ZB-1, ZB-2: Scratch Build Cylinder design for limited Production by Zero Bubble.

[Davidh]

Hello all,

Thank you all for the feedback. I have really appreciated it. I have been making cylinders for almost 20 years. I have tolerated leaks and have taken measures to mitigate as I've made one offs, but when you are thinking about making these things for production, your mind expands and the implications become greater. A Simple thing like O rings and leaks become magnified.

I spent today dealing with 'O' rings. Last night I cut the grooves in the flange of the respective front and back end caps that I will use to prototype all the arrangements. Slipped the rings on and then placed the cylinder in my pool. So far so good, no leaks that I can see. The end caps are a tight fit and then 'O' rings press very firmly into the polycarb.



I then gave both end caps a coat of resin. In this heat, 30 degrees, dry in no time.

So this morning I checked the resin coat on the caps. Nice and dry, 10 am and already 32 degrees, high humidity. The plan today was once again to test the 'O' rings but this time I'm in the pool too.
After cooling off for 40 mins or so and holding the cylinder down half a meter I had a slight drip inside. Should do better than that.

I took the cylinder out and had a look at how the O rings sit in their grooves. The resin dried unevenly in the grooves creating high and low points. This would need to be sanded or turned out. I placed the caps on the lathe and turned down the high points. I then sanded some further to level down the grooves. Slip on the 'O's and fix the caps back on and back for a swim. After twenty minutes more of a leak.

Back to have a look at those rings.

Along with the question of how oversize that flange should be, to which I thank all the contributions, I have also been thinking about the groove to compression relationship with the 'O' rings. Should the groove be wide and shallow to help squash the o ring even further or narrow and deep. Currently the grooves are narrow and a little deep. The O ring just rises like a bump just above the radius of the flange. Pressing against the inside of the polycarb I noticed a couple of potential problems.



We've all done it,

I think there may be a couple of factors that lead to this. 1. Not enough groove for the O ring to sit in? 2. That this O ring is too sloppy. Get a tighter one? Opinions please. I have filed a 45 chamfer along the inner edge of the polycarb to help deflect the O ring down as it slides into the underside of the tube. This does assist.

Theoretically the water can only get in two places with O rings. Above and below. The O ring needs to seat well with both. This next photo shows what I think is the most important thing to get right with O rings.



As you all know, when an O ring presses against the side of the tube you can see a black line where the rubber is pressing against the inside, especially in the right light. This should tell me at least the outside of the O ring is creating a seal. I can't see beneath the O ring to see what is happening in the bottom of the channel, all you can do here is get it smooth and consistent.
By the afternoon I managed to get no leak. I believe this is because I got at least one O ring to press firmly against the inside of the Polycarb. A thin black line barely does the job. A thick black line consistent and all the way around should provide the line of defense we are all after.



I think that the thicker the line the better. I want both rings to press firmly and created a thick black line all the way around.




So my question is. Wide shallow channel or narrow deeper channel? or does it matter as long as the O ring is squashed and there's a thick black line inside the polycarb? Suggestions please?

Something for HWSNBN,

I look at all your photo's where you have 1--20 cylinders sitting on a bench, and I think to myself, wow, he has machined all those caps and they don't leak. I have one of you cylinders and It don't leak.
Do you have a percentage of caps that you test and have to fail as they leak?, or is your routine so good after all these years that you can pretty much make every end cap watertight?

I'm guessing that after making hundreds of these you have in place procedures or steps that guarantee no leaks. I would appreciate your wisdom on this.



David h.

[RCSubGuy]

David M. has perfected the art of cylinder production, and I've seen him in action.

There are two tolerances to pay attention to:

1.) The interference of the cap and the tube
2,) The depth of the o-ring groove.

If you machine the end caps to consistent tolerances of size of each piece of tubing (and as you already know, each cap needs to be individually machined to match each individual piece of tubing), you have the base dimension from which you can machine down your groove. David uses a self-made tool that has the correct width, and a stop that halts cutting at the correct depth.

In my experience building my own cylinders in the past. Larger o-rings are far more forgiving than smaller ones, and offer a larger sealing face on the inside of your tube.

[He Who Shall Not Be Named]

Good question, David. I'll take some shots of the tools and process and work it up into a proper response tonight. You're on your way. Soon, getting those flanges and grooves right the first time will be old hat.

David

[Rick Teskey]

If you use just one or ring it may cut your leaks in half try one or the other.
if fit is a little loose add a shim under o ring I use an automotive pin stripe , they come in various widths , cheap too try

[Davidh]

Hello Rick,

Yes I re-tested today, no leaks. One O ring on the twin O ring end cap.


