Today was not a perfect day for electro sea trials. There was a bit of a breeze, so I had to go both ways and average the speeds. And so many speed figures are listed in 3 significant digits even though the accuracy of the observations does not support that degree of resolution.
I also rigged my most accurate DVM to read voltage into the controller. So now there are two more columns: battery voltage, and power in watts into the controller. I decided that this was necessary in light of the significant voltage sag at higher power levels.
Again, this data is not as accurate as I would like it. If I had a helmsperson who could steer a straight course with minimal use of rudder while also maintaining an ordered current, I could make better observations. Also, because of the wind, I did not bother with anything less than 5a. I did not bother recording prop speed this time. The relationship between prop rotation speed and vessel speed is already established and neither the recording method, nor actual operating parameters have changed in that regard.
5a 1.55kts 49.6v 248w
10a 2.35kts 47.2v 472w
15a 2.95kts 46.7v 700.5w
20a 3.25kts 46.6v 932w
30a 3.7kts 46.1v 1383w
40a 4.05kts 45.8v 1832w
60a 4.6kts 44.9v 2694w
100a 4.9kts 44.8v 4480w
130a 5.3kts 45.1v 5863w
The last current observation is only approximate, as the needle was off the scale and pegged hard. Could have been a bit higher than that.
Multiplying power by .7 should be a pretty good estomate of power in the motor. This is basically a 3 phase AC motor and the controller is of course going to take its share of power. Personally I don't really care about the precise power in the motor, since it is not going to give me a direct approximation of battery energy consumed or remaining. But for the idly curious who can't be bothered with number crunching, that would be by my calculoestimation about 4.1kw, well within the nominal 5kw continuous rating for this motor.
Conversely, since the full power battery current is in doubt, if I assume the controller is outputting 5kw, then the actual battery current would be 158a. But here is another thing... I didd not massage the data to make it fit, and somehow I recorded a voltage at full power that was higher than the voltage at 100a. What if I made a mistake, and entered 45.1v when it was really 44.1v? Assuming 44.1v and 130a would make the controller power to be 5733w, or an estimated 4kw into the motor. Or if I assume the motor is getting a full 5kw, then ... nah. couldn't be. 161a into the controller. From the behavior of the meter as it approached the stop, I dont believe it could have actually been over about 140a and I am actually thinking that my estimate of 130a is pretty close. The full power figures are kinda moot anyway since I will seldom operate at that current level. If the situation is dire enough to require maximum power I am not going to be anal about the exact actual figures, anyway
Anyway, I think I need to find a computer with windohs on it so I can run the controller configuration utility, before concluding testing of the 5kw motor.
One lesson learned from this is if cruising at near hull speed is going to be a normal practice, lead batteries are not going to be very practical even for the budget minded. The better discharge characteristics of LiFeP04 or even NiMH will be worth the large performance boost, especially since you can also discharge them much deeper than lead batteries without seriously shortening lifespan. For up to half of hull speed or so, flooded cell lead batts are fine, if money is tight.
Another lesson... the different voltages shown, with different loads, make assuming nominal voltage, even used relatively, to be misleading, for the purpose of recording performance data. Certainly, since an ordinary D'Arsonval type analog ammeter is so inexpensive and easy to install, battery current is a logical operating parameter during ordinary operation of the vessel. But for analyzing performance, both the current and the actual real world voltage are needed, and also watts, for convenience.
Third lesson... make sure your current shunt matches your ammeter, or else construct a corrected meter scale card.
Fourth lesson learned... just because your connections WERE tight, doesn't mean that they ARE tight. Stranded copper compresses, and there is always vibration and movement to deal with. Outgassing batteries can create huge messes of green gunk on copper terminals. Water should be checked in at least one cell per battery before each day's use of the propulsion equipment.
One fact of interest is that for a thumbnail projection, Peukert effect can be conveniently ignored, at a discharge current equal to the 20 hour discharge capacity, in amp hours. Ths would be 11 amps for my bank, so conversely, at 11a battery current, I should be able to go for 10 hours and still not discharge deeper than 50%. Interpolating from the test data, this would give me a range at 11a, of about 25 miles with no contrary wind or currents. I know I am drawing a lot of conclusions in the face of a mountain of unanswered variables, but I am thinking this will hold pretty close in real world conditions. At 5a, I would actually gain a bit relative to the 20hr rate, so I should be able to count on 22 hours run time, and a range of 34 miles, at 1.55kts. Boggles the mind, even if I am certainly NOT going to stand at the tiller for 22 hours making half of a liesurely walking speed.
---In electricboats@yahoogroups.com, <no_reply@yahoogroups.com> wrote :
Looks like thunderstorms today so even when the batts are topped up I won't be able to speed test. Hopefully first thing in the morning. With this discovery and correction in my instrumentation, and hopefully rigging a 20 point light so I have legal navigation lights, it will still have been a productive day. Maybe I will borrow a windohs computerr from somebody (I run linux and the configuration utility for tthe controller only runs in windoze.) I will make sure that the max battery current and motor current are where they need to be. I didn't think about that before but the DC into the controller should be about 1.43x the motor current, and that may be part of the reason I never got the expected power level out of the old controller. It was limited by the max battery current setting, I think, which I probably set equal to motor current limit. Documentation for these Kelly controllers is not what it could be.
| Reply via web post | • | Reply to sender | • | Reply to group | • | Start a New Topic | • | Messages in this topic (18) |
No comments:
Post a Comment