboard in the manner described in this article:

https://www.passagemaker.com/technical/ ... ank-design

Quote from article: The insulating material can be made of high-density polyethylene, closed-cell foam, or high-density synthetic rubber. Whichever material is chosen, it's important to ensure that it cannot absorb and hold water.

The surface of the tank where the material is to be adhered should be degreased and then lightly abraded using a 3M Scotch-Brite pad or 220-grit sandpaper. Degrease the surface again after abrading and then apply a polyurethane bedding compound. Finally, bed the insulating material (it, too, should be degreased, abraded, and degreased again) in the sealant.

The bottom of the tank should be treated similarly. Many builders install tanks on 1- or 2-inch-wide, 1/4- inch-thick (at a minimum) strips of one of the abovementioned insulating materials, spaced 1 or 2 inches apart. The strips should be placed athwartships, port and starboard, to allow water to drain when the vessel rolls. The primary objective of this non-hydroscopic exercise, an ABYC requirement for all metallic tanks, is to keep the tank from remaining in contact with water, particularly stagnant water.

Re-reading several times (I'm old and slow) I have come to the conclusion that you can move the tank out of its niche- and cut both (all) sides but you can't move it out of the engine compartment -w/o cutting it up? So you can get access to the tank on all four surfaces to cut it into thirds?
Access is an absolute nightmare. The only way I can access the "back" and bottom sides (relative to the photo of the tank I included with the first post) is by cutting an opening on the side of the tank that you can see in the photo (which we will call the front). The back the tank is bolted through the engine room wall by tabs indicated on the drawing as "CSA 05 007". Once I unbolt that, I can slide the tank forward about 10 inches so I can reach into the opening I have created and cut the back of the tank without damaging the engine room wall.

My intent is to use a sawzall for most of it and a multitool and a lot of blades and patience for the surfaces that are difficult to access. Honestly, the challenge of making clean cuts in hopes of re-using the tank may be insurmountable. If that's the case, I may just decide to hack the existing tank to bits, just to get it out.


By adding threaded or bolt-on tabs top and bottom to joint the three tanks together, athwartships, you could be sure the three wouldn't shift side to side as the boat rolled. Not sure what is above and below in terms of structure? but another set of bolt or thru-bolt tabs could help anchor the tanks into the hull.
Could you elaborate on how to make connections between the tanks? Perhaps you have some photos of the tabs you suggest? I presume these tabs would be made of thick aluminum and connected together with stainless bolts and a healthy dose of Tef-gel?
As long as the tanks would be removed and opened up end to end- you can inspect for internal corrosion resulting from water bottoms that may have been left in the original tank? Also you might be able to etch the tanks inside and out? and add a sump to the low point with a small drain where you could regularly check/drain the very lowest point to see if you're collecting water?
Luckily, the inside of the tank is nearly pristine. I have been very diligent about filtering every single drop that goes in and adding BioBor JF biocide. It's quite ironic that I have almost zero bug or corrosion inside but the damage came from the outside!
leftcoast32
Posts: 10
Joined: Mon Nov 08, 2021 4:20 am
2

Re: Help with large (900liter / 243 gallon) aluminum diesel tank design

#9

Post by leftcoast32 »

kmorin wrote: Tue Nov 09, 2021 1:56 pm
First if the tank is supply fuel for a diesel engine (?) doesn't the return stream of fuel used to cool the injector piping return to the tank warm or hot over time? So, if you're concerned w heat build up- doesn't seem like engine room heat would be primary source? I'd be more concerned with return fuel heat dissipation? rigging a water to fuel (liquid to liquid) or an air to fuel (gas to liquid) Heat Exchanger shouldn't be all that difficult?

Either tube and shell type or even radiator type of exchanger would cool the tank's contents below any critical temp where the fuel might 'boil off' and become an issue?
The original design does indeed have a return from the injectors and that heat source is not a problem over the thousands of boats this boatyard has built. I am just not sure how much heat is added by the injectors themselves given that the tank is 900 liters.