David,

looking forward to your write up big time.

Thankyou.

David H

[Scott T]

Could you mount the o-ring in a bed of silicone to ensure no leaks underneath especially if you are using a shim?

Another question is could you make a third seal using a wrist band/elastic band over the gap between end cap and cylinder.

[Davidh]

Hi Scott,

Thanks for the suggestion. Because I am designing these units for limited production, I want as much as possible to incorporate proven technology that is repeatable and reliable. The "O" rings should be one of the easier things to seal as they don't rotate or reciprocate. The only downside is the large surface area. I really think this design needs a reliable sealing method that is simple and consistent. Hundreds of units using 'O' rings already in use proves to me that this system works well. I just have to get the design and practice right.

The key to this will be accurate templates and working out the right procedures. QC and all..



David H

[Davidh]

Hello all,

I have over the last couple of weeks designed and started making most of the parts that comprise the shaft supports. These are the round sections that support and seal the prop shafts as they exit the
rear of the end cap. I have focused on the stern end cap of the ZB-2. The twin shaft end cap. I suppose this one is the most challenging and will probably take the longest to perfect. I am also going through a bit of a twin shaft submarine phase right now. I've just completed the Delta IV and having too much fun with the Papa...

I have followed the work of HWSNBN and have designed my set up in a very similar way to how David does it. I have made the various parts from different materials not for any particular reason other than I like using the lathe and these materials were at hand. Brass, Renshape, and Delryn (Acetyl). Nice materials to turn.




I'll let Blender, (awesome program that nobody else seems to use) do the talking. The outer ring is currently turned in Renshape. The Shaft support housing is delryn and the Seal inner support is
brass. The shaft extension is stainless and the U cup seal is rubber. Why are most of the components red? Well most of my boats are Russian so... I also don't want my end caps to be confused with
Mr Fifty shades of Grey.....



Here's a question, yes David that means you.
I have seen that you glued your U cup seal into cylinder section. Then you butt up the Oilite bearing from underneath. I was planning, especially for the testing phase of all these parts to
simply silicon everything. This is because I will eventually take the whole assembly apart and re-think the whole process if I encounter leaks. Or everythinig works and then these parts will go the tooling stage. Can I get away with siliconing all these components or is superglue an essential part? I can just see the U-cup tearing as I try and pull it form the recess it sits in. Would Silicon be strong enough to resist the turning moment of the U-cup as it seals the rotating shaft?



The white piece of Delryn on the left is the shaft support housing. It silicons down over the end above the U-cup.



I am still mulling over the diameter business to do with Polycarb coming in different tolerances. I may have a further question or two for the masses..
I am looking forward David to the write up on procedure and end cap shaping... No pressure....



This project has brought the realization that so many seemingly small technical things that we give fleeting attention to can actually generate lots of questions. I have go a few questions later about getting the extension shaft onto the motor shaft. What seems like a straightforward process actually hides an involved process. Well for me anyway as my attempts to drill out the shaft for the motor were very successful at creating a lovely smooth centric.... but ,....slipping shaft.....

Don't be shy people,


David H.

[type7]

That is what engineering is all about. Not everyone enjoys it, hope it works out in the end.

[Davidh]

Hello all,

Thanks type 7. Yep, its the challenge we are here for. I've always enjoyed a model engineering challenge. I like the fact that If I build something myself I should be able to fix it and make it better.

Anyway, back to the cylinder. As mentioned previously the next challenge and thing to get enlightenment on is the Shaft extension. I will take a piece of 4mm stainless steel and cut to size. Now initially I put the required size drill bit into the tail stock of my lathe and moved it in. Pretty soon it was smoking and not much else. Should have known. Researched drilling Stainless steel and Cobalt tipped bits are what you need. So I went up to the local Bunnings (Oz hardware chain) and found a nice little 2mm bit.

So pushing that in on the tailstock did the job. Some machine oil and nice little curly bits coming off. Then took the part and placed it on the motor shaft. Slightly too small. I decided to install a slightly larger 2.1mm drill bit. Would this cut? Not a cobalt bit but a lot less material to cut and really just scraping the rim. Gave it a try and worked nicely. Thinking it would just push on a little tight and then a good hit with the hammer should keep it in place. The shaft slid on nicely.... too nicely..




So how tight should the interference fit be? Do I have to hit the shaft really hard to press it down. I have found that it is difficult to even get it to move into the hole. I went back over Davids notes again and noticed that he said he tapers the motor shaft before hitting the shaft in place. I guess I'll have to try that tomorrow. I will have to come up with an ideal distance that the extension sits on the motor shaft and make sure that I can attain this consistently so that one shaft is not longer than the other, and over multiple units.

