Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length.

[TimCo]

I have recently rejuvenated a 40-year-old "Moonglow" model with its original Marx Decaperm Special (direct-drive, not geared), by equipping it with 2 x 2S 5000mAH LiPos in parallel and a Tio Marine 15 (i.e. with LiPo cut-off) controller. Unfortunately, with LiPo-mode selected in the ESC, the motor occasionally stutters, like a "misfire".
At Mtroniks suggestion, I have added capacitance at the supply input to the controller and beefed-up the wiring to what I consider to be an extreme degree, but the results are still no good. Three videos from the model boat are now on YouTube to illustrate:

1.   Original 14awg Y-adapter for the two large batteries, with single 30A auto fuse in in-line fuseholder, between battery and ESC, and with 10,000uF capacitor connected to back-to-back Deans connectors, connected to the male Deans plug that is soldered to the ESC’s red and black wires.   http://www.youtube.com/watch?v=A_R9sqKAo-g
2.   12 AWG Y-adapter with 25A blade auto fuses, one per branch, connected to 2 x 22,000uF 16V capacitors soldered directly across the ESC’s red/black input wires. The fuses were also soldered directly to the thick copper buss-bars connecting the two capacitors. The only detachable connectors were the Deans plugs for the batteries. At 10A motor current, the volt-drop in the leads has been reduced from 93mV, total, both leads, to around 20mV.  http://www.youtube.com/watch?v=Ru7YG1f0MjM  . This additional assembly adds a rather undesirable 60g to the model – more than the weight of the ESC.
3.   As 2, above, but with the ESC’s LiPo battery protection turned off, i.e. NiCad mode selected. You can just see in the videos a partial reflection of the ESC’s LEDs in the front windscreen, LHS: in this one, the blue LED is not showing.  http://www.youtube.com/watch?v=2r69SuPs_lw  . Here, the only unwanted speed fluctuations are caused by debris and weed catching the prop! No “glitches" at all! Total running time was 33 minutes, to avoid over-discharge of the batteries which might normally be expected to run for around 1 hour.

Before it is suggested that I remove the supply fuses in order to reduce the voltage drop a tiny bit more, I would like to reiterate that with a single 5000mAH 2S LiPo connected directly to the ESC, the glitching problem was still evident, at currents of 10A or less, with this motor.  It seems to me that, in LiPo-mode, the ESC is responding spuriously to small and, particularly, very brief, dips in the supply voltage caused by motor “noise”. With this in mind, I have tried adding 0.1mH in series with the motor, but got no improvement.

I would like to ask whether anyone has experience of this ESC, combined with LiPo batteries of only two cells? In LiPo mode (blue LED on)? With what motor?

[dodgy geezer]

No help to you, but I also have the same problem with small ESCs running in EeZeBilts with 4.8v NiMh batteries. I'm running on the limit of the voltage everything will operate at, and the ESCs always cut out intermittently in the same way...

[Subculture]

Sure it's not radio issues? Have you tried it with another receiver and/or transmitter?

The voltage certainly should dip on the lipos, you're not exactly pushing them in that boat are you.

[TimCo]

It's not a radio issue. The performance is the same with three different Tx/Rx sets on 3 different frequency bands, although the videos cited are all with a 2.4GHz set. In any case, the third video was included to demonstrate that the ESC performs OK so long as it is in NiCad-mode, i.e. LiPo mode de-selected (blue LED off).

As regards speed variations with small ESCs with 4.8V NiMH, perhaps Mtroniks' recommendation to add 10,000uF across the battery leads, as close to the ESC as possible, might work, if you can afford the weight increase. For example, I have a 10,000uF 25V electrolytic here that weighs 12g.

[Subculture]

Okay, so it sounds like the Mtroniks lipo monitor is a bit too sensitive. I guess switching it off defeats the object of using that model of controller in the first place.

[dodgy geezer]

As regards speed variations with small ESCs with 4.8V NiMH, perhaps Mtroniks' recommendation to add 10,000uF across the battery leads, as close to the ESC as possible, might work, if you can afford the weight increase. For example, I have a 10,000uF 25V electrolytic here that weighs 12g.

Interesting idea, and worth trying - though the problem with the small EeZeBilts tends to be space as well as weight..

[TimCo]

I have challenged Mtroniks to modify their LiPo cut-off, and await their reply. In the meantime, I shall run in NiCad mode, at least, when the cameras are on, and keep an eye on the clock. A separate LiPo monitor display/alarm is one possibility.

Today, I discovered that I am not the only one to notice this problem - see  http://www.tcrcboaters.org/phpBB2/viewtopic.php?p=775&sid=0f9f6490fdb5f7380f49f5cfc6cc2f66  where a similar thing has been noted on a Tio Marine 30.

[TimCo]

Finally, the 1970's Decaperm failed. Although repairable (armature winding soldered joint broken at commutator), the armature was never well-balanced and the plain bearings were wearing out, so I went brushless! A small RCS 2845-size water cooled motor, 2100kv, 40A max., with a Sea King 60A ESC was fitted. This combination is around 150g lighter than before and can deliver around twice the power (250W at top speed).
The new combination is also more energy-efficient at moderate speeds: for example, on 2 May 2013, the model was taken to two venues and run for 50min at the first and 35min at the second on the same batteries (2S2P Li-Po, 10Ah total) and around half the charge was left at the end (i.e. discharged only to 3.75V/cell). Videos from this day, from both inside and outside of the model, can be seen at http://www.youtube.com/watch?v=Hdq276KMCg8 (Rudyard Lake) and http://www.youtube.com/watch?v=MglUI87mX4w (Caldon Canal at Denford).

[malcolmfrary]

Could have been interference generated by the failing motor.  It doesn't have to go via the ether - odd voltage spikes go through wire and components equally well.  Suppression does just that - it does not eliminate interference, it just reduces it in most cases to a level where it is not a nuisance.  If the source level is too high, it still gets through.  2.4GHz is widely held to be "proof" against RF interference - it isn't, its just that not many ways of generating interfering signals have been found yet.  They are just as prone as anything else to spurious spikes getting into the power, and that includes the regulated 5 volts fot the control end.  Thats not to say that the ESC isnt over sensitive.