Hi Guys,
Finally getting around to catching up on my write ups now that I am on holiday, my last testing session down at the lake was on 17/09/16 and it was very successful, armed with a new Turnigy marine 120A ESC, it has cured all of the running issues that I was experiencing, it was however not without its problems in the beginning.
I bought the ESC, like most of my RC stuff, from HobbyKing, I already have two of these lower rated ESCs in my brushless Trent so already had a programming card for it, I didn't come across any programming issues for the Trent as I only wanted to change the ESCs to forward and reverse operation, which is the first option on the menu.
The real confusion arose when I wanted to start changing the timing settings, which as the picture shows, is item 09 in the menu system, but the LED readout on the card only went as high as 04 items!?!?
In the end after a lot of searching around the internet on various forums, it became clear that this programming card is shipped with various sets of software and stick on menu plates, which do not always match, fortunately, the programming card was displaying the correct number of menu items, with the correct number of options per menu item as according to the manual, so using that as a guide I programmed the ESC and all appeared to be fine.
So with new ESC installed and operational I went down to the lake armed with a couple of large Overlander outrunners and my 6S 5000mAh Lipo pack, testing consisted of installing each motor in turn, programming in a timing option, then recording a static thrust test on the slipway using the POV GoPro, at the end of each run I recorded the results down in my notebook and this is what I found.
Outrunner 5045/10 720KV on 6S (16128 RPM)- 26.25º Half Throttle 17A Full Throttle 84A
- 22.5º Half Throttle 20A Full Throttle 85A
- 18.75º Half Throttle 15A Full Throttle 76A
- 15.00º Half Throttle 14A Full Throttle 76A (73.7A on lake)
- 11.25º Half Throttle 15A Full Throttle 75A
- 7.5º Half Throttle 15A Full Throttle 74A
- 3.75º Half Throttle 16A Full Throttle 73A
Outrunner 5055/06 580KV on 6S (12992 RPM)- 26.25º Half Throttle 16A Full Throttle 67A
- 22.5º Half Throttle 15A Full Throttle 66A
- 18.75º Half Throttle 15A Full Throttle 66A
- 15.00º Half Throttle 15A Full Throttle 78A (on lake)
- 11.25º Half Throttle 14A Full Throttle 65A
- 7.5º Half Throttle 15A Full Throttle 58A
Now it must be stressed that these figures are not going to be super accurate, the watt meter didn't want to settle quite a lot of the time, so most of the figures are averaged to try and give some kind of meaningful comparison.
Comparing my result to other tests done online, I am not very surprised with the outcome, I started testing at one end of the timing range, the motor will run, but will more than likely be running very warm with a high current draw, you then cross a threshold and the timing starts to match the motor configuration much better, which is indicated by a sudden drop in overall current draw, this can be seen in the 720KV results when I drop below 22.5º of timing, half throttle is reduced by 5A and full throttle draw is reduced by 11A!
As far as I understand, this proves that 18.75º with the 720KV motor is the best compromise for a balance of max power and efficiency, whilst keeping within the motor's maximum operating rating of 80A, which means there will be no overheating issues, so no melted windings!
Looking further into the results it is important to note that static lakeside testing will give different results to running up and down the lake under normal operating conditions, this can be seen from the, "on lake" figures in the chart, the reason behind this is that the loadings within the drive are different whilst running static as apposed to free running on the lake, the brunt of the force being transferred from the impeller comes back up the impeller shaft and into the coupling and on a couple of occasions I had the coupling slip and the impeller start the catch the inside of the impeller tunnel, when I do my next redesign I will be looking into different couplings methods and may introduce a force plate to isolate the motor and coupling from longitudinal loading.
After static testing was completed for each motor, I chose what I thought was the best timing setting and sent the prototype out for a couple of power laps around the lake for comparison. The 720KV motor was sent out with a timing setting of 15.00º, which gave an overall lower full throttle reading than the static testing, but I think this is more down to the motor not spooling fully before the ESC stepped in to protect the battery from over discharging. The opposite was found for the 580KV on 15.00º of timing, with a full throttle power run returning a higher figure than the static tests, but looking at the figures, it would appear that this motor would respond better to a lower timing input and at a later date I will go back and retest the 580KV on the 3.75º and 0º settings.
As with any development process, a fair amount of trial and error is required and this project is by no means any different, every time I take another step forward in overall output I run into one or more problems to be solved. I am fairly confident that I have got a handle on the motor / battery / ESC combination issue, which means that I can put the power into the drive, I am facing new challenges at the upper power levels due to the fairly in-compressible nature of water, effectively, to make better use of the power I am putting in, I need to change the design of the jet unit to allow for a more efficient through put at high speed, allowing for easier ingestion and expulsion of water.
By smoothing out the flow through the unit this should stop the impeller trying to pass more water through the stator housing than it is capable of flowing, because if you exceed the available flow rate out of the unit, all you will end up doing is causing the internal pressure of the unit to rise, which can be very crudely measured by the flow of water coming out of the cooling water port, which is tapped off the side of the stator housing, so the higher the stator housing pressure, the more cooling water output there will be.
Cooling water doesn't flow below 15% throttle and upto to around 70% it is quite proportional to throttle input, above and beyond 70% the output increases significantly, suggesting that pressure is increasing disproportionately to throttle response, some ideas to combat this that I am going to explore are:
- Increased overall length of the tunnel section - This will allow for a longer, larger inlet which can be shaped to allow water to flow upwards more naturally instead of relying on the the drive sucking water out of the lake
- Increased stator housing diameter - At the the moment the stator housing is smaller than the impeller tunnel, a bigger housing allows for a greater volume to flow
- Impeller design - Searching on the internet there is quite a wide variety of impeller designs, I intend to try different configurations to see what affect there is on performance and current draw
On this outing I very briefly tried putting a ramp behind the inlet on the bottom of the hull, the idea was to slow the flow of water passing the inlet and force it up into the jet inlet, this kind of worked up to a point, but at higher speeds the water pressure in front of the ramp built up until the back of the hull would jump to equalise the pressure, the cycle would then repeat which lead to the back of the prototype jumping up and down at high speed, so the ramp was removed and the whole exercise chalked up to experience.Moving forwards I am starting the design and build of the MK2 prototype, which will consist of twin jet units in the large white and orange fibreglass hull I bought a while back, this means larger jet units will be required and to this end I have just bought a larger 3D printer, which will allow me to make tunnel housings upto 250mm in length if I want to.But the MK1 prototype is not finished with just yet, I intend to go back to the lake and re-test the 809KV outrunner I have, as well as the seemingly indestructible 1900KV inrunner, I also have a 775 brushed motor that I might test as well just for kicks.One last improvement that I have on the cards is a redesign of the reversing scoop mechanism, it takes a real beating hammering across the lake and the bolts that hold it together keep falling out and although it does make the prototype move backwards, there is no real control, so I will try and address these issues as well.Alex
Author
Topic: Water Jet Drives & 3D Printing (Read 96696 times)