The 4mm stainless that I ordered is not quite 4mm. It's an extremely tight fit for the Blue U cup seal. Where os the balance with how tight the seal should be around the shaft. I don't want too much friction, too much load on the motor pulling excessive amps. I shaved off about .1-.2 of a mm and then found that the seal slipped on nicely, making sure that it was still sealing all around. I have also had to shave off a little
for the Brass couplings that I use. You can see some subtle variation in the photos. The 4mm oilite bushes fit beautifully.

In the meantime, I found some of the bolts I need to fix the front of the motor to the end cap. These bolts are a really unusual size. 2.6mm. I've had to order more in. I marked out where the motor needs to be mounted and where the shaft will stick out. After marking this point I drilled initially a 2mm hole for the shaft to go through and then a larger one behind on the inside for the recess for the wider bearing mount at the front of the motor.




This is the second of the two motor mounts. I have coated the Renshape outer ring with resin, sand it back and will spray it with primer.
The end cap I am working with I am only using for arrangement, layout and testing for watertight integrity, as it is precisely machined to fit the diameter I am working with. I have made it smooth and have
given it shots of primer however will not be too concerned as after I have done all the testing and layout will transfer all the detail onto the oversize end cap as the final master.


David H

[Davidh]

Hello all,


I gave myself a year to perfect this cylinder caper, but at the rate I'm going this thing might be wrapped up in a couple of months. Still lots of challenges and progress ahead.

After reading Davids write up on extension shaft fitting I got onto tapering the motor shaft by grinding it on the grinding wheel. Rotating at an angle and getting a taper happening. After a while it became a challenge to get the taper even. Now here comes the big question. How do you tap the extension shaft onto the motor shaft and get it absolutely centered? As I tapped both motor and shaft on the back of the vice to press it on I was guessing that it was going on straight. I have been thinking, can you create some jig that precisely aligns the extension with the motor? What amount of off centre do you tolerate? If any. I played around with tapping at a slight angle, being careful about side loads on the motor and all. I have the shaft as close to centered as I can get it.








This is the assembled motor mount test. As you can see the shaft is more centric the closer you get to the motor. ( the motor is running at nearly full power, which I would rarely do ) Clearly the outer tip of the shaft shows the most amount of off centre-ness. Any deviation further down being suppressed by the Oilite bearing, the seal inner support, the silicon supporting the whole structure and absorbing side loads and the seal. Although here you can see the chafing against the lip ain't good as the seal is also oscillating against the shaft. Its wearing and giving me some blue plastic powder worn off from the shaft rubbing. The machined surface against the lip may not be absolutely smooth so some 1200 grade grit may fix that.

Here I am piling up questions for David and he's still getting over a cold. Hang in there mate.

Anyway I siliconed the Seal in place in its recess in the Brass inner support. Came back the next day, checked the seal, some slight movement with the shaft eccentricity, and then plonked it on the pool. No leaks. Wow... I then hooked up a battery running at low speed and plonked it in the water, thinking, its gotta leak now. 2 minutes in the water, no leaks, WHAT! I really like this mount design. It feels smooth and solid and as stated, its not leaking at the moment. Its a nice set up and easy to get to all the parts for replacing if need be.



Things haven't worked this well since Tommy Flowers turned on Colossus and scratched his head as to why everything worked perfectly.

Anyway David, Hope you are better and look forward to hearing your thoughts.


David H

[He Who Shall Not Be Named]

Hello all,

I decided last year that I would give the design and development of Sub kits a break for a year or so. I would spend 2020 fine tuning the kits I have, selling them and adjusting some smaller parts and
just generally lifting the quality of parts, here and there. I am happy with how my latest design , the Delta has turned out and don't think I need to develop a new kit. I guess I'm running out of Russian boats that I like..

I have received emails from people who want to know about the best cylinder to fit one of my boats. I have referred them to subdrivers and other vendors according to the needs that they outlined. However I have been thinking for some time about building a basic cylinder design that could complement the kits I have and make another option available for the market. I have decided to concentrate on one basic design that as much as possible will be interchangeable. The design will feature a single shaft stern end cap "ZB-1" (Zero Bubble-1) for the MIKE, BOREI and RESOLUTION and a twin shaft equivalent "ZB-2 for the PAPA and DELTA.

I intend to go down the same manufacturing path as HWSNBN with regards to Casting the endcaps. I have been curious about doing this for some time. I think that over the past couple of years I have developed the skills and familiarization with this medium and what you can do with it. I realize that this will be low production and the cost of producing these castings is a fair outlay but we do this for the passion.I certainly do. I will be prototyping the endcaps out of Renshape. This is great since Hardrock (Scott) moved down to Nowra (south of Sydney) and left me with a stupidly be piece of Renshape. Thanks Scott, it's awesome!.

