EDF units

[Konrad]

I just saw these on your site.

I really like that the rotor (motor) and impeller (fan) are integral. Reminds me of the Huffington fans of old. Now I hope these new fans have are much higher copper fill.
Love the triple bearing support none of this cantilevered design stuff!

Looking at the 90MM unit I'm a bit puzzled why a 6 cell wind. At this size of model I'm looking for 8 cell minimum. (Most quality low voltage ESC like the Castle Creation Edge can go this high) But most HV ESC will allow you to run 10 to 12 cells. Are there any plans to bring in any lower KV (higher cell set up) 90mm fans?

Can we get some more photos (Spec's) of the motor

EPF Mercury II 90mm EDF 12 Blade 1900KV (Wide Inlet)

From EPF Hobby's 3 years of research and development comes the Mercury II series of Electric Ducted Fans. Entirely constructed from CNC machined, high tensile strength aluminum alloy and static and dynamically balanced right at the factory, they'll amaze you with their turbine like sound and...

alofthobbies.com

6 Cells to my mind is about right for the 74mm class.

All the best,
Konrad

 

[jvaliensi]

The internet says the 6S 90mm will pull a maximum of 145A and produce 4.8kg of thrust, at peak.
My JP Fan 90mm 6S does 130A and 3.8kg thrust.
The big inlet may help. I don't believe it will do 4.8kg, but I'd be happy w/ 4kg.
I like the wires out the front. That is elegant.
I may try one.

 

[Konrad]

The JP Fan, that's what this reminds me of (Inlet guide vanes). Unlike the JP fan it looks like the impeller is supported on both sides. (You might already know that I don't like cantilevered designs).

Not having actually run one of these or even held one (Yet), concern with the design is the low stator count (4).

I'm wondering what is the RPM limit. I See JP-fan lists their 90mm as having an 88K rpm limit. On Aloft's web page I see where the fan in dynamically balance to 60K rpm.

If on six cells this draws 145 amps that means one needs a 180 amp ESC. At these power levels a high voltage ESC is often looking like an economic alternative. Also at 145 amps the cycle life of the lipo is drastically lower. A 12 cell set up drawing 70 amps looks far more practical.

To my way of thinking the 6 cell 90mm is a result of the low cost MOSFET limit of around 25V. (This is reminding me of the old 7 cell round cell trap, where the chargers couldn't charge batteries above the 12 volt source voltage) One might be able to make a case for the 6cell 90mm use in the low cost Freewing foam junk. But if actually looking to upgrade I'd think most would leave the 6 cell set up far behind. And if setting up a higher quality jet than a foamy one wouldn't even consider a 6 cell 90mm set up.

Again I feel that in the 70mm to 80mm fan the 6 cell set up is practical. In the 90mm class the stress on the battery (more than 120 amps) has far too high a cost in low battery cycle life.

 

[jvaliensi]

Don't write off 6S 90mm. I think I'll have good success with my 6S JP 90mm. Certainly Sebart did, as it was the factory set up for the mini Avanti S.
However, I plan to build another w/ a 12S JP90mm fan. And all your points are valid.
My batteries are 3-years old and still putting out the amps.
I just lost two packs - one because I left it plugged in a RC car with a very small drain, the battery went to 0V. Another 6s pack just dropped a cell, but it was at least 3-years old. I've actual had a 4 or 5 batteries loose a cell and become garbage over the past 4 years. It's fun when it happens during a flight.

 

[Konrad]

Remember the "factory" is trying to meet market expectation. Far too many folks are using 6 cells in 90mm EDF because of the limitation in the Low Voltage ESC. At some point the cost of a low voltage ultra high current ESC exceeds the high voltage low current ESC for the same power.

I think 120 amp is the practial limit for most sport application for the lipo battery.

Time is a factor in lipo life, the 2 year point is when folks can start to see a drop in battery performance. But what we really should be looking at is charge cycles when talking about battery life.

 

[Wayne]

Hey guys, these are super nice fans. We were test flying the 74mm a while back and we very impressed, so here we are now. Sounds as good as she looks and the performance seems to be what is stated. I have seen reports of people running higher cell counts with good results, but we have not tested this.

