Jump to content

Catastrophe!


Jonathan M
 Share

Recommended Posts

On zero throttle, move one stick at a time and observe the readout for that servo on the Watt Meter (do each aileron servo separately by unplugging first one then the other), then repeat at 50% and 100% throttle - the difference in amps drawn for each servo should I assume be the same.

Link to comment
Share on other sites

I am not sure about why you think the control surfaces were in spin mode. The force or the crash could have left them anywhere. My money is also on power . We must not think for one minute that the ratings quoted on the power train components are safe. I allow at least 20% headroom. They are not necessarily describing the items wrong. But just not the way we expect them to mean. Also interference or make of radio should always be put very last as the possible cause of a crashing if you want to be getting near the real reason. As one who has been there I do feel for you it’s never nice to see it happen to anyone I hope you have not scrapped the model though you or someone else will get the inclination to repair it sometime

Link to comment
Share on other sites

Cheers Gangster - I agree, the power issue seem to be the most likely one... allowing for the fact that the loop problem took place at full-throttle and the spin problem at no throttle...?

The control surfaces were all in the same post-crash positions exactly as I'd first put them into to enter the spin: up elevator, left rudder and finally ailerons (see photos at top of first page of surviving surfaces and servo-arm positions). I agree that one would normally expect much more random deflections after such a hard impact, but - as I say - they were all exactly where I'd originally had them before loss of control.

Interestingly, I was just looking at my own Gangster 63 build-thread (started two years ago before I got distracted by other things but now thinking about making this year) and came across these contributions:

Posted by Phil Francis on 24/04/2017 19:37:45:

Don't use New power Servo's unless you want a world of hurt. Tried a couple a while ago they don't centre well, 1/4" either side of centre.

Posted by Percy Verance on 24/04/2017 20:48:13:

My experience of New Power servos isn't particularly good either. Those I bought were not good at centering well. I shan't buy anymore...........

Futaba 3001's or HiTec equivalents for me all the way....... More often than not, budget servos will eventually come back and bite you on the proverbial.

Not saying the NP servos were definitely connected to the Wot4's catastrophe. I won't know anything until I've run the tests, but the common denominator in the two instances of loss of response - the loop and the spin - both involved the elevator!

They were certainly very noisy and didn't seem particularly fast or responsive - I kept having to fiddle with rates and expo, trying to get the model to respond smoothly to stick movements, rather than too little and then suddenly too much that I kept experiencing!  They certainly didn't centre properly.  So I'm definitely going to fit one of the established brands on the new Wot4 (very happy with the Futubas on the Acrowot and the Hitecs on my Middle Phase).

Edited By Jonathan M on 25/01/2019 17:41:54

Link to comment
Share on other sites

Jon,

I think if this happened to me I would be trying to establish in the first instance why the throttle control behaved as it did. If I read this correctly you fully closed the throttle (stick) on two separate occasions from full throttle but the motor continued to run at full speed. I would now be thinking that for whatever reason, at least in the first instance anyway, the throttle stick is not working properly. I have to say that for me that would be a really major incident; from the safety angle just for a start...

It seems to me that the processor in the ESC will have to receive a valid pulse width signal on the throttle channel at all times to operate; if it did not it wouldn’t run the motor at all. This signal has to come from the receiver which in turn gets it from the transmitter, it’s difficult to see how any random component or any fault in this little power chain could create and maintain such a signal on it’s own. Also very significantly the transmitter’s throttle cut-off switch, which I consider just holds the throttle channel pulse chain at the fully closed position anyway, effectively closed the throttle. I don’t think it could have done that if the problem is not in the transmitter and related to the throttle stick.

And then almost simultaneously another separate fault appears to cause all the servos to stop working… …is that now perhaps stretching credibility to some extent, I don't know, but I think I’d certainly want to give the transmitter a very close coat of looking at to make sure it was 100% ok. In my experience, in a now long ago working life, I found that little occasional random faults and glitches on electrical kit generally, if every effort to crack them was not made at the time, 9 times out of 10 will sooner or later occur again.

Just another niggling thought, it seems as though the throttle channel doesn’t really like faithfully obeying full stick movements sometimes. When you initiate a flick roll where do the sticks go? I’d say generally in the corners at full stretch. So now you may be in front of me and saying perhaps the other sticks are also following suite… Sounds like an extreme long shot, bordering on the limits of visibility maybe, but pretty easy to prove one way or the other.

Good luck…

PB

Link to comment
Share on other sites

Jon regarding the throttle, have you tried looking at the "Monitor" screen on your Spektrum Tx and wiggled the sticks - slowly, fast, suddenly, any which way - just to check the sliders on the display continue to show the stick positions correctly and smoothly with no jumping? This will show if the sticks are behaving correctly without having to connect any servos etc.

