Look after your connectors!

[Allan Bennett]

Some time last Autumn my Multiplex Twinstar crashed shortly after taking off, due to the ailerons going berserk.  Having glued the front end back on and, I thought, tested everything, I took it up again some time after New Year with exactly the same result 😒

 

A week ago I took my Flair Magnatilla to the field for its first outing in many months, but found the ailerons going berserk during ground testing.  It didn't take long to find that wiggling the wing servo connection caused it, so I took Maggie back home unflown.

 

A common factor in these two models is that I use gold-plated pin-headers mounted on a PCB and glued to the fuselage side.  The reason for this is (a) so I can plug the wing in single-handed while holding the wing in the other hand and (b) so I can have as many channels as necessary all on one plug.  The first thought of mine and my flying buddies was that the header pins were perhaps worn or corroded, but when I took them out to replace them they looked to be in perfect condition.  I can only assume therefore that the female connectors inside the plug were either corroded or, more likely, loosened by regular plugging and unplugging.  I replaced them, of course, and flew without incident last weekend.

 

I've been using this system for my wing connections for some years now on many models, but my TwinStar and Maggie are flown most so get the most plug/unplug cycles.  Seems to me now that replacing wing connectors, of whatever kind, on a regular basis should be considered.  I've now ordered a few of 4-Max's Max-lock connectors to see if they might be better than my pin headers.

[Chris Walby]

Horses for courses springs to mind. My son had a PZ Artizan (foam, mid wing aerobatic) model which had detachable wings. Thing is it fitted in the car so the wings never came off although the servo leads and extension had keepers on.

 

One day it all checked out and then mid flight it went berserk, I don't think I would have got it down let alone without any damage. On inspection the servo lead to Y plug was causing the servo to travel from one extreme to the other and most places in between. Unplugged it and plugged it back together and never had a repeat of the problem.

 

Less connections the better and if you do need connections the connectors made specifically for the job may be the best option.

[Philip Lewis 3]

Connections that are made and broken should always be sprung connectors, they are made to do that, if you plug in ailerons leads regularly change the connector periodically as the connector was never made to be connected and reconnected time and again. 

 

MPX, XT90/60/30 and 4mm banana plugs are all examples of sprung cpnnectors. 

[Peter Jenkins]

Interesting point!

[Philip Lewis 3]

This is a major factor in consumer orientated connections which are in mass use. Your standard 3 pin plug is as good an example as any. USB C is not smaller than Micro USB but significantly one of the main reasons why USB C came about as a replacement was because of the early failure rate of micro USB, USB C has a much higher coonect/disconect failure rate, the other reasons were power capacity in charging devices and data speed transmission.

 

Obbviously the connections we use in this hobby are not tested to the extend of mass use connections but we can certainly learn from the principles. 

[SIMON CRAGG]

Good heads up.

 

Think I might treat my extension leads that are regularly make / break, to a dose of Electrolube.

 

Can't do any harm.

[Graham Bowers]

8 hours ago, Philip Lewis 3 said:

Connections that are made and broken should always be sprung connectors, they are made to do that, if you plug in ailerons leads regularly change the connector periodically as the connector was never made to be connected and reconnected time and again. 

 

MPX, XT90/60/30 and 4mm banana plugs are all examples of sprung cpnnectors. 

@Philip Lewis 3I recently had a debate with myself about connections for a 4 servo wing and favoured Ashlok type connectors as the pins and sockets have crimped cable insulation support, whereas Multiplex have no inherent support. After reading your post I fished around on the internet to see what commercial cable suppliers use for support of multiplex connectors and it seems to vary from a bit of heat-shrink at one end of the spectrum to these Jeti connectors at the other. https://www.nexusmodels.co.uk/jeti-mpx-soldering-couplers-power-jms-mp6p.html

 

Do you have a view on best practice of how to deal with potential failures at the stiffness transition of a soldered servo cable to a multiplex pin please? Jeti connecors or pot them, for example?

 

By the way the Ashlock pins and sockets look to be non-sprung, however the truth is often in the fine detail and I've never held one in my hand. Are they sprung, do you know? https://www.fighteraces.co.uk/product-category/airframe-accessories/wiring-connectors/

 

[Brian Cooper]

It is refreshing to see this topic coming back into a collective consciousness. 