Here was our test bed:

We thought this might be a little underpowered, but the fan made so much thrust it was a non-issue. This was a prototype plane with a prototype motor with a prototype ESC. Sadly the ESC let us down by getting so hot that the darn thing desoldered all of the mosfets after a few minutes of zooming around. Seems the heatsink was not making contact with the mosfets. Plane came down hard, snapped the wing joiners, and dented the front ring of the motor. Everything but the ESC will fly again.

I will tell you these things look so good, they make great desk art! A shame to bury them in a model. I have no idea how they make these things for the price they charge.

 

[Konrad]

... Everything but the ESC will fly again. ...

Even the wing joiner?

Was theESC a new test product or were you testing a line you carried such as Power Up? Can you share with use the brand and model # of the ESC that failed?

 

[Nigel]

I believe it was Neuron ESC, but I was surprised it lasted that long - it was an 80 amp ESC easily pushing 120+ amps, more realistically 140 during the vertical punch outs and momentum changes. Considering how it performed even with the known spacing defect for the heat sink, I'd say it did pretty well. I think Chris dumped a 120 amp ESC in there or so for the re-maiden.

 

[Konrad]

Ok, we can't blame to OEM for the mis-application of their product.

But if the MOSFET weren't in contact with the heat sink this indicates that FrSky has a manufacturing issue and should be recalling these ESC. Reputable firms like Castle Creation (CC) do recall products that are found to have manufacturing issue such as as these. You might have noticed that CC has a manufacturing tag on their products for the last 10 or more years for exactly this purpose.

 

[Nigel]

It's a known defect from an older batch I believe. Wayne has already contacted Frsky about it and we check the ESCs that we do ship out to ensure that they are in contact. I doubt FrSky will recall them, most likely choosing to give out a warning/fix future batches.

 

[Konrad]

A service bulletin is another way to address issues like this. Can the defect be seen without disassembly of the ESC? Does the discrepancy vary between unit to unit? Or is the problem a batch issue? Can the issue be fixed by the end user with "normal" Hobby tools (soldering iron and screw driver)? Normally Aloft has a notice on their sales page of known issues with products.

FrSky Neuron 80 Amp ESC

Features:Smart Port enabled and telemetry data transmission supportedTelemetry data for ESC: Voltage, Current, RPM, Power Consumption, TemperatureTelemetry data for SBEC: Output Voltage, Current.High performance 32-bit micro-processorHigh-precision current sensor ACS781KLRTR-150U-T(Resolution...

alofthobbies.com

I agree FrSky most likely won't recall the product. Has FrSky ever issued a hardware warning/service bulletin?

All the best,
Konrad

P.S.
I'd like to see a place on this forum where we can discuss electrical components and their issues, away from electrical aircraft.

 

[Nigel]

I don't think there was a point on putting a warning on the product page - especially when the failure was so few and far between, compounded with the fact we quality check them before they ship. I do believe that you can clearly see if the MOSFETs aren't making contact - I tried to find a better picture, but this is the best I could find. It's not an enclosed/cased ESC, so it's blindingly obvious and easy to spot.

As for can it be fixed - I'm not sure. I forget whether it was a spacing issue with the standoffs or simply no thermal paste. It wouldn't be the easiest thing to fix, but I'm sure anyone with a little experience under their belt could manage.

 

[Konrad]

Not so blindingly obvious and easy to spot, as we have evidence that at least someone at Aloft missed it once.

Now if we the customer are are made aware of the issue, then yes it looks like an easy fatal flaw to spot. And from here it looks to be an easy thing repair. just goes to support my view that the customer experience has fallen as a priority at FrSky. Any QA guy or QA program should have caught this, like you said; it is blindingly obvious and easy to spot.

All the best,
Konrad

 

[Nigel]

Very easy to miss when you don't even know it was there in the first place. ? I couldn't agree more on FrSky's interesting quality control management. Relying on distributors and even customers to find flaws is quite negligent. But at the end of the day, at least we know about it before it could cause more serious damage than a test platform.