Fitting snakes

[Ron Gray]

2 hours ago, Geoff S said:

I've never had any success cutting 2mm threads for some reason. Of course one of the problems is that the 2mm push rods sold by SLEC which are threaded at one end only are rolled threads (like bicycle spokes) and the rod itself is less than 2mm dia.   I just buy 2mm threaded rod from Modelfixings and take the easy way.

 

I'm not a fan of snakes and I mostly use the outers as guides only for either metal push rods or Bowden wire for closed loop controls (it makes them easy to replace, too).  Apart from closed loop I prefer 3mm carbon rods with 2mm threaded insets epoxied in at each end.  They're amazingly rigid and dimensionally stable with temperature, which snakes aren't always.

 

Geoff

I use rod that is>2mm, usually from my spares 'bin' to ensure that a good thread is cut.

[kc]

This rather poor quality picture shows what I meant.    The opened out clevis has 'bolt' inserted ( with blob of solder to form a head on a threaded end ) and it's just the same as the locknut principle - tightening one against other eliminates the slack in threads and it's unmoveable until you hold one and turn the other then it releases.

Edited February 21, 2021 by kc
extra wording added

[Geoff S]

1 hour ago, Ron Gray said:

I use rod that is>2mm, usually from my spares 'bin' to ensure that a good thread is cut.

 

I haven't tried your method.  I did the 'obvious' way and held the rod in the vice and turned the die with the die holder. I always found it slipped off the rod rather than cut a thread.  Perhaps doing it your way round is better. I'll try it that way next time.  Most of my 2mm rod is 14 gauge stainless steel bike spokes which won't thread anyway, so  have to be careful which rod I select ?

 

Geoff

[Nigel Sharp]

I like to support the outer over it's entire length - takes a bit of forward planning where the servo height is.

 

I cut some softish to medium 1/4 sheet, glue some sandpaper around 1/4 dowel and route out the channel along its length. Then I bond the outers to the balsa with cyano or similar and use electric tie wraps to add a little security.

 

Then that rod is bonded to the fuselarge former(s) and or local cross braces.

 

What you don't want is secured ends and saggy middles as applying load in flight can lead to flutter etc. Just what John mentioned eariler.

 

 

[Trevor]

To screw the threaded extender into the snake inner, I just put the extender in the drill Chuck and use the low speed (screwdriver) setting.

[Ron Gray]

2 hours ago, Geoff S said:

 

I haven't tried your method.  I did the 'obvious' way and held the rod in the vice and turned the die with the die holder. I always found it slipped off the rod rather than cut a thread.  Perhaps doing it your way round is better. I'll try it that way next time.  Most of my 2mm rod is 14 gauge stainless steel bike spokes which won't thread anyway, so  have to be careful which rod I select ?

 

Geoff

Ah, yes Geoff, 14 gauge stainless, I wouldn't even attempt it!

[PatMc]

1 hour ago, Nigel Sharp said:

 

 

What you don't want is secured ends and saggy middles as applying load in flight can lead to flutter etc. Just what John mentioned eariler.

 

 

So long as both ends of the outer are securely anchored I don't see how it makes any difference whether the length in between is supported or not. It's the two exposed parts of the inner is that's doing the work, the rest of the inner can only be the same length as the outer whether it's loose or supported between the anchor points.

If a fuselage is roomy enough I use push rods or pull - pull for the elevator & rudder. I usually only use snakes when the rear fuselage is narrow such as a glass fibre glider fuselage. In most cases supporting between anchor points would be impractical.  When I do use "snakes" I prefer metal inner with plastic outer. I silver solder a metal clevis on the control surface end with length adjustments catered for at the servo end only. 

[Martin Harris - Moderator]

My preference is the simple Z bend at one end - if you're using snakes, the arguments for precision go out of the window and I've never had - or seen - one fail.  KISS every time for me!

 

[Nigel R]

17 hours ago, Basil said:

SMC, recomended I use the Sullivan GoldN rod semi flexible (Blue) type. There appears to be a hole of approx 2mm in the yellow/gold inner. What do you use to achieve a fit to horn/servo.Please.

Bas

You should have all the required hardware in the pack with the rods?

 

I use the blue sullivan snakes by preference. Some good fitting advice here. 

 

Only thing I can add,  it is worth using a locknut on the metal clevis. Don't forget sullivan come with imperial threads. 

[Jonathan M]

1 hour ago, Nigel R said:

You should have all the required hardware in the pack with the rods?

 

I use the blue sullivan snakes by preference. Some good fitting advice here. 

 

Only thing I can add,  it is worth using a locknut on the metal clevis. Don't forget sullivan come with imperial threads. 