 

Those of us who have been in the game for a number of years have learnt about the frailty of  connectors, but we take it for granted that everyone else knows about the problems when, in reality, they don't. 

 

The regrettable fact of life is that after years of plugging them in (and out) our connectors wear out.  

Leaving them plugged in doesn't give immunity from issues either because eventually a "film" grows between the contacts, which then gives a poor "jittery" connection. 

 

So... Catch 22. . . Plugging them in helps to keep the contacts clean and bright, but it also makes them wear out. 

 

Personally, I adopt the kind of thinking which is more associated with full size aviation.  Namely, fix it before it breaks. . My connectors are replaced after three years' service. 

Don't forget your on/off switches either. They wear out too. 

 

I also approach all details of a model, especially the radio installations, with the mindset of thinking I would actually be in the thing.... and would not want to be flying in it with anything substandard, weak, or bodged. 

 

Remember, with aircraft, if there is anything that CAN fail, then it ruddy well WILL (sooner or later). 

.

 

[J D 8]

                   I use WD40 contact cleaner. This product is formulated to do what it says on the tin.

[Allan Bennett]

5 hours ago, SIMON CRAGG said:

Good heads up.

 

Think I might treat my extension leads that are regularly make / break, to a dose of Electrolube.

 

Can't do any harm.

 

26 minutes ago, J D 8 said:

                   I use WD40 contact cleaner. This product is formulated to do what it says on the tin.

 

Unfortunately I didn't put the female parts of the connectors under a microscope after I replaced them but if, as I suspect, they had loosened rather than corroded then contact cleaners are not going to be much help.

[EvilC57]

2 hours ago, Graham Bowers said:

@Philip Lewis 3I recently had a debate with myself about connections for a 4 servo wing and favoured Ashlok type connectors as the pins and sockets have crimped cable insulation support, whereas Multiplex have no inherent support. After reading your post I fished around on the internet to see what commercial cable suppliers use for support of multiplex connectors and it seems to vary from a bit of heat-shrink at one end of the spectrum to these Jeti connectors at the other. https://www.nexusmodels.co.uk/jeti-mpx-soldering-couplers-power-jms-mp6p.html

 

Do you have a view on best practice of how to deal with potential failures at the stiffness transition of a soldered servo cable to a multiplex pin please? Jeti connecors or pot them, for example?

 

By the way the Ashlock pins and sockets look to be non-sprung, however the truth is often in the fine detail and I've never held one in my hand. Are they sprung, do you know? https://www.fighteraces.co.uk/product-category/airframe-accessories/wiring-connectors/

 

For a 4 servo wing with effectively 12 connections, I’d probably use a 15-way D-type connector as used in computers etc.. We used them a lot in military spec equipment when I worked in the aerospace industry. If they’re good enough for them, they’re good enough for me. I used the 25-way D-type below in a large twin engined model a while ago, and the connector has never given any problems.

 

You can see that each soldered pin has heat shrink sleeve over it for reinforcement, in this case I think this is sufficient. But in other cases where more strength is required, other than fitting a backshell if convenient, I’ve partially potted the wires in hot melt adhesive if necessary.

 

Edited April 18, 2023 by EvilC57

[Cuban8]

Manufacturers saving on their costs by not using gold flashed terminals has led to a lot of problems. I'm pretty sure my original Futaba 6 channel outfit that cost plenty in the early 80s was all gold flashed including the servos.

 

[Nigel R]

2 hours ago, Graham Bowers said:

Multiplex have no inherent support

 

MPX connectors are pretty good. Sprung loaded, solid connection, and wrap up a decent number of pins. You have pointed out the one obvious weakness though - no strain relief.

 

Fortunately it is an easy fix - you can pot the connector in nice cheap hot melt glue.

 

There are fancy moulds available to do this. Not cheap of course.

 

Or, you can wrap a bit of electrical tape around the connector for a makeshift mould. With a bit of care, this can look neat and is functionally as good as the fancy moulds.

[Nigel R]

 

2 hours ago, Brian Cooper said:

Those of us who have been in the game for a number of years have learnt about the frailty of  connectors, but we take it for granted that everyone else knows about the problems when, in reality, they don't. 

