1/48 scale Type VIIC U-201 build

[Monahan Steam Models]

Nick,

I received my Arkmodel 1:48 scale Vll today. I have a question for you about the hull sections! Did you use the provided screws and the Plastruct Bondene solvent cement to glue the hull half sections, or just the solvent cement?



Rob
"Firemen can stand the heat"

Rob,

I used both types of solvent only because I was originally having trouble with the orange label solvent. David helped solve the issue and that was due to oil contamination of the molded parts. Scrub the hell out of them with scouring powder and degreaser before you do anything. I have yet to do this to my parts as I stopped glueing parts together after this was discovered and switched my attention back to design and rough assembly work. I do plan to use both the solvent as well as the screws provided when the time comes to bond everything together. I recommend dry assembling the kit a few times with the screws. This will familiarize you with how things go together and allow you to see what screw holes need opening up and adjusting. Better to solve those issues now.

Nick

[Monahan Steam Models]

Great post! I love information like this- nice job..

Thanks BG!

I’ll admit that I was a little skeptical if I still needed the diaphragm equalizer if I was able to achieve the 100% oil fill. Yes, David don’t kill me, I’m just being honest and look, boy was I wrong! Yes it the diaphragm equalizer definitely helps!

Going to go eat some humble soup now…

Nick

[He Who Shall Not Be Named]

Thank you sir! That diaphragm you suggested trying, came in handy for the heated oil expansion! Didn’t see that one coming but, duh!

Nor I. But, thinking it through: the viscosity of the oil dragging down the motor armature and the 'water-hammer' between gear teeth is making the motor pull more current, hence the heating. You're taking us all to school, Nick. A most interesting (and well presented) thread. Tribel knowledge is the better for this kind of work. Nice thing about r/c submarines is that servo response speed is not critical.

[Monahan Steam Models]

Nor I. But, thinking it through: the viscosity of the oil dragging down the motor armature and the 'water-hammer' between gear teeth is making the motor pull more current, hence the heating. You're taking us all to school, Nick. A most interesting (and well presented) thread. Tribel knowledge is the better for this kind of work. Nice thing about r/c submarines is that servo response speed is not critical.

I completely agree with everything you have said. In regards to the heat build up, another source to add is the drag created by the servo horn pressing down on the o-ring. This is an area that could be dialed in a bit better to free up a small bit of resistance. Also the cooling effect of the servo being in the drink may help reduce some of the heat build up.

Interesting stuff regardless. Collaborative experiments like this and shared knowledge is what makes this sort of thing fun! We all learn from one another! I certainly wasn’t the first to try this. I followed what Bob Gato, Sub Ed, Bob Martin, Steve H and others did way before my attempts and now have implemented your recent piece of the tech puzzle into the equation. Good stuff!

Nick

[He Who Shall Not Be Named]

I completely agree with everything you have said. In regards to the heat build up, another source to add is the drag created by the servo horn pressing down on the o-ring. This is an area that could be dialed in a bit better to free up a small bit of resistance. Also the cooling effect of the servo being in the drink may help reduce some of the heat build up.

Interesting stuff regardless. Collaborative experiments like this and shared knowledge is what makes this sort of thing fun! We all learn from one another! I certainly wasn’t the first to try this. I followed what Bob Gato, Sub Ed, Bob Martin, Steve H and others did way before my attempts and now have implemented your recent piece of the tech puzzle into the equation. Good stuff!

Nick

Yeah, O-rings present a lot of friction to rotation. A cup-seal would be ideal, but that means inserting a very short length of brass tube over that portion of the splined output shaft, and then working out a means of mounting the cup-seal to the top of the servo body so its lip can engage the brass tube. (A fools-parade at that point, I'm afraid, Nick. We start chasing our own tail. Diminishing returns for time and effort).

David
Rube Goldberg is my Personal Savior

[Monahan Steam Models]

Yeah, O-rings present a lot of friction to rotation. A cup-seal would be ideal, but that means inserting a very short length of brass tube over that portion of the splined output shaft, and then working out a means of mounting the cup-seal to the top of the servo body so its lip can engage the brass tube. (A fools-parade at that point, I'm afraid, Nick. We start chasing our own tail. Diminishing returns for time and effort).

