I have just detected winch creep on the new v2 winch that I purchased when they first came out. If I sheet in slowly the winch consistently moves to the same point ok. If I sheet in fast the winch goes to the same point then creeps off letting the boom out slightly!!! I have checked out the tx/rx etc and have narrowed it down to the winch.

Has anyone had a similar problem? 

Yes the winch in my new DF65 appear to be doing this.

On 19/07/2023 at 13:36, Paul Middleton said:

Yes the winch in my new DF65 appear to be doing this.

Was there an answer to this or do you just live with it??

I'm afraid the explanation for this gets a bit complicated and requires a rudimentary understanding of control theory.  The short answer is that you would need to re-design (re-program) the winch internals to change its behaviour.  Some winches (e.g. Kingmax ones) can be re-programmed to change some of the control parameters (e.g. dead-band).  I don't know if this is possible for these winches but this may be a fundamental control algorithm problem that you can do nothing about.

Can you do an experiment for me please?  With the winch fully sheeted in, push on the boom to simulate a sail loading.  Push hard enough to move the winch a bit and then try to hold the same pressure.  Does the winch stay where it is, or does the force it produces slowly increase to try and return the winch to the original position?

Many sail-winches (particularly the smaller ones and nearly all the old analogue ones) use a very simple control algorithm that is fine for a servo that operates about a lightly loaded central position (like a rudder) but is not very good for a servo that needs to hold a given position under varying load.

If you do understand control theory then I think what is happening is that, when the winch is moved fast, the winch very slightly overshoots the set point.  The control algorithm appears to have an integral term which slowly increases the error until the winch moves back in the opposite direction.  Either the winch has a pretty wide dead-band so that it stays noticeably 'further out' or they've got the control algorithm wrong and the reverse integration term isn't working properly to move it back again.  You would need to put an oscilloscope on the motor terminals to see exactly what is going on.

As it happens I've been playing with winch control algorithms a bit recently as I'm rather disappointed with the performance of my Kingmax winches.  They seem to back off quite a long way under load - which is not ideal on a 10R.  What I have learned so far is that controlling these things accurately is more difficult than you might first think as the motor's torque v current curve is hugely non-linear.

 

 

My DF65 winch that showed this symptom actually fully stopped working about 2 sessions after I put in my comment above and went back (for a replacement) and I fitted my spare winch whilst this process the return.  

I suppose these things are mass produced in China with an interesting attitude to quality control so you are bound to get the odd 'dud' one! I have Whirlwind winches in my other boats - Dave Andrews knew what he was doing! John, I broadly understand what you are saying so I'll leave it at that! Of course this is only an issue when either sheeted fully in or out.

I have just found the same problem with a DF95 fitted with the new winch and not used for 6 months. I have replaced the winch and everything is OK. It is obviously a manufacturing quality problem. 

I happened upon the Tushy/Tim vid about latest 65 n 95 products today, and a new winch is mentioned which is better for something like 'endpoint stability' - sounds like the same creep we're mostly describing here? Anyone have this 'v3' winch and confirm it solves that one-time creep-out?

I find it irritating to set up, when just the few mm alters boom angles quite a bit - compensating sheet length for what it will (probably) do on the water.

I'm tempted by the v3, if it does have a better control loop (== a performance ìmprovement....?).

Tho 90 quid for a servo is a lot.... (presumably not enough volume for economies of scale compared with a general market servo)

As I've been sailing DF65 and DF95 from the early days I've had all versions of the winches available through that time.

In my opinion the latest version is the best so far ....... but ..... is not always perfect. It is definitely an improvement on the slightly earlier similar looking grey version, and much better than the even earlier blue (and black?) one.

The price of just the winch on it's own does make you wonder if it is better overall value to buy a complete new boat with it installed !!

Be careful if buying one as the previous grey version is very similar in looks, the big difference in operation is that when you connect the rx battery the new winch will SLOWLY go to its commanded stick position, after it has got there it will then operate at normal speed.