Ccpm Interaction?

[ChrisB]

Need advice from experienced CCPM users on any heli but my problems are on my 2 600N`s a standard and a Pro

Pro first...

When i originally set this up there was no noticable swash tipping when going thru the full collective range but i have noticed that in flight from a hover to full climb positive pitch the heli veers left,but in a full negative climb it also seems to want to go left,when i now look at the swash from zero pitch to negative the swash drops to the right a BIT and from zero to full positive it seems to go left a BIT and i mean a bit..

This is obviously the reason for this but will a Cyclock address this problem?
If so How will it achieve it?

It seems that sometimes i may fly thru this problem and other times it really annoys me..

On the standard N it goes back cyclic in a positive climb and forward cyclic in a negative climb,but there is no real noticable tipping of the swash,i have purchased a levelling tool to check it properly

All linkages on both helis and servo wheels are exactly the same length and position so the problem is coming from one or more of the servo themselves

Again i am wondering if the cycloc would cure this,,if so how roughly would you go about tuning these problems out,i assume you have to do it with the heli connected to a pc while you fiddle..

Also what are the Phasing issues that the 600N does not officially suffer from and would this non existent problem cause any of the above

Neither front flip cleanly


These problems are one of the reasons i have got another Rappy

[AndersUK]

I you are confident that the geometry is 100% accurate then its probably the servos. Using the leveller get it dead level at 0 degrees (ideally mechanically, if not with sub trim) and dial out the top and bottom (what ever you have it set at +&-9, +&-10 etc?) with the ATVs.

To answer your question on the Cyclock, most definitely. I have heard that there isn't much you can't correct with one of these.

[mchristyuk]

What you are seeing is a set of mismatched servos..

Now normally one servo travels further than the others in BOTH directions.. however from the sounds of it yours is traveling too far in one direction and not enough in the other.

The way I fix this is by flying the machine and adjusting the travel adjust in the transmitter. The Cylock won't solve the problem for you.. you will just have to use it to make your adjustments rather than your transmitter.

I only get my machines roughly right on the bench and fine tune in the air.. at the end of the day its how they fly that is important.

Lets take your machine for example. For arguments sake lets assume it is the aileron servo (could be pitch) of your setup that is causing the CCPM swashplate to not travel correctly. Say you give it full pitch and the servo doesn't move enough causing the swashplate to tilt and the model to go right. Therefore go into to your transmitter and increase the travel adjust on the right side only for the aileron servo. Keep increasing it until the model climbs straight.

Then you need to do exactly the same thing for negative pitch.. therefore put her inverted and again climb out.. adjusting the travel adjust on the opposite side of travel until the machine climbs straight.

The same applies to your elevator situation. Use the travel adjust on the transmitter to alter both ends of the throw until your machine climbs straight.

Hope this helps..

Cheers

Mark

[ChrisB]

Mark

I think you have hit the nail on the head there,i didnt see it like that but having read your post it seems clear whats happening,i am pretty sure that on the pro the right hand side servo (pitch channel) is the one thats misbehaving..Its lower than the other 2 on full negative hence right input and higher at the top hence left input..If i adjust the travel limits either end it should sort it

i am using a DX7 and currentyl the travel limits are all set to 100% out of 150%

I will let you know

[Boggy]

if you have the leveling tool Chris and can dial out interaction on the dx7 dosent that make a cycloc a bit of an expensive white elephant? or does it do more than this?

[jamiem]

have you tried using the align 600 specific servo horns? they removed all the interaction i had on my 600e, but did nothing to solve the oh-so annoying phasing issue (obviously!). They do them for futaba, hitec and jr and cost about £3 a set.

[ChrisB]

Jamien

I have the Align horns on the Pro which is the worst affected,my servos are Hitec 6965`s

This is how i have checked it up to now

Powered her up and set the trims to centre and no sub trim,measured the distance from the 2 front servos top ball on the horn to a fixed point on the heli and they are both identical so the horns are exactly the same position,i measured every link rod and they are all identical.

Further to Marks suggestion i have done the following

At zero pitch i measured the distance from the top main shaft bearing mount to the outer edge of the bottom of the swash with a digital caliper and it was the same on both sides,i didnt bother with fore aft as the problem is only in the roll axis and it eyeballs level anyway..
I then measured the swash to brg mount at full pitch and the right hand side was slightly lower so i increased the ATV until it matched,i then did the same at full negative and adjusted as necessary

So now from full neg to full pos the swash is perfect,there has been no ill effect from adjusting the ATV`s on the relevant servo when doing full travel aeileron or elevator input so on the bench at least it appears to be an improvement

I will fly it tomorrow and fine tune as needed,hopefully this will cure the problem without the need for a Cyclock..

I will advise when i have flown it and do the same with the other 600N of mine,i am also going to fly my 600E as i am sure thats OK but want to check

[AndersUK]

The Cylock won't solve the problem for you.. you will just have to use it to make your adjustments rather than your transmitter

Mark

I have to disagree here. The Cyclock will solve the problem as it will provide sufficient adjustment to dial out pretty much most types of interaction whether from poor matching servos or otherwise. Agreed tweaking the ATVs, as we both stated, will do the job to a limited degree. But, its a crude way of doing it until you have used a Cyclock.

Quote from Ash's review

"This is an interesting question. Up until I bought the cyclock I was under the impression that my DX7 was doing an excellent job of keeping all of my servos in sync and driving them accurately. It wasn't until I installed the cyclock and started looking at it's configuration parameters that I realised that my setup actually wasn't that good at all. With the cyclock my cyclic servos are moving faster, more accurately and with less interaction than I could ever achieve with just the DX7 transmitter."

http://www.trextuning.com/cyclock.php

Cheers

Andy

[mchristyuk]

Andy,

I'm afraid I'll have to disagree with you

True the Cyclock will solve the problem.. but its a bit like buying a new TV to get all channels instead of retuning your current set!

How is adjusting the travel on the transmitter cruder than doing it on a Cyclock? It's doing exactly the same thing! The issue is with the servos being badly matched.. either the Cyclock sends the servo the pulse width commands to move to the correct point.. or the transmitter does it.. One is built in to the transmitter for free.. the other isn't.

"With the cyclock my cyclic servos are moving faster, more accurately and with less interaction than I could ever achieve with just the DX7 transmitter."

How on earth is the Cyclock making servos move faster?!.. The transmitter sends a command to the receiver.. the receiver sends that on to either the Cyclock or the servo.. either way it is the same pulse train.. the Cyclock may adjust the pulses to remove interaction.. but there is no way on this earth it can make the servo move faster.. unless it can see the future and know what command is coming next!

My understanding is that the DX7 sends the cyclic commands all together so the interaction will only becoming from badly matched servos. Older transmitters don't send the cyclic commands together so here a Cyclock would help a little.

The servos may be marginally more accurate going through a Cyclock.. but.. most setups aren't getting maximum accuracy anyway from the equipment at hand. For example a standard PCM setup (And the DX7) has 1024 resolution. However you can only get this if you run 150% travel adjust.. anything less and you're reducing the accuracy of the system.

It shocks me how people will throw money at a problem rather than spend a bit of time and care to set things up properly with the equipment they have..

The Cyclock is no doubt a good device.. but it's like a governor.. a nicety..

I feel like I'm sounding like a right old f@rt now.. but it does concern me the number of machines I see badly setup utilising a heading hold gyro, governor, and something like a Cyclock to mask the setup issues. The only way you can get the best performance out of these devices is to make sure the machine is setup properly in the first place.