David
Rube Goldberg is my Personal Savior

Because I have a habit of doing things my own stubborn way, I chose a o-ring that had an ID much smaller than what would typically be used in a ideal rotating fit. The shaft splines already presented a less than ideal smooth sealing surface for the o-ring so I chose to significantly undersized the ID of it in order to stretch it doing two things. First the stretching would reduce the o-rings overall installed height and second would force it to seal better on the splines. The sealing is now dependent on the squish between the bottom of the servo horn surface and the servo case surface. So the O-ring travels with the shaft. Drag is adjusted by the squish factor. A better chosen o-ring thickness could help reduce drag more perhaps. The mineral oil does seem to help provide lubrication to the O-ring so far.

Nick

[redboat219]

the cooling effect of the servo being in the drink may help reduce some of the heat build up.

Nick

maybe add some heat sinks or a U-shape brass tube for the oil to circulate by natural convection to cool the servo?

[rwtdiver]

Rob,

I used both types of solvent only because I was originally having trouble with the orange label solvent. David helped solve the issue and that was due to oil contamination of the molded parts. Scrub the hell out of them with scouring powder and degreaser before you do anything. I have yet to do this to my parts as I stopped gluing parts together after this was discovered and switched my attention back to design and rough assembly work. I do plan to use both the solvent as well as the screws provided when the time comes to bond everything together. I recommend dry assembling the kit a few times with the screws. This will familiarize you with how things go together and allow you to see what screw holes need opening up and adjusting. Better to solve those issues now.

Nick

Nick,

Thank you very much for getting back to me, I really appreciate your advice, and will certainly follow it.

I am going to start my own build blog on this forum of my Arkmodel, so as not to interrupt yours or anyone else's theme! I am sure I will have a lot of question, so I hope you do not mind if I chime in on your build with some of those questions?!

Thanks again Nick for all your help and input!

Rob
"Firemen can stand the heat"

[Monahan Steam Models]

Romel,

The amount of heat that built up was not very much and I was stressing the servo a lot more than what it would normal see during normal operations. Adding special features to help cool the oil are likely not needed.

Rob,

No problem at all and ask as many questions as you need. I do follow your builds so if you want to ask a question that specifically relates to your build in your thread, I will see it and do my best to help. Or if you want to ask a question here on this thread that works too.

Nick

[Monahan Steam Models]

This morning I wanted to get some real data regarding the drag or resistance the newly waterproofed servo was experiencing since the modification. Real data is always better to look at verses guessing or lots of arm waiving.

I used the bench top power supply set to 6 volts constant output and set the current draw at variable. This allows you to view the amps or in this case milliamperes or mA draw in real time. I also used a servo tester set to the automatic function in order to get repeatability across each servo test.

First up was to test a standard unmodified SG90 servo. The current draw in automatic mode averaged about 120mA.

Next up was to test the modified servo as is from yesterday. The current draw in automatic mode averaged about 485mA.

The next test to the modified servo was to remove the servo horn and o-ring. The current draw in automatic mode averaged about 235mA.

The final test to the modified servo was to reinstall the o-ring and servo horn to find the optimal amount O-ring squish or sealing without leaking oil and still having the equalizer diaphragm work. The current draw in automatic mode for this scenario averaged about 260mA.

Nick

[He Who Shall Not Be Named]

This morning I wanted to get some real data regarding the drag or resistance the newly waterproofed servo was experiencing since the modification. Real data is always better to look at verses guessing or lots of arm waiving.

I used the bench top power supply set to 6 volts constant output and set the current draw at variable. This allows you to view the amps or in this case milliamperes or mA draw in real time. I also used a servo tester set to the automatic function in order to get repeatability across each servo test.

First up was to test a standard unmodified SG90 servo. The current draw in automatic mode averaged about 120mA.

Next up was to test the modified servo as is from yesterday. The current draw in automatic mode averaged about 485mA.

The next test to the modified servo was to remove the servo horn and o-ring. The current draw in automatic mode averaged about 235mA.

The final test to the modified servo was to reinstall the o-ring and servo horn to find the optimal amount O-ring squish or sealing without leaking oil and still having the equalizer diaphragm work. The current draw in automatic mode for this scenario averaged about 260mA.

Nick

Excellent test protocol, Nick; you verified -- and segregated -- the two causes of drag on the drive train: fluid viscosity drag (mechanics working in a liquid environment); and mechanical drag (the O-ring drag on the output shaft). Your observations and application of your findings led you to the adjustment of the O-rings tension between horn and servo case, which reduced the off-the-chart half-Ampere draw to a quarter-Ampere draw (still about twice the in-air current draw, but reasonable).