 

I've often wondered about the necessity of that:  i.e. if there is the same stock hardware at each end, then surely even if the inner rotates a bit so its final position moves a mm or two, its actual length stays the same?  And it would have to rotate an awful lot before one end or the other comes undone!

[Nigel R]

A pack of 2-56 nuts is very cheap insurance.

 

Whilst the length may remain the same, I'd rather it didn't rotate at all - catastrophic failure ensues if it comes out.

 

On small stuff, like 20 class or less, I just use a plastic clevis on the ends instead of the supplied metal ones, as plastic ones won't rotate.

 

Here's my own 2p on installing a pair:

 

Edited February 22, 2021 by Nigel R

[Jonathan M]

True.  Although a slight smear of PVA on the threads also 'thickens' things up slightly but not so much that one can't adjust the thing.

[Nigel R]

There's many ways to skin the cat ?

[kc]

Lots of my models have used snakes and I always straightened them by inserting a length of piano wire inside to keep them straight whilst using masking tape to fix them to 1/4 sq balsa.   Of course the piano wire was removed after taping.  The snakes were crossed so they were as straight as possible between servo and horn.

However my favourite model uses a birch dowel pushrod instead of snakes and that is superior - never changes trim when it's hot or cold.   That model uses closed loop for the rudder so no chance of pushrods clashing.   Better system and probably lighter too.

[Nigel R]

I've only ever noticed one or two clicks of trim change with the Sullivan kit.

 

5 minutes ago, kc said:

That model uses closed loop for the rudder

 

Closed loop is the best.

[John Stainforth]

13 hours ago, PatMc said:

 

So long as both ends of the outer are securely anchored I don't see how it makes any difference whether the length in between is supported or not. It's the two exposed parts of the inner is that's doing the work, the rest of the inner can only be the same length as the outer whether it's loose or supported between the anchor points.

If a fuselage is roomy enough I use push rods or pull - pull for the elevator & rudder. I usually only use snakes when the rear fuselage is narrow such as a glass fibre glider fuselage. In most cases supporting between anchor points would be impractical.  When I do use "snakes" I prefer metal inner with plastic outer. I silver solder a metal clevis on the control surface end with length adjustments catered for at the servo end only. 

That's what I thought at first, but the reality proved me wrong. It does not take much slop in the pushrod system to create quite a bit of give in the control surface. A 5 mm throw on the TE of the surface might correspond to about a I mm throw on the horn, and if one is using a 1 m (order of mag) pushrod, that can be achieved by only about 1/1000th change in the length of the rod. First, I checked the outer sleeve and found that I had secured this very well all the way along, so it did not seem to be the problem - it had to be in the pushrod itself. The minute slop in the holes in the servo arm and control were contributing to the unwanted movement in the surface, but only to a small degree. So the culprit was definitely the pushrod of the snake. I am a big fan of Slec products, and whilst they are simple and practical they are not exactly precision engineering. The Slec snakes have quite a lot of slop between the OD of the pushrod and the ID of the sleeve, and the pushrod is rather flexible. When the pushrod is relaxed (control surface neutral), it adopts the shortest (straightest) path inside the sleeve. But when a counter force is applied to the control surface, the pushrod is put into compression and its flexibility allows it to conform to slight curvatures of the sleeve. Worse still, because of the sloppy fit of the pushrod, it can cross-over from one side of the sleeve to the other, following the longest possible path. (I am using very exaggerated terms here to make the point; in reality, these movements are very small.) I suppose once the pushrod has taken up this longest possible path (effectively shortening its length) it has nowhere else to go and with sustained compressive stress the plastic of the pushrod itself can then compress slightly. The carbon fibre pushrod, I used as a replacement for the plastic one, was a better fit inside the sleeve, was also much less flexible so could not flex and buckle so much, and is also much less compressible.

Does anyone here know if there are better quality snakes on the market than the Slec and Sullivan ones? I would be prepared to pay more for quality here, because once a snake is built into a fuselage it would be very difficult to change out later.

[Martin Harris - Moderator]

5 hours ago, Jonathan M said:

 

I've often wondered about the necessity of that:  i.e. if there is the same stock hardware at each end, then surely even if the inner rotates a bit so its final position moves a mm or two, its actual length stays the same?  And it would have to rotate an awful lot before one end or the other comes undone!

As I mentioned earlier, a Z bend at one end prevents rotation but there is the possibility of vibration fretting threads and a lock nut protects against this.

[John Stainforth]

28 minutes ago, Martin Harris - Moderator said:

As I mentioned earlier, a Z bend at one end prevents rotation but there is the possibility of vibration fretting threads and a lock nut protects against this.