My bigger concern is, whether or not the heat from the combined engine and generator operating in the engine room on an uninsulated tank will cause the fuel to boil off. Wikipedia tells me that diesel vaporizes at 300degF, so I am almost certain the engine room won't get that hot, so maybe I'm worrying about nothing. It was just that the boatbuilder insulated the tank from the engine, so I was curious if there was a reason for that. As mentioned earlier, I don't want to attach insulation to the vertical tank walls again because there was evidence of poultice corrosion under that insulation when I removed it, indicating the insulation was holding stagnant water/moisture against the aluminum.

I presume you've heard of or seen engine room tanks that are uninsulated?

Joining the tanks with 'commoning' hoses and valves at their bottom surfaces would allow the tanks to level side to side when drawing down - only consideration is the fill rate would have to be less than the common leveling connections. If, for example you have reduced port ball valves instead of full port ball valves between the tanks- you'd need to learn to adjust the fill rate so you don't overfill the tank connected to deck fill plate before the other two could equalize levels.
I posted a link to the ball valves I have purchased. Hopefully those aren't reduced port ball valves? The commoning hoses will be 1.5inches and the three "new" tanks will be 250-350 liters EACH. I imagine that'll be enough but you're right, I just need to be careful when fueling up. The fuel fill hose is 2 inches, FYI.
leftcoast32
Posts: 10
Joined: Mon Nov 08, 2021 4:20 am
2

Re: Help with large (900liter / 243 gallon) aluminum diesel tank design

#10

Post by leftcoast32 »

welder wrote: Tue Nov 09, 2021 4:18 pm Why can't you cut an access hole in the deck to pull the tank out and install the new one?
As this is a center cockpit sailboat and the engine room and tank are below the cockpit, the surface above the tank is actually a cockpit seat and also a shelf inside the boat. Rebuilding that would be a nightmare!

The engine room wall "behind" the tank is a common wall with a passageway inside the boat. As much of a pain as it is, the only way to get the tank out is to cut it to pieces and remove from the engine room lid.
amel-54-bronwyn-cockpit-700x461.jpg
amel-54-bronwyn-cockpit-700x461.jpg (114.91 KiB) Viewed 8759 times
leftcoast32
Posts: 10
Joined: Mon Nov 08, 2021 4:20 am
2

Re: Help with large (900liter / 243 gallon) aluminum diesel tank design

#11

Post by leftcoast32 »

welder wrote: Tue Nov 09, 2021 4:20 pm And could you put a fuel bladder on deck to use until you get to a better place for repairs?
Unfortunately the closest "better place" for repairs is New Zealand and that's over two thousand nautical miles away. I don't feel comfortable living with a fuel bladder traveling to the other, even more remote islands we will be visiting on the way there.

Due to covid and cyclone seasons, we don't anticipate getting there until Nov 2022 and we anticipate the passage to there to be one of the roughest on our entire circumnavigation. All things that point to us trying to get it fixed now.

I wish I could wait - certainly would be a lot cheaper!
kmorin
Donator 08, 09, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24
Posts: 1735
Joined: Mon Aug 18, 2008 1:37 am
15
Location: Kenai, Alaska

Re: Help with large (900liter / 243 gallon) aluminum diesel tank design

#12

Post by kmorin »

reacting to the reply @ #7 post.

I'd go to the shop and ask to discuss bend radius for 0.2" or 5mm or 0.187" (3/16") mtl and confirm the bend radius of the top die? Yes, Trust but Verify. If you get the :normal- "we've been doing this for 20 years" and "using a knife edged steel die is fine"... I'm not sure you're being served for the global long haul?

that is: if you are forced to build w/o using old tank parts.

More to follow when I can read fully your replies.

