Weird glitches underway - troubleshooting suggestions

[He Who Shall Not Be Named]

John,

That's a good test protocol for total current draw off the BEC, can't argue with the methodology of the test either -- very much 'real world' circumstances. And great follow-through with your argument that you don't think (in your case anyway) that the observed problems with the ESC was not an overloaded BEC; you actually put your position to a demanding (the loads presented to the servos) test and chronicled the result. Well done, sir.

.85 Ampere's is well within the MTroniks BEC's safe operating range.

And your above posting is just what this thread is about -- the interaction of producer and customer. Yup, it's in the right thread, John.

David,

[toppack]

With an intermittent problem like that, I think the only way you will know for sure if it's the ESC or not, is to try another ESC, (or maybe different receiver if you have one handy). I know it's not easy to change out the ESC, but that's probably the best way to isolate the problem.

[Subculture]

The very best BEC's are switchmode devices. Very wide voltage range, high current capability easy to find a model that'll supply 3A, and very efficient- they run very cool, unlike linear devices which are decidedly inefficient, especially at high voltage drops.

Unfortunately, the only company I've seen implementing switchmode BEC's on their ESC's are Castle Creations, and these are brushless ESC's.

If you want a standalone switchmode BEC, then Dimension engineering come recommended.

http://www.dimensionengineering.com/

[VanguardUK]

Yes tried the Tx off / on - no change.
The BEC is on board the M-Tronics ESC is a 1.5 amp one.
I use 7.2V battery. The BEC may not be an issue as the leveller and fail safe still worked and these need 5 volts from the RX, and the RX gets the 5V from the BEC. All other servos etc were working too.

By the way- what ESC where you running that had the BEC issue?
J

Mtroniks - to be honest all my servos worked as well - it was only the motor that would fail. - this would go along with the BEC failure theory, which is a worry as I've just installed one in my Vanguard for it's test run tomorrow!

[Slats]

Mtroniks - to be honest all my servos worked as well - it was only the motor that would fail. - this would go along with the BEC failure theory, which is a worry as I've just installed one in my Vanguard for it's test run tomorrow!

I am not following why all the servos still working would be a BEC problem.
What was the battery condition like at the point where the motor stopped working?

Please report back in here with the details of the Vangaurd run.

I am going to be trialling as per Rick's suggestion a different brand ESC. The one I'll use has only a 1amp BEC - its an Electronize - been using these in multichannel heavy targets for at least 15 years, never had any problems.
I won't get to the lake though for a couple of weeks.

Best
John

[Slats]

Guys,

It is the BEC!!!

Re tested today and found big problems with the auto leveller (not a KMC one).
Anyway hit 1.8amps!

Also been checking out Heli / Parkflyer forums.

The bottom line I get from these other disciplines is BECs on ESCs almost completely use linear BEC technology - by products are heat and or ratings of amperage that vary with input voltage. Also the stated amp rating is usually a peak figure.

These other disciplines all say the same thing - rule of thumb - more than 2 servos - separate BEC and of the switch mode variety.

I have just bought 2 of these:

http://www.euromodels.com.au/shop/index.php?main_page=product_info&products_id=812

I hope this will beat the problem for good!

Thanks for your input and patience.
Best
John

[Subculture]

I have one of those. They work very well.

[toppack]

Re tested today and found big problems with the auto leveller (not a KMC one).
Anyway hit 1.8amps!

Curious, Is that 1.8 A a combination of current draw of the Leveler-circuit and the Servo connected to it?
If so, the servo may be drawing excessive current, not the leveler circuit?

[He Who Shall Not Be Named]

Well done, John. You've tracked this problem to ground and have, through your careful research and actual testing, helped identify the likely cause of glitching in systems employing more than two servos (and a host of other devices) sucking energy off the receiver bus serviced by only the ESC's BEC.

Agreed, we need to either procure or have manufactured discrete switch-type BEC's to power up the many servos and other devices and recommend this devices inclusion in all our SubDriver's.

As I see it the only modification of the current SD hook-up arrangement would be to remove the ESC's red wire from the receiver plug; hard-wire the switch-type BEC to the battery plug, and output the higher capacity switch-type BEC into the receiver 'battery' port.

Very good work, John -- you've moved the ball forward here.

David,

[KevinMc]

A word of caution on switch-type BEC circuits- Compared to a linear regulator, swich-mode regulators are very "noisy" (electrically speaking) and have the potential to reduce your receiver's range. I'm not saying they're unsuitable, just do a thorough range check when installing one in your system before "hitting the deep".

KMc

[toppack]

To supplement the ESC's BEC (which is a 5.volt regulator), I use Voltage-regulator chips such as the LM7805, in TO-220 type package. Which is 5.volts at 1.Amp and will handle several high-torque servos with no problems, if mounted on a small heat-sink of some kind (such as an aluminum servo-mounting-tray or bracket). Will handle 2.Amps with a large or well ventilated heat-sink.
Simple 3 wire hookup: In, Out & ground (the mounting tab is ground also)
Cost about $2. each.

Here's link to Data-sheet for LM7800 series Regulators:
http://www.fairchildsemi.com/ds/LM/LM7805.pdf

[mtroniks]

Hi John.

Like the 'targets' term, explanation appreciated.

I still wonder about the BEC...servo current usage can be very 'spiky', a test meter would tend to show an average value, and not show spikes the perhaps could upset ESC operation.

I'd be tempted to try a capacitor connected across + & - on a spare channel of the receiver, something like 1000uF, 16v (observing polarity!). This will act as a reservoir, hopefully providing energy to satisfy those current spikes when required.

We must get around to doing a nice stand-alone high current BEC unit...

It's possible I'm barking up the wrong tree John, but this is my gut feeling on the subject.

[Slats]

Ok appreciate all that and exactly what Steve (Mtroniks) said is the best explanation of my latter testing. The amp meter showing the intial .87 Amp was indeed likely and average (or mode - most populous current) - -average or mode in the fact it stayed on so long I could photograph it, the spikes I saw around 1.8 amps are what I believe to be the problem. Sorry- I should explain, the leveller issue I mentioned; it started behaving badly (erractic interacting with throttle controls) this week. Even with the leveller out of the RX and just the servos in spikes were observed above 1.5 amps

Kevin - I am now very concerned about the noise issue of the switch BEC - is range the only anticipated problem? Can the BEC be in some way isolated from this noise? What do you recommend?

Rick- will check it out what you suggested - and I believe this is the sort Skip Assay used to produce.

Thanks everyone.
J

[Slats]

I have one of those. They work very well.

Andy,

re Kevin's comments about switch types being more electrically noisy - have you experienced any interferrence or randge problems using this BEC?

J

[Subculture]

No problems to date.