 

Part of the problem is that there is some rubbish out there. Cheap, naff extension leads. RX switches that are frankly unfit for purpose. Cables that are too lightweight in either strand count or insulation type. Etc.

 

2 hours ago, Brian Cooper said:

Don't forget your on/off switches either. They wear out too.

 

I like quality (i.e. branded) DPDT switches with enclosed wiper connections (this one is good).

 

They will wear out, but have a longer life in our applications than the usual cheap crud sold by hobby shops.

 

The wiping action helps to avoid the build up of the problematic film - "self cleaning" is the phrase, I think. A toggle switch does not have this property.

 

[Graham Bowers]

9 minutes ago, Nigel R said:

 

<SNIP>

There are fancy moulds available to do this. Not cheap of course.

<More SNIP>

 

Thanks, didn't know they existed.

[Philip Lewis 3]

You can use these to give some strain support on MPX connectors (as in the picture) and I've used these for the servos in the wing which allso offer some support but plenty of other good choices available.

 

As for switches I would only ever use the fail on type but as my planes are all electric these days I don't use any switch, I just plug the battery straight in using xt30 as the connector, after all I'm going to have to connect the  drive batteries up anyway so another connection when the cannopy is already off isn't going to make a great deal of difference, it eliminates another point of failure.

 

This is a ver healthy debate.

I  

Edited April 18, 2023 by Philip Lewis 3
Spelling

[Graham R]

2 hours ago, Graham Bowers said:

Thanks, didn't know they existed.

Worth their weight in gold. very secure encapsulation

[Graham Bowers]

4 hours ago, Philip Lewis 3 said:

As for switches I would only ever use the fail on type

Fail-on switch sounds sensible. Googling doesn't return much, would you mind telling more please?

[Philip Lewis 3]

Pretty much all electronic switches would fail on, think of them as like a relay except that the when the relay isn't energised the switch is on, they actually use power to hold themselves off, so if that "relay" were to fail the switch would be permanently on and would not then turn off, when you look at the technical description it normally quotes a current drain in the off position and that is why.

 

Electronic switches have two other advantages, there are no moving parts so nothing to wear and because they are electronic they are sealed to the air so imune from contamination and corrosion.

 

Here are some examples

 

https://www.bondaero.co.uk/details/p1085152_20817837.aspx

 

https://www.t9hobbysport.com/frsky-nfc-power-switch

 

https://www.t9hobbysport.com/hall-magnetic-switch-with-battery-redundancy

 

They are also sometimes reffered to as fail closed switches because when they fail the contacts are closed.

 

 

Edited April 18, 2023 by Philip Lewis 3
My awful spelling!

[Phil Green]

Even high-quality connectors such as used for RS232 leads etc are typically rated for maybe 200 plug/unplug cycles.

Our nasty tinplate servo sockets are nowhere near as well-made or as durable.

 

Edited April 18, 2023 by Phil Green

[Graham Bowers]

12 hours ago, Philip Lewis 3 said:

Pretty much all electronic switches would fail on, <Snip>

Thanks, I understand which technology is being referred to now. I was optimistically wondering if there were fail-on versions of simple "wiper" switches.

[Graham Bowers]

8 hours ago, Phil Green said:

Even high-quality connectors such as used for RS232 leads etc are typically rated for maybe 200 plug/unplug cycles.

Our nasty tinplate servo sockets are nowhere near as well-made or as durable.

I imagine many theses have been written on the subject of connector durability and this thread has been a bit of a prompt.

 

It seems the best we can do is have a plan, and that may be different for different models.

 

Think I'll start by sorting my servo connector bits into "gold plated" and "tin plated" as I know I have both. And be a bit more selective in future!

[Graham Bowers]

21 hours ago, Nigel R said:

There are fancy moulds available to do this. Not cheap of course.

By the way, I found 3D print files for such molds. https://www.printables.com/model/32349-multiplex-6pin-hot-glue-gun-mold

Time to dust off the Prusa 😉

Available for XT60 too.

[Bob Burton]

On 18/04/2023 at 09:36, EvilC57 said:

For a 4 servo wing with effectively 12 connections

Why would a 4 servo wing have 12 connections rather than 2 for power and 4 for the data lines ?

[EvilC57]

-

Edited April 19, 2023 by EvilC57