Cheers,
Kevin Morin
Kenai, AK, USA
kmorin
kmorin
Donator 08, 09, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24
Posts: 1735
Joined: Mon Aug 18, 2008 1:37 am
15
Location: Kenai, Alaska

Re: Help with large (900liter / 243 gallon) aluminum diesel tank design

#13

Post by kmorin »

Leftcoast32,

Image

here is a tie bar of 1/2" 6061 T6 Extrusion (flat bar) used to bolt two side by side tanks together.

Image

the tapped 1/2" block is not fully welded to the tank side in this image. But I think you see the idea?

as to stress in bending from too sharp die? These strain lines don't always show up right away? Might look up an older post by Jetty Wolf, he replaced a below decks tank in his welded skiff, its possible his tank was strain hardened and corroded out due to incorrect tooling? Should be able to find his post here somewhere?

Image

Hope this helps clarify my remarks a bit?

More to follow,
Cheers,
Kevin Morin
Kenai, AK
kmorin
kmorin
Donator 08, 09, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24
Posts: 1735
Joined: Mon Aug 18, 2008 1:37 am
15
Location: Kenai, Alaska

Re: Help with large (900liter / 243 gallon) aluminum diesel tank design

#14

Post by kmorin »

lc32
regarding the mounting of sheet plastic on the tank surfaces; Solid sheets mounted directly to the metal will have a different expansion and contraction rate - so, if the sealant remains very elastic when cured- I suppose sheet plastic could be glue to the tank sides- I'd be concerned about the eventual intrusion of water and resulting corrosion?

Unmentioned in the tank article you linked is etching the mill scale off of the aluminum. This is an important step for tank longevity.

I have a thread on another site that discusses tanks- http://www.glen-l.com/phpBB2/viewtopic.php?f=7&t=28367 maybe worth reading? Not sure you don't already know what I'd collected there? But a review could help answer some questions- from my point of view.

still reading the posts,

cheers,
Kevin Morin
Kenai, AK
kmorin
kmorin
Donator 08, 09, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24
Posts: 1735
Joined: Mon Aug 18, 2008 1:37 am
15
Location: Kenai, Alaska

Re: Help with large (900liter / 243 gallon) aluminum diesel tank design

#15

Post by kmorin »

lc32

this remark/answer is about fuel tank heating. The fuel returning to a full tank is one whole lot different heat contribution than one returning with the same heated fuel is in a 1/8 full tank! Reaching 280-300 def F of tank contents seems pretty hard to believe happening via radiated engine sidewall heat!

The engine room heat would (need to) be radiated into the fuel, via the engine side wall radiation to the tank side wall then to the fuel. Or the engine heats your engine room to 300+ degrees and therefore the ambient engine room temp warms the tank ? Not happening by radiation of room convection! The engine room is vented to get fresh air to run.... so even in the tropics where its warm.... I don't see the engine space reaching 300F??

However, the return fuel has been heated by conduction- the liquid is in hot metal tubes- this is returned to the fuel tank to reduce injector pump cavitation and is a much greater source of BTU's in the tank's fuel than side wall radiation and convection in the engine room. I think you might have the BTU sources confused in magnitude?

Besides, if you put a pc of 0.080" sheet 1/2" off the side of the tank exposed to radiant heat sources, the heat will #1 largely reflect, #2 HEAT will cause the hot air (expanded) to rise up the chimney formed between the 'hot plate/sheet' and the tank wall, and #3 very little of the radiant heat can 're-radiate' to the tank wall!

So if the voids are vented- even in tropical climates- the voids will form a cooling flow just by heat rising and the chimney affect. So if all voids have a forward facing funnel and an after facing funnel- the void will constantly have a heat differential induced cooling flow even in hot weather. If the boat is becalmed this will diminish to very little but the rising heat will cool the engine and tank voids.

I don't see the need to put sheet plastic ON the tank wall, it seems a path to create eventual corrosion and not real gain in heat shielding. If you want to cool that tank put a water to fuel (tube and shell) passive heat exchanger on the return fuel line to use seawater to cool the engine heated return fuel.

Hope this came across in text? My thoughts on the tank cooling issues.