The scientific method on display here! Good stuff, sir. The output transistors of the PCB's H-bridge thanks you!!

Observe and record the characteristics of a system -- work out a hypothesis on what is the likely cause of the observed characteristics -- formulate a theory of how that system works and apply theory to change the systems characteristics.

https://youtu.be/bTlXZf5qsZI

David

[Monahan Steam Models]

Since the last update of this build thread where I was experimenting with waterproofing sg90 servos, I redesigned the aft motor/servo mount to accept a pair of Savox IP67 rated micro servos. I made this choice to use these servos instead of the sg90 type servos based on the fact that they are better constructed out of the box and draw much less current during normal operation compared to the sg90 servos.



Here’s the new version of the aft motor/servo mount with the slightly larger Savox micro servos installed. Linkages to the stern planes and rudder pushrods still need to be made up.



The new forward servo mount with the Savox micro servo and bow plane linkage installed in the type VII hull.

Other work with this build has been focused on the “in the wet” wiring harness build as well as the aft WTC main battery access simplification. Have made great progress in these areas and will share those build updates soon once I get a chance to document and complete the work.

Nick

[Monahan Steam Models]

Was able to spend some time today on the WTC’s aft module section.

Before deciding to break the original WTC up into three separate modules, the order of operation to gain access to the batteries was to disconnect all external hoses and electrical connections and then pull the whole WTC out of the vertical split at the aft of the hull. The next step was to to pop the forward compartment end cap with it’s equipment shelves attached to the end cap and disconnect the data and power connection. Then I could remove the aft end cap, disconnect the data ribbon cable and pull the aft equipment shelves out to gain access to the batteries which still required removing two small long bolts that doubled as fasteners to hold the upper equipment shelf to the lower as well as act as a stop to keep the batteries from sliding out. This process wasn’t hard and was relatively optimized in the original design. It would only take a few minutes to do. Still it wasn’t ideal.

The new horizontal water line hull cut made removing the original WTC a lot easier but I felt there was still an opportunity for improvement with the complex original WTC system. This brought the sectional modular WTC design revision into play which further reduces the amount of stuff that needs to be disconnected and removed from the hull each time the boat is serviced or operated.

Today’s work involved replacing the lower aft equipment shelf with a design that allows the two LiPo battery packs to be removed from either side without the need to unbolt anything or disconnect any wires other than the battery plugs. It was also time to get rid of the 8 wire rainbow ribbon cable which was the original data connection to the forward WTC compartment. The new 17 wire water proof connection in the aft end cap now carries those data signals along with the data signals, positive/negative power for the rudder and stern plane servos and the four wire usb connection to the Arduino control on board. So that wiring needed to be done too.



Here’s a side view of the newly populated lower equipment shelf. The short side rails that kept the batteries in place have been removed from the design.



The new side rails feature a tongue and groove installation with the lower equipment shelf.



Here is one of the new side rails slide into place about halfway.



And another photo of it in place.



The wiring mess that I started out with this afternoon for the new 17 wire connection. Note the ribbon cable is still in place.



The new lower equipment shelf populated with all the gear. No space is wasted. This area features both the 12 volt and 5 volt main busses. The main fuse. The main battery connection. The remote on/off board as well as the high amp on/off switch. The 3 amp 5 volt reduction board as well as the BLM module.



One of the last soldering job connections to the new 17 pin harness was the usb data plug for the on board Arduino unit. Now I can modify the PID settings and Kalman filter code used in the static diving routines while the boat is water tight.



The following photos show the aft module sealed up. Again no space was wasted. The last bit of wiring tidying up work needs to happen to the main high/low power connection at the lower right of this photo.





Other than that, I’m happy that now all the aft module takes is sliding off the outer acrylic tube and removing the batteries for servicing.



Well that’s besides pulling these four connections…. Still way simpler than what I had before…..

Nick

[SubDude]

Nick, are you leaving the antenna bundled up? Very smooth...I mean it is just SMOOOOTH!!!

[Monahan Steam Models]

Nick, are you leaving the antenna bundled up? Very smooth...I mean it is just SMOOOOTH!!!

Steve, thanks!

The antenna is just bundled up for now. There is a hole for a brass nipple to be installed in the aft end cap that the antenna will pass through. Still need to finish the work needed on the end caps.

Nick