I had one of those Grapnaur bow thrusters. Put it into it's own little WTC. It ran ok. Then I left it under 24" of water over night and found that the seals leaked. I guess it's good enough for a surface craft.
I was curious what it would be like building one, and how it would perform. My first prototype looked like:
Actually, my initial test did not have the hole in the center of the face plate. I was more than disappointed to find it was no better than the Grapnaur thruster. Both pumped about 2.5 gallons of water in 22 seconds.
I looked at the pump and thought it really wanted to be a centrifical pump, so I drilled out the center of the face plate. Ohh, then it would pump the same volume of water in 12 seconds. Now that's what I was looking for.
The size factor of the was all wrong though. I really wanted to have opposing bow and stern thrusters so the 60" Sierra could really turn on a dime. So I started wondering around looking for something I could turn into a compact thruster pump, and I found 2" copper end caps. This is what the end product looks like:
There are only three parts to make: the impeller, the housing, and the motor face plate. This is how I made them.
First I cut the impeller parts roughly to size.
I marked the circle oversize because I new I was going to use the lathe to bring it down to size.
Then I soldered the a collet onto the impeller plate. I use a 1/8" stainless steel rod to ensure the collet was aligned with the hole in the plate. I would use less solder if I did it again...
The handy thing about that stainless rod: the solder could not actually bind to it, but there was sufficient grip that I could use it on the lathe and I turned the disk to size.
After turning it down a sharp tap released the 1/8" stainless rod, and it was time to cut slots for the impeller blades.
Here you can see me using a 1/8" milling bit in the collet to ensure I was properly aligned vertically.
After getting the right vertical height I switched to the 1/16" bit and cut the slots for the impeller blades. This ensured they were perfectly opposed to each other.
These are the parts prior to soldering on the blades.
And this is a rather out of focus picture of the result.
I actually dropped this onto one of those magnetic propeller balancing things and filed away some of the solder so that it was nicely balanced.
Next it was on to the pump housing. First thing to do is drill the 5/8" hole for the input / output tube. This definitely required a jig to give me the correct alignment.
I had to wrap tape around the 2" copper end cap so that it would be snug in the hole I could drill. Note that the end cap is 1/16" proud from the hole. I did this on purpose so that when I clamped it into place the end cap would be held in place and not rotate under the drill.
This is it all clamped and ready for drilling.
You can see the tape around the end cap. Hmm. I had the jig setup so that the center of the 5/8" drill went on the edge of the end cap. It was easier to make this way. I wonder if the pump would have been more efficient if the input / output tube was closer in to the center of the end cap. I guess that's one of those things I will never know.
This is the very very cautious and well clamped drilling operation.
I say cautious because the bit really liked to grab the copper and phyically pull the chuck of the drill press spindle... Still after all the drilling was done and the clamps removed:
I was pretty happy with the result. I had tried manually using a dremmel tool to achieve this before. It did not work out well. This was something I could use.
Time to build the input / output tube. First I cut it to length then slipped it into place and marked the metal to remove.
With these markings I then removed the bulk of the material. This was an iterative operation as I got closer and closer. In the end it looked like:
Sorry it's a little out of focus again. The tube has under 1/16" proud all around. When I got it to this stage I dropped it into a vice applied a little flux, and wrapped a bead of silver solder to it.
Then I used a torch to heat from the bottom of the assembly. So by the time the solder melted on the top the surface tension and gravity allowed it to flow around the entire joint.
Then I finished the job of grinding the tube so that it was flush to the end cap.
Finally on to the aluminum plate I used to mount the motor to the pump housing. I started by drawing the circle and four mounting tabs on the aluminum, then cut it roughly to shape.
Here you can see the rough version attached to the motor.
You may have noticed some of the pictures for the pump housing had four slots cut in the front edge. You can see why now. In my first pump I left cutting those slots till the end. That made it very awkward to work on (after the in / out tube was attached). The second pump I cut the slots before doing anything else.
To bring the aluminum plate to size I wrapped the motor with tape until it was a snug fit for 2" copper tube then used the copper tube as a filing guide.
There was a fair amount of filing to get the aluminum to the correct circle, and then to get the slots to fit nicely with the cuts in the pump housing.
In the end it looks like:
I had thought that I would need a mechanism to hold the motor in the housing, but with both pumps the friction fit was perfectly adequate.
Note I did not go for tight tolerances on the impeller. I'm sure the occasional piece of flotsam will make it through the pump. I figure 1/16" gaps everywhere will allow it to be non-sensitive.
As I mentioned earlier it works quite effectively. I have a 65 turn motor on it to generate significant torque. When it is running at full speed it needs to be over 1" submerged or the hole in the front of the housing will suck air from the surface.
I have used the pump extensively in tests and have left is submerged for days, then taken it in and out of the water. Obviously the copper, brass and stainless parts have seen no ill effects. The old sign of un-happiness is:
That replaceable lead that connects to the brushes. It's pretty minor, and coming from a replaceable part. I keep thinking there must be something conductive I can coat it with to prevent the water getting to it.
Anyway, it was a fun little project. I'm hoping the bow and stern thrusters will keep the Sierra from running into other craft, or obstacles.
Of course, if you just want to take an Grapnaur bow thruster and run it wet, then drill a hole in the face plate and I think you will get an impressive improvement in the amount of water it pushes. But I have not tested that the motor on the Grapnaur bow thruster may rust more that the motors I am using.
It's a great day for R/C sub building.