I am still in the research phase of this project and have been watching HWSNBN's write ups on cylinder prototyping in "todays work" and " the subdriver becomes modular". I have especially been looking very closely at how everything to do with the shafts are done. I have spent and will continue to spend hours looking over parts lists for seals, o-rings, oilite bearings, bushes, shafts, motors and Polycarbonate tube. I have started ordering parts in bit by bit and started playing around with fit. I have also started doing the sketches to get my ideas right and looking closely at the ones HWSNBN has appear in some of his pics.

My design will be similar to the arrangement I currently employ in my boats. It will use a pressurized system with a single cylinder, no partitions, a pump at the front and the water going to an isolated PVC type ballast tank further forward. A simple, old set up that works well. I am allocating about a year to develop these designs. To what level of detail I will go to with regards to internal components and layout I'm not sure yet. I will at least design and make enough to stop the water coming in from either end!

There is an enormous amount of work to do, innumerable avenues I could go down. Any number of parts options i could take and truckloads of prototyping, testing and development. I am setting myself about a year to do this.






I don't think that contributions for this endeavour will appear as my regular friday night write up. Yep, my social life rocks.... It may be a bit more intermittent. It will be interesting to see what these two designs look like when I have finished with them.

I will start getting into more detail of specifics in the next write up. I will certainly be throwing out the odd question when stumped.


David H

I suggest, to answer your questions, you read over my latest contribution to the, 'today's work' thread. I go into detail on o-ring groove cutting and bulkhead machining. Sorry about the delay addressing this, but I'm still fighting this years bought of the flu.

David

[Davidh]

Hi David,

Yes, I have been looking over the ‘todays work’ posts with a lot of interest. I have been aware that you have been ill and holding off on the questions, however as I progress,
more and more questions are cropping up. I have plenty to ask you.

I have made progress on the shaft design and supports, and am happy with how things are going.

When you are ready..

David H

[He Who Shall Not Be Named]

Hi David,

Yes, I have been looking over the ‘todays work’ posts with a lot of interest. I have been aware that you have been ill and holding off on the questions, however as I progress,
more and more questions are cropping up. I have plenty to ask you.

I have made progress on the shaft design and supports, and am happy with how things are going.

When you are ready..

David H

Blast away, David.

David

[Davidh]

Hello all,

Thanks David, as you can see above there are questions about the interference fit of the shafts and how you go about fitting them I won’t repeat the questions here. I think I’ve also found the answer to the question regarding getting away with only using RTV silicon to hold the whole support assembly together. I will need super glue. But I would love any further info on the shaft interference fit.

P



I have been working on the oversized stern endcap and been working out the locations based on the feedback I have been getting from the test design that I have been working on in parallel. Part of this design consideration has been as to how I will hold the end cap during the machining process for each individual endcap that I produce. I understand clearly what you said about machining each part , it’s like machining individual cylinders sleeves in an engine. ( I have been watching heaps of old YouTube videos about the production lines in WW2 for the Wright R2800 radial and also the legendary RR Merlin, man that was engineering). Anyway I have had to consider how each endcap will be secured to the chuck. Pretty quickly it became clear I would need a fixture clamped by the chuck with the end cap bolted to it. Holes for the motors would double up as the securing method for this process.

I have designed and made a fixture, but I think it is deficient and will go into the reasons why. What I would love to see is photos of your design.



The principle issue I have is that I cannot get it centric. I knew that I would need a dial gauge and saw yours. Got mine out but am having a hard time getting everything dead center.
The design I have will be gripped by a three jaw chuck. This is what came with my lathe and I haven’t got anything else except a face plate which I could use. My fixture is wood, that is fine but the three jaw chuck grips a smaller diameter wood block behind. Over time I can see that the jaws will mark and cause the wood to compress meaning that each time the centricity will deviate unpredictably.. I will have to machine up a plastic or metal section to clamp. One that won’t crimp with the jaws tightened.





Because I have a 3 jaw chuck I cannot tighten each jaw independently. Unlike a four jaw chuck the three jaws are all tightened in unison with a spiral screw arrangement. The only answer I can see it this is to subtly adjust the nuts holding down the endcap to the fixture. The two screws are a smaller diameter than the holes for the twin motors.
In the above pic I managed to get the endcap close enough to scrape out another mill or two off the inside diameter.





So as mentioned I would love to hear your thoughts and see pics of your fixture.

Thankyou David.

David H