Wattmeter Quandry?

[SIMON CRAGG]

As some of you may see from my other topics, I have been converting an 80" wingspan PT-19 from IC to electric. I have had a LOT of help from this forum. However, I have come across a problem today, which I just cannot get my head round. Earlier today my brand new Wattmeter arrived, as my old one was suspect. Plugged everything in and:

17X8 TGS PROP

6S 5000MAH 35-50 PACK

100 ESC

SK3 5055-430 MOTOR

71A

23.41V

1710W

7800 RPM

77% Prop loading.

Nowhere near the e.calc figures, but then again they do not list TGS props.

After trying a few more props, I balance charged the Lipo and tried the 17X 8 again, expecting better results:

59.89A

23.70V

1432W

7800RPM

76% Prop loading.

I am completely baffled as to why the large drop in readings, with a fully charged balanced battery. Can anybody shed any light on this, as I am starting to get frustrated with it, after so much effort!.

[MAD Dave]

Experimental error Simon? Suggest try again a couple more times to consider the range of results. Talk Sunday.

[PatMc]

You have got better results - same rpm for lower current. Perhaps the motor bearings have loosened up.
In fact the new rpm figure is probably higher than the first but the difference isn't enough to register, does the tacho reads in 100 rpm increments?

[Chris Walby]

Hi,

I eluded to this in your other post, if you are expecting accurate results then you need to spend serious money on test equipment and conduct the tests in a very controlled environment. If any of the parameters change it will change the results.

Air temperature, motor temperature, ESC temperature, prop balance and probably most likely battery temperature will all impact on final results.

I am sure Dick can provide far more accurate data regarding battery performance, but for competitions people take great care and time as they push closer and closer to gain ultimate performance.

 

Try this, but don't change anything just do each test over a 5 minute period

Three tests of 20 seconds at full throttle with 2 minutes between each test. Record the results at the end of each 20 second run.

IMHO my money is the results will not be repeatable, but feel free to prove me wrong. At least it will give you a bench mark of repeatability.

PS - as for the watt meter, measuring dc it may be accurate but the waveform generated from the ESC will be very unlike smooth DC. Instruments can be very non linear with high frequency noise imposed on the DC.

 

 

 

 

 

 

Edited By Chris Walby on 02/11/2019 06:18:46

[gangster]

Not sure I would want to run the motor at full chat for 20 seconds On the ground In the air the prop unloads. Also we have to be caref if anything that gives a digital display. Because we can get a display that reads out to a few places of decimals we have been accustomed to assume an accuracy that is not there and we don’t need anyway.

[SIMON CRAGG]

Thanks for the replies. I am going to try and calibrate the throttle again (done via a programme card), and change the Deans connectors to EC 5'S, as I have just read that Deans are only good for up to 3s!. Still got a vertical learning curve!.

[gangster]

Not sure about Deans only being good up to 3s. I have always assumed that The suitability of a connector was dependent on the current. Looking at your set up it is possible you could be over their limit. I only use deans but with 4S 50 amp max.

[Gordon Tarling]

Simon - I've just run your given configuration in Ecalc and it's showing 1755W, which is pretty close to what you're getting and ties in very well with the numbers I get on the same setup in my Carbon Cub. Not sure what prop data you're entering, but I always use APC-E in Ecalc unless there's good reason to use something else. Check your other numbers too. As gangster says, connectors are current limited normally and Deans are OK for use up to 50A, so I'd recommend using different connectors. I use EC5 connectors in my Cub.

Edited By Gordon Tarling on 02/11/2019 09:34:06

[Simon Chaddock]

Simon

Are by any chance these figures the maximum readings as recorded by the Watt meter?

If so they need to be treated with a pinch of salt as they may represent just an instant peak rather than a true average power reading.

I really only use a Watt meter to give me confidence that I am not exceeding any components limit and to give a 'ball park' figure to the Watts being used.

[SIMON CRAGG]

Gordon, I have just sent you a pm.

Simon, yes max readings. But they are nowhere near the e.calc figures!.

[Martin Harris - Moderator]

What did your program estimate for prop rpm? I'd suggest that your rev counter is the most likely instrument to give consistent results so as you're achieving similar results on both runs and if they are somewhere in the predicted range, with indications that you're well within the capabilities of the equipment, why not just fly it and see if the model performs as you'd like it to? You can always experiment with different props afterwards, keeping an eye on maximum current draws and component temperatures as a check on both specification and cooling effectiveness.