Cheers,
Kevin Morin
Kenai, AK
kmorin
leftcoast32
Posts: 10
Joined: Mon Nov 08, 2021 4:20 am
2

Re: Help with large (900liter / 243 gallon) aluminum diesel tank design

#16

Post by leftcoast32 »

Thanks Kevin,

The pics help a lot.

In this situation, would you be concerned about water getting trapped between the bar and the tank? They are also considering putting a bar across the tops of the tanks, which I presume would it even more likely that water gets trapped between the bar and the tank?

Being that I won't likely (hopefully!) be separating the tanks too often, could I just apply a polyurethane sealant before attaching the bars to the tanks?
kmorin wrote: Tue Nov 30, 2021 3:14 pm Leftcoast32,

Image

here is a tie bar of 1/2" 6061 T6 Extrusion (flat bar) used to bolt two side by side tanks together.

Image

the tapped 1/2" block is not fully welded to the tank side in this image. But I think you see the idea?

as to stress in bending from too sharp die? These strain lines don't always show up right away? Might look up an older post by Jetty Wolf, he replaced a below decks tank in his welded skiff, its possible his tank was strain hardened and corroded out due to incorrect tooling? Should be able to find his post here somewhere?

Image

Hope this helps clarify my remarks a bit?

More to follow,
Cheers,
Kevin Morin
Kenai, AK
leftcoast32
Posts: 10
Joined: Mon Nov 08, 2021 4:20 am
2

Re: Help with large (900liter / 243 gallon) aluminum diesel tank design

#17

Post by leftcoast32 »

The tank builder suggests that we coat the entire tank with epoxy to prevent future issues.

I have read differing opinions with some saying that any coating may start to fail, allow water in and invite corrosion. While others say it's a great way to protect the tank.

What are your thoughts?
kmorin
Donator 08, 09, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24
Posts: 1735
Joined: Mon Aug 18, 2008 1:37 am
15
Location: Kenai, Alaska

Re: Help with large (900liter / 243 gallon) aluminum diesel tank design

#18

Post by kmorin »

LC32,
Coatings are only as reliable on aluminum as the prep- if the tank is well etched after mechanical abrasion- sand blasting being the best sander next, and an alloydne (chromic acid resulting in a chrome oxide film over the bare/raw aluminum) solution put on while the etched metal is still wet from the water rinse? Then primer and topcoats put on correctly with all the joints/crevices/thread hole's edges protected from 'pulling up' while the paint dries- then epoxy coating will help fighting bilge corrosion.

But if the tank isn't coated with sound procedure then those who say it will speed up crevice (poultice) corrosion are correct. Kind of an all or nothing sort of circumstance.

As to tie bars locations? I put them on the vertical surfaces but painted the tanks (described above) including the face of the tapped blocks- which were on all four vertical sides of these tanks. The after side you see in the posted image was copied on the forward side and the insides have much larger blocks and a 'straddle' mount to fit over the glassed in wooden keel (4x12) as shown. The outboard sides also had 'engine mount' style fittings and tapped blocks to hold them. All vertical surfaces were painted on both the tie bars, straddle mount fittings and those engine mount style outboard. these tanks were suspended between the VKBeam and the outer two hull stringers already built into the hull.

Cheers,
Kevin Morin
Kenai, AK
kmorin
leftcoast32
Posts: 10
Joined: Mon Nov 08, 2021 4:20 am
2

Re: Help with large (900liter / 243 gallon) aluminum diesel tank design

#19

Post by leftcoast32 »

Kevin,

It turns out the HDPE Starboard is not ideal as the Sikaflex 291 and 292 that I have will not adhere to it. Apparently nothing sticks to HDPE.

Based on this page, I should adhere the non-hydroscopic insulating material to the tank, at the location where it rests on the supports. That way, no water can get between the insulating material and the aluminum.
https://www.passagemaker.com/technical/ ... ank-design

From your description below, it doesn't seem you would attach the insulating material to the tank's bottom surface?