A friend's Dx6i Tx was misbehaving (I cant remember which channel) and I solved it using electrical cleaning spray in the relevant potentiometer. A new pot may be better if a fault is found. As Peter says above the throttle cut just tells the Tx to send fully down (125%) to the Rx by-passing the stick. If the stick potentiometer was faulty this could explain the power problem and why it was initially cured using the cut button.

 

Edited By GrahamWh on 25/01/2019 21:17:11

Link to comment
Share on other sites

Posted by eflightray on 25/01/2019 18:50:18:

Catastrophe ?

Many moons ago, the group I flew with had a trophy for the best (most stupid) crash.

It was made from cardboard, and depicted a 'cat's-ass' trophy. wink

Much cherished if won. All clubs should have one.

.

My cycle club's award for the 'crash of the year' is 'The Flat Cat Award'. It was instigated in 1990 when I was downed (very seriously as it turned out) by a cat deciding that the shortest way across the road was through my front wheel. I think my wife and I are the only married couple to 'win' it (2 separate events).

Geoff

Link to comment
Share on other sites

That's rotten luck, some years ago I managed to flatten my rx battery on entry to a spin. The plane spiralled down from on high and came to earth some distance downwind. Happily the earth was so wet and muddy that I got away with just a broken prop for my troubles. One question though, what's a three point spin?
Link to comment
Share on other sites

Posted by eflightray on 25/01/2019 18:50:18:

Catastrophe ?

Many moons ago, the group I flew with had a trophy for the best (most stupid) crash.

It was made from cardboard, and depicted a 'cat's-ass' trophy. wink

Much cherished if won. All clubs should have one.

.

We've had one for years. A Merco 61 (I think) that looks like it's been run over by a tank, mounted on a tall wooden plinth. Very coveted trophy with many of our club's top flyers' (!) names on it. TBH, I think I can do without the accolade.wink

Link to comment
Share on other sites

The fact that the control surfaces appear to be in the position required to enter the spin, and that the aeroplane wouldn;t come out of the spin, argues for a loss of power to the receiver, causing it to become non-responsive. I'd be looking at testing the current draw on those 4 servos all being run to their extremes at once and the stability of the current supplied to the Spektrum 6210 receiver under those conditions.

The model was described as being almost brand new - approximately how many flights on the board before the total loss?

Link to comment
Share on other sites

Peter, Graham, Leccy, et al

Now back home from 24hrs away. Just checked the TX monitor which shows throttle (and other travels) all working fine. Also just checked the throttle-stick on my IC Acrowot's throttle-servo, which works fine. This leads me to conclude that I might provisionally rule out the TX end of things?

While Peter is right in what he observes about the throttle cutting via the switch when bringing the stick down didn't immediately work being related to a potential problem with the TX, it is also feasible - if it was indeed a poor connection disconnecting aloft at extreme-G or under other duress - that on both loop occasions it was the removal of such duress by the point that the model had entered the vertical dive part of the loop that coincided with my hitting the switch that reset the throttle connection, rather than the TX switch what did it? Conversely there would have been no removal of extreme conditions during the spin (thus none of the controls responded), so this also points to a problem in the air rather than on the ground.

Next up - tomorrow - is to remove the whole caboodle of servos, RX, ESC and motor from the box of wreckage parts and rearrange them statically on the bench powered by a fresh LiPo. The electronics all look physically fine and hopefully the motor isn't terminally damaged and will run mounted on a block of wood held by a vice. I'll do all this with the watt-meter connected up inline to carry out a full set of tests.

I won't be able to statically test for extreme or negative-G (vis the conditions at the top of the loop and in the spin), but will be able to wiggle each connection in turn, etc, and will report back afterwards.

The other thing - Leccy's question - is that the model had flown from new for a total of only four sessions (four Lipos per session so about sixteen flights in all) but most of these had been relatively sedate initial testing and trimming flights, and it was only on the fateful day itself that I was beginning to push the model harder.

Jon

PS: I'll reiterate this: I now fully recognise that it was an error to not ground the model immediately after the very first sign of trouble and this is a lesson I've fully taken on board for the future.

Edited By Jonathan M on 26/01/2019 20:46:14

Link to comment
Share on other sites

Martin, the Confetti-Incident took place last Thursday afternoon, so not particularly cold, maybe a rather pleasant 10°C at lunchtime with no real wind-chill, and I easily didn't need to wear gloves.

Good point Don! But its worth giving it a go.

One thing though, it isn't my imagination: those NP servos were so very s-l-o-w. When I checked the TX earlier this eve on the Acrowot the Futabas were lightening-fast by comparison. And I've just also tested the TX for full working on my 1m DLG which is powered by a 1s 300mAh Lipo and its two KST08 and two EMAX 8g servos responded similarly fast. I'm not saying the servos were at fault necessarily, just that this is a lead worth pursuing, which might lead towards the true fault...?