P.S. I agree with Chris about the accuracy of typical Wattmeters - I found mine to be about as much use as a chocolate teapot but put it down to it being faulty - perhaps people are putting too much faith in them?  I have had decent results from my telemetry sensors.

Edited By Martin Harris on 02/11/2019 10:57:43

[SIMON CRAGG]

Martin

E.calc says 78500rpm, and I am getting 7800rpm.

This is the only thing we agree on!

e.calc+ 78.3a, 1631.8w

me: 59.0a, 1415.0w

I have tried calibrating the throttle, changing the timing etc. but I always get more or less the figures above!

All very strange.

Only thing left is to change from Deans to EC5.

Oh................and FLY it!

[Simon Chaddock]

Simon

In the IC world, where RPM and static thrust is about all you can really measure, pulling the predicted revs for a given prop is good enough!

Don't chase Watts, its only a power input not a power output, just fly it.

[Frank Skilbeck]

One thing to be aware of is how quickly you open the throttle, all electric motors will take more current as the accelerate upto speed, which is why you need to check the readings once the motor is running at speed.

One puzzle on your wattmeter is that on the 2nd run, you get the less watts at the same rpm and the volts are slightly higher, if the volts are slightly higher then you'd expect a slight increase in rpm (and amps and power),

If you do the V=IR calc on both readings then the system resistance has gone from 0.32 ohms to 0.396 ohms, less than 0.08 ohms, which could be explained by a different ESC, connector temp etc.

Also on a 6s system you get a fair spark when you connect the battery which eats away at the connector, I switch over to anti-spark connectors at 6s

But I concur with Simon just fly it, if you want some real measurements then use some telemetry, the Unisens E sensors are great for this.

Edited By Frank Skilbeck on 02/11/2019 11:58:28

[SIMON CRAGG]

Thats the plan, fly, fly, and more flying. I know its a safe and reasonably powerful system that is below the spec. of the components.I have got a bit bogged down by facts and figures, so its time to get up the field at get on with it.

[Peter Jenkins]

In my experience, the biggest impact on performance in an electric setup is the make of prop. I regret to say I've never heard of a TGS prop. If you use an APC Electric prop that is likely to give the best results unless you go up to more expensive carbon props. But be prepared for significant changes in power with different makes of prop with the same diameter/pitch.

[Peter Beeney]

I’ve always tried to figure out this stuff using my own brand of KISS, Keeping It Strictly Simple, so I then have to generally judge these result by what I actually see in the air.

I’ve never used a wattmeter but I do have some useful kit, mostly which is now like me, getting very old; although unlike me it still seems to be quite reliable…

These are respectively a voltmeter, a clip on power meter and a contact thermometer. Plus a slightly more recent micro tacho.

I always check the unloaded speed first, this gives me a true known datum point, the rpm per volt; mostly the figure is as stated but not always, one motor recently in a ARTF with no information printed on it at all, had a kV of 600. The model’s specification sheet gave it as 850. Then bolt on a prop, that might be one that I just think is suitable or perhaps more likely one that’s recommended in the manufacture’s spec., (but it will be an ACP i/c prop), and check the revs again. As a starting point the figure would want to be at least within 20% or better of the unloaded revs. At the same time I’m definitely checking the current flow, and if I’m really pedantic about it the battery voltage as well; but really I’m only concerned about the current, I know roughly what the voltage will be and I’m not particularly interested in the watts anyway. Then maybe some experiments with other props, generally until I get the motor turning as close to the original unloaded revs per minute as practically possible. I’d also want to give it a fairly long flat out run too, and check that nothing is getting overly hot; although the current passing through should now be well within limits; the ESC temp check at around seven eighths/three quarters throttle.

Finally an eyeball flying test to see how the model performs compared with what I’m expecting. Of course, much of this is going to depend on the model and how it’s expected to perform, such as the comparison between a Cub and Corsair for instance. But I’m always up for a bit of wellie and I’m sure that a reserve of power is never going to be a bad thing, there must have been times when pilots of both models and full size had been very appreciative of a few extra hosses…

Probably a bit of a basic and primitive method by modern standards but I’ve personally never had any problems. And, indeed, because of the increasing popularity of electric models over the last few years I think the electrical knowledge generally gained by aeromodellers has increased by leaps and bounds; which can only ever be a good thing.

PB