I want to absolutely make sure I do this right because the the current damage is related to the aluminum resting on carpet, which created the perfect conditions for poultice corrosion.

I have access to a building materials store here. What material would you suggest I purchase as insulating material? Would you attach that material to the tank bottom where it rests on the supports?

Thanks!
kmorin wrote: Mon Nov 08, 2021 5:15 pm

Rather than solid nylon/sheet plastic bedding I'd want to run the surface on the table saw to create ridges/valleys/troughs where water can escape and not form sealed flat puddles against the tank's walls- anywhere. I often recommend vinyl carpet runner- as it comes ridged and isn't reactive to aluminum.. while Starboard isn't reactive, it would be vastly improved if the surface stopped capillary action by having grooves cut in the tank side support surface.


Cheers,
Kevin Morin
Kenai, AK
leftcoast32
Posts: 10
Joined: Mon Nov 08, 2021 4:20 am
2

Re: Help with large (900liter / 243 gallon) aluminum diesel tank design

#20

Post by leftcoast32 »

FYI here's where the tank will be mounted. This is a top-down view. The drawing is in French because I'm on a French island. :)

The original tank, being a bit over 6 feet long, only had two fiberglassed over wooden wedges as support underneath. Because we're now installing 3 tanks, I had 4 more fiberglassed wooden wedges added.

I've also attached a photo of how the wedges look from the side. The bottom of the tanks are sloped into the boat and the wooden wedges match this sloping angle.

Part of me wonders whether the epoxy coating on the tank bottom will be sufficient to prevent poultice corrosion from occurring where the tank rests on the supports. But I fear without an insulating material, over tens of thousands of miles of seaway, the epoxy will fail and invite water between the tank and the failed coating.

So what would you suggest I use between the tank and the supports? The tank shop suggests we add a sacrificial aluminum sheet or bar, but I am skeptical. My idea of HDPE won't work, as mentioned above.
drawing copy.jpg
drawing copy.jpg (200.4 KiB) Viewed 8398 times
Picture1.jpg
Picture1.jpg (187.41 KiB) Viewed 8398 times
kmorin
Donator 08, 09, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24
Posts: 1735
Joined: Mon Aug 18, 2008 1:37 am
15
Location: Kenai, Alaska

Re: Help with large (900liter / 243 gallon) aluminum diesel tank design

#21

Post by kmorin »

lc32,
as mentioned above, I would look for some insulating material- like sheet plastic but... not a full sheet film. If you took 1/8" HDPE and put it on the table saw and put a grid recess pattern (checkered) about 1/6" deep cuts- then the capillary action that creates a wide area thin film will not form as strongly so the grooves won't allow water to cover the entire tank bottom area where the plastic is against the tank's bottom plate.

By creating 'leak paths' (grooves) and checkering the sheet- you break up the capillary action paths that allow water to stay against the metal- yes painted bottoms with plastic film (preferably not flat sheet) should work to bed the tanks.

If clear vinyl 'carpet runner' is available- this in an inert plastic that has ridges running in one direction on the surface- and would provide air circulation (not high flow) as well as drain paths for moisture that accumulates. I've used this product and recommended it before. Main idea is to reduce large areas where the water can stand.

If painted the tanks would have to move quite a bit to scratch the paint off- if it is well applied as I've mentioned above?

If a sacrificial plate or bar was put on the bottom of the tank- then the interface between the tank wall and the "thickener" would need to be sealed - and air tested!! otherwise when it the adjacent metal heats and cools the face between the two pc of metal will make a perfect capillary/suction of droplets of water running off the tank sides! Then you'll have created a crevice cell ! If you use thickener plates then make sure they're welded all around?

Air testing it to make sure the interstitial space between the two pc can't pull a vacuum through a cold lap or crater crack and get moisture into the space.

cheers,
Kevin Morin
Kenai, AK
kmorin
Post Reply Previous topicNext topic