Time for beer! beer

Link to comment
Share on other sites

I've now run tests on the workbench with the following results:

  • The transmitter throttle-stick is NOT at fault - it works perfectly when tested on my other models (Acrowot throttle servo, MiddlePhase landing spoilerons, DLG landing flapperons).
  • The receiver is NOT at fault - all servos work correctly and it measured just under 5.5v at all states of -use (elevator, rudder, ailerons, throttle/ESC); also when powered by a 4.8v NiMh in the power-socket and a normal servo in the throttle-socket, it reduced down normally from 100% by just pulling down the throttle-stick.
  • The motor was intact and with a 13x8 prop peaked at 62A at full welly (the old wrecked one was 14x7 and had peaked similarly when previously tested); the ESC is rated at 60A with a 10sec peak of 80A.
  • However the EXACT SAME PROBLEM occurred on the bench as it did during both loops when trying to reduce from 100% throttle (with LiPo, ESC, motor and prop connected up as normal) - only the throttle cutoff switch stopped full throttle, after which it worked normally again; also, just as in the air, the problem did not occur when applying anything up to one rachet-click's worth of full throttle, only at 100%. As this all took place in a static situation, I can therefore rule out a poor connection for this side of the problem.
  • The new New Power servos are ridiculously slow: 60° took 0.7secs (compared to 0.4secs or less for other servos, e.g. Futaba 3003, Hitec HS85MG, and vintage Hitec HS300 servos!), the only redeeming feature being that they had no problem lifting 0.8kg of lead, albeit at the same slow speed!

Now, none of this directly explains the loss of control during the later spin manoeuvre (when the throttle was at idle), however as the ESC is clearly at fault in not reducing throttle from a full 100%, it is extremely likely that it is also the only cause of complete loss of power to the RX and therefore loss of control during the spin.

Remedies:

  • I need to speak to the original retailer tomorrow. The ESC was clearly faulty and resulted in the destruction of the model and one expensive LiPo. Also the four NP servos are so ridiculously slow that they're effectively useless for smooth flying. These really need to be replaced by one of the usual suspects which I know have the necessary speed (if not quite ever as advertised, at least not calibrated to 'geological' timespans).
  • I have already ordered a replacement Wot4 with a different power-train from a separate retailer. This has a slightly lighter and less powerful motor (the model doesn't need for me the unlimited vertical the 40 motor was producing) and a different brand of ESC. I'll also be playing safe in future and will use a spare AAA Eneloop 800mAh NiMh to power the RX. The overall weight will be much the same, but the NiMh can be placed further back to achieve a more satisfactory rearward CG which I've discovered is needed to spin properly!

Jon

Link to comment
Share on other sites

PS - I forgot to add that before connecting the motor I tested the power-draw of the various servos, both without any load individually and all together by rapid stirring of the sticks, and also by hanging 0.8kg of weight off the third hole out on the arms of each of them in turn. I don't think there's anything adverse here: from 2.2W (0.4A at 5.5V) just sitting there doing nothing, to an average of 5.0W (0.9A) swinging the stick from one end-point to the other, to a max of 12.4W (2.25A) by going bananas with all the sticks at the same time. By comparison the smaller Hitec HS81 servo draws 0.25A under expected flying-surface load.

Edited By Jonathan M on 27/01/2019 22:17:57

Link to comment
Share on other sites

May I add my pennyworth into this forum. I am mainly an i.c man, but I have made several forays into the electric scene. One thing that I was very particular about, was the power suppy to the Rx. I have always adopted a seperate power supply to the rx, and not trusted the 2 or 3s Lipo/Li-ion batteries to maintain a constant steady voltage to the rx,especially when one has upward of 5 or more servos for the battery to contend with. I have several Esc's with the red wire removed from the harness and taped back, so that I can use a 7.2 battery to just power the Rx. It's extra weight, I know, but worth it in my mind, for peace of mind.(pardon the pun)

Link to comment
Share on other sites

Ok now you've got a failure mode I've not previously even dreamed of.

I would not be so quick to pin it on the esc. Overlander sell rebadged hobbywing esc as far as I know and they are usually good. I have the exact same esc which has been used without issue in a very similar application on spektrum rx many times.

Full throttle and only full throttle caused the esc to latch the throttle input until throttle cut was applied.
That is the method you describe to reproduce the issue?

Can you confirm that you set the esc for the throttle sick movement?

Does this problem happen if you reset the model memory or try it with s clean model memory?

I would absolutely try this with the new esc when it arrives without changing the rx or existing motor and prop.

Also test with different prop. With No prop. Test with a servo tester in place of rx. Also test with the bec power removed and a nimh to supply rx power.

I think we need to rule out motor load as well as esc and tx and rx setup here.

You're not write at the bottom of the rabbit hole just yet.
Link to comment
Share on other sites

not totally convinced that the esc is guilty beyond doubt. You do appear to have been pushing what in all honesty is really a cheap Chinese esc to and beyond the limit. I would play safe and use an 80 amp esc. The max current could be less in the air than on the bench but I would use that as a safety zone rather than relying on that. Cooling is also a factor. Personally I might keep that esc for say a 25 size model

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

 Share

×
×
  • Create New...