HobbyKing.com New Products Flash Sale
Reply
Thread Tools
Old Mar 09, 2009, 05:19 PM
Registered User
Joined Feb 2009
3 Posts
Question
TF8000 advice

Hello all,

I'm in the process of bench testing a TF8000 to be used somewhat unconventionally as mass trim for a VTOL model as a controls project in conjunction with another propulsion system. The short story is, our bench tests don't reflect the data gleaned from others here, and from the test data of ductedfans, the vendor. We have some ideas, but I'd be curious to hear the opinions of those with more experience than I.

Our setup is the TF8000, Lehner 3060/13, Schulze 40.160H, with 14S Tanic 4500maH/30-35C. Data is being monitored on an Eagle Tree v3 150A. For bench testing (from which below data is derived), we're actually using 4S deep-cycle marine lead acids, off the chargers at 52V, as eventually we'll be testing a number of these in parallel.

We calibrated the controller on startup to 2000us/1000us full/neutral PPM, exactly, from a microcontroller feeding the signal (as opposed to servo tester). We then worked our way incrementally up to 65% of full throttle (1650us), ramped to for 10 seconds, then held for 10 seconds. The problem is, we're already seeing 160A+, with the batteries dropping to 42V under load, so still 6.7kW. However, we're only reading 85N or so through a load cell to a DIO. Ductedfans has provided us with a bench test point indicating 133N@6.37kW, admittedly probably rosy, but we're not even close.

Here's our list of possibilities:

1.) Voltage dropping so low sends fan off design. With LiPos you'd probably see 5-10V higher.
2.) Ablative friction of rotor on the housing. There was some mild scoring, but this is unlikely and shouldn't be sustained.
3.) Downstream wake effects from having a large bundle of wires interrupting the exit flow, which we haven't minimized/organized yet.
4.) Error in Eagle Tree v3 logger. However, max possible error on Allegro current sensor stated at 5%.
5.) Error in load cell. We've calibrated extensively, but might verify on a scale rig.

None of these seemed to be a silver bullet. Any thoughts?

Current draw was quadratic with a linear ramp, trending upwards (See pic). RPM sensor pretty much freaked out (the sensor was of the back emf voltage tap sort). At high voltages (48V nominal), even after freshly calibrating for each new test, the motor didn't come on until 1150us or so, with a mighty kick. At 12V, it was quite gentle and came on pretty much straightaway, ramping from 1000us. No load tests showed RPM was logarithmic with throttle input. Anyone have any insight on the Schulze 40.160H and what might be happening here?

Many thanks

[IMG]48V_65perc_throttle.jpg[/IMG]

Edit: as an aside to any who might have/will use the TF8000 in the future, the collet for the rotor mount is tapped to M6. The screw referenced in the manual (auf Deutsch), and the screw provided from the factory is an M5. We didn't catch this the first time and damaged our rotor.
Ithacacian1 is offline Find More Posts by Ithacacian1
Last edited by Ithacacian1; Mar 09, 2009 at 05:27 PM.
Reply With Quote
Sign up now
to remove ads between posts
Old Mar 09, 2009, 09:24 PM
Registered User
Joined Jun 2002
860 Posts
Ithacacian -

Just a couple of very general thoughts/questions;

Are you sure the Lehner motor is the correct motor for this fan? Reason for this question is that Lehner has a non-standard way of specifying Kv. This might have led you to using the wrong motor.

Nothing wrong with using a microcontroller to run the ESC, but is the microcontroller shielded from the motor/ESC and associated wires?

That is where I would lOOk.

Ron
Ron Pogue is offline Find More Posts by Ron Pogue
Reply With Quote
Old Mar 10, 2009, 01:53 AM
Registered User
Joined Feb 2009
3 Posts
Ron,

Thank you for responding. Our motor was that recommended to us and sold as a package with the tf8000, by our vendor ductedfans. Wired in delta, we've found specifications listing the 3060/13 at 604Kv, translating to a full throttle RPM under 30,000 RPM with load, which is pretty close to what the tf8000 should be operating at. Its tough to tell whether our motor came close to that operating speed on account of the shoddy rpm data.

The microcontroller is powered off of USB, completely separate from the motor power source. I've looked at the signal with a logic analyzer and its fine.

I see how the motor might be an issue, but I've seen people on here using the 3060/10 with 12S, which is 785Kv, where 785*(12/14) = 670. Not off by a mile. Could you elaborate on the Kv issue? I assume you're not just referring to delta vs wye.

I appreciate the help.
Ithacacian1 is offline Find More Posts by Ithacacian1
Reply With Quote
Old Mar 10, 2009, 03:09 AM
Registered User
Joined Jun 2002
860 Posts
Quote:
Originally Posted by Ithacacian1
Ron,

Thank you for responding. Our motor was that recommended to us and sold as a package with the tf8000, by our vendor ductedfans. Wired in delta, we've found specifications listing the 3060/13 at 604Kv, translating to a full throttle RPM under 30,000 RPM with load, which is pretty close to what the tf8000 should be operating at. Its tough to tell whether our motor came close to that operating speed on account of the shoddy rpm data.

The microcontroller is powered off of USB, completely separate from the motor power source. I've looked at the signal with a logic analyzer and its fine.

I see how the motor might be an issue, but I've seen people on here using the 3060/10 with 12S, which is 785Kv, where 785*(12/14) = 670. Not off by a mile. Could you elaborate on the Kv issue? I assume you're not just referring to delta vs wye.

I appreciate the help.

Hi -

Kv refers to the rpm per volt, of a free running, unloaded motor. A motor loaded by fan typically will turn 80 - 90% of the unloaded figure.

Wish I had answers for you, but I have no personal experience with the TF8000 nor the Lehner motor. It is fairly new to the EDF scene.

About the best a guy could do is use the "search this forum" button in the upper right-hand corner of the front page of this forum. A search using the terms lehner 3060 revealed quite a bit of useful info.

Ron
Ron Pogue is offline Find More Posts by Ron Pogue
Reply With Quote
Old Mar 10, 2009, 08:11 AM
Registered User
Lboogie's Avatar
Sebastian Fla
Joined Nov 2005
600 Posts
http://www.lehner-motoren.com/

Regards,
Leonard
Lboogie is offline Find More Posts by Lboogie
Reply With Quote
Reply


Thread Tools