I use the lock nuts, plus a dab of glue, not so much to prevent rotation but because the screw fittings are often a very sloppy fit in the metal clevises. The nut stabilises the whole thing, preventing it from flexing from side to side (which could lead or unwanted or even catastrophic wear).

[leccyflyer]

If the clevises are a sloppy fit it's worth making sure that they are the correct clevises supplied with the particular threaded insert. I saw a Wot 4 auger in some years ago and the failure was due to that - a clevis that seemed to fit, but was a fairly sloppy fit - either a metric on an imperial thread or vice versa. It was holding just enough for the control surfaces to move during the preflight, but in flight it let go - it was also set up to give up elevator with pulling on the clevis. Result was a dive into the embankment, with nothing to do but shut the throttle.

[John Laird]

I have and continue to use Sullivan rods in my big vintage models and always anchor them along their length as  flexing here reduces the throw at the horns.

 

I recall reading somewhere that these rods ( and any of similar material ) are subject to length expansion in hot weather - never bothered me with my vintage models but it might affect trim on aerobatic models.

 

I also use 1/8" OD carbon tubes as the hole is a slightly slack fit for the quick link threaded wire ends.  I bend the wire slightly to get a friction fit then when I have the length right with the link on, I drip thin cyano down the tube to glue the wire in position and also bind the end of the carbon tube - the cyano wicks thru the "bundle" of fibres making up the tube and strengthens it.  Roughen the wire first.  I also support the carbon tube along its length it it a long rod.

[Nigel Sharp]

 

In regard to soggy middle ?

 

For the little effort involved in securing the outer firmly when I have invested lots of time and sometimes money then I will support them properly.

These things are mostly plastic and flexible. Even if the outer is secured at each end then by the look of some supports  in the photo's you can get deflection of these fixings.

 

If you get high aerodynamic forces on rudder or elevator then that can push the inner back into the outer and a bow happens between the ends its rather like pushing carbon fibre poles into tent pockets true the pole then holds up the tent because the ends are held firm. But take note on our RC models one end is always free to float e.g. the end attached to the wiggly bit. In that case what would happen to the tent? Yes it would collapse.

 

Another example would be an overloaded shelf it bends between the supports and the ends lift a little essentially the distance between the end faces shortens as it makes the chord of the arc.

 

The converse happens if you have a bend in the snake inside your fuslage you apply control force but when flying at high speed like a dive aerodynamic forces take over straightening the snake meaning you lose resolution, worse it doesnt pull out -> but the same thing.

 

So I'll stick to securing the outers. Generally my plane to gound incidents are caused by the ground to stick interface rather than soggy snakes.

 

[kc]

Nobody commented on my gadget for inserting threaded ends into snakes - so either you all understood or perhaps nobody did!

So here is a slightly better picture of the gadget made from an opened out metal clevis and a bolt ( actually a spare threaded end with a blob of solder to form a head)

The bolt is simply screwed until it meets the threaded end and then it 'locknuts' it enabling the threaded end to be driven into the snake, then when bolt is unwound the clevis bit simply comes off without unscrewing from the snake.

 

 

Edited February 22, 2021 by kc
extra wording

[Ron Gray]

Another method for inserting the threaded metal rods, hold the thread rod in a soft jawed vice then screw the snake onto it (ribbed snake inner works easier doing it this way). You could always use a 2mm tap to cut the thread in the inner.

[Geoff S]

1 hour ago, leccyflyer said:

If the clevises are a sloppy fit it's worth making sure that they are the correct clevises supplied with the particular threaded insert. I saw a Wot 4 auger in some years ago and the failure was due to that - a clevis that seemed to fit, but was a fairly sloppy fit - either a metric on an imperial thread or vice versa. It was holding just enough for the control surfaces to move during the preflight, but in flight it let go - it was also set up to give up elevator with pulling on the clevis. Result was a dive into the embankment, with nothing to do but shut the throttle.

 

One reason I always bin any non-metric threaded parts,  usually American.  They appear to fit, until, as above, they don't.  The only time I'll use them is on an ARTF where replacement is very difficult because they're part of the structure.

 

Snakes are my last resort when I can get nothing else to fit. 

 

Geoff

[Martin Harris - Moderator]

I do prefer Sullivan clevises to the largely unbranded metric fittings.  Having witnessed flutter removing a fuel tube "retained" clevis, I really like the sprung keeper that clips the Sullivan together. 

 

They also have an interlocking join where the rolled tubing section for the thread is joined.  I've seen cheap metric fittings which have opened up slightly - or perhaps were never formed properly in the first place?

 

A small stock of 2-56 or 4-40 locknuts is a cheap price to pay for the added peace of mind from using these higher quality items. I've never had a pin detach from one either - I can't say the same for some 3mm clevises I once bought!