twofer

[webby2]

I am going to place the test stuff here.

Right now I have only tested the benchmark setup, I am still making a few more parts for the test-bed itself.

The motor will be run at 3V for all tests.
I am using the same loaded motor with the same 17.2 ohm resistor across its terminals for all of those tests.

The motor with nothing else
0.03A to 0.08A

The motor in the housing
0.05A to 0.08A

The motor in the housing with a second housing output shaft connected directly to the first housing output shaft
0.08A to 0.09A

The motor in the housing with a second housing output shaft connected directly to the first housing output shaft with a motor in the second housing
0.19A to 0.21A

The motor in the housing with a second housing output shaft connected directly to the first housing output shaft with a motor in the second housing and a 17.2 ohm resistor across the second motor terminals
0.29A to 0.31A
Voltage across the resistor
0.855V

The motor in the housing with a 65T gear attached to the output shaft and the second housing with a 33T gear attached and no second motor
0.9A to 0.11A

The motor in the housing with a 65T gear attached to the output shaft and the second housing with a 33T gear attached WITH motor
0.32A to 0.38A

The motor in the housing with a 65T gear attached to the output shaft and the second housing with a 33T gear attached with motor and a 17.2 ohm resistor across the second motor terminals
0.43A to 0.47A
Voltage across the resistor
0.534V

[webby2]

due to extra frictional losses with my system I had to increase the voltage to 6V

Here then are those numbers with a 6V input.

They are
no second motor
second motor
second motor with 17.2 ohm resistor across the terminals

benchmark 1:1
0.12 to 0.14
0.24 to 0.26
0.49 to 0.52 v out 1.96 to 1.97

benchmark 1:2
0.15 to 0.18
0.42 to 0.47
0.74 to 0.76 v out 1.48 to 1.52

system
0.37 to 0.41
0.53 to 0.56
0.68 to 0.72

Observation: with the 2:1 benchmark the load on the motor caused it to heat up and slow down, so I cooled it off several times and ran a short loaded test, the 1:1 benchmark did not appear to exhibit this to the extent of the 2:1 and the system did not appear to exhibit this problem to that extent either.

[webby2]

My system is a geared transmission that is a 1:2 step up ratio, for every 1 RPM in on the input  the output has 2 RPM.  This is a constant ratio.

It uses the twofer gear to communicate the input work through the system and to the output.

I am encouraged by the test results and am now planning on building a better version using more efficient motors and generators to hopefully loop the system.

[webby2]

Thought of something stupid.

How much does it cost me to run the system without the twofer gear engaged?
at 6V it costs between 0.31A to 0.34A . 
That is a lot of the increase in draw,, so maybe I need to start with better motors and generators.

[webby2]

Another stupid thing,,

I do not have the system balanced, at all.
I spun up another test-bed that is about the same mass and stuff as unbalanced as twofer was and had about the same draw readings, well, then I balanced that frame and the draw numbers came down to between 0.16A and 0.24A,, and that was not "perfect".

[webby2]

I am printing off a bunch of new parts for twofer.

I am getting rid of some redundant parts and a new frame so that I can balance it.

I was not going to disassemble it but since I am a group of one I can do as I please   it does not really matter since I still have all of the .stl files to reprint it as it was.

[webby2]

I have the new frame and stuff installed and ran the tests again and the numbers were so close to what I already have it was incredible.

Not what I was looking to happen but that means I am not seeing the resistance added by the mechanics of the system properly.

[webby2]

probably not the correct way of going about this.

Since the extra drag of the system is high compared to the motors abilities I thought that I would just increase the input voltage until the motor was t about 1A draw.
10V fully loaded
0.98 to 1.01

Then I set the same input motor up against the same output load in a 1:1, directly coupled together
0.74 to 0.8

Then I set the same input motor up against the same output load at a 1:2, geared up
1.07 to 1.12

The numbers are still really close

[webby2]

been running more tests and one thing is for sure and that is I am overloading my motors,, short run times before I can feel the motor heat up.

I am trying to rationalize what it is I am seeing, my time only costs me my time but buying parts to try and make a "real" test-bed costs me money,, so I am trying to not spend anymore than I have to.

I can run twofer in a few ways but the more important part is that I can run it just as a 1:2 step up gearbox and when I do this the draw for a loaded run is slightly higher than the 1:2 step up geared system.

10V
twofer loaded
1.17 to 1.19

benchmark
1.07 to 1.12

in a few tests they were closer but it looks like this for the most part.  It does not help that I can only run them for a minute or less,, I am trying to not let the motor heat up and reduce the input draw that way.

[webby2]

twofer has many parts, it has 2 frame pieces and a few large gears.  I thought to check on the windage issue to then only spin up the frames and two large gears, keeping it balanced, and the draw was a lot.

10V
only the lower frame

0.15 to 0.20

The lower frame\upper frame and 2 large gears

0.29 to 0.38

and then I also noticed just how much of a fan this thing makes,, not good for cheapish building,,,

I kind of already knew that I would need to pass a lot more work through the system to mitigate these losses or maybe some kind of cover???

[webby2]

what I think I am seeing is that twofer when run as just a 2:1 gearbox  behaves very similar to a conventional 2:1 gearbox but when run as a twofer, and if I can mitigate the windage losses, then it would have the draw much closer to a 1:1 gearbox while providing for a 2:1 output.

[webby2]

I am still playing with the windage stuff,, but I have a partial system built up with enough of the parts installed to run a quick spin up to see what my windage losses are and then I will assume that they will stay about the same for the full system.

This was just a windage test,, not all the parts are installed,, worth repeating

6V 0.14A to 0.16A

I would say that that is a significant reduction in the input cost for twofer, previously it was 0.31A to 0.34A for the same kind of test.

Now to make the rest of the parts

[webby2]

Things are not testing out very well for my concept.

I now have the air resistance thing down to a minimum but the costs have not dropped like I thought they might, so it is still working just like a standard gearbox.

I pulled apart one of the motors and it looks like there is a thermistor or some such inside so I am assuming that my test numbers are not real either.

[webby1]

so there is a mystery force of opposition that I did not know about.
This force is seeming to be interesting and so I am building a few test-beds to do the same kind of thing but in different modalities to see if I can figure out how to use this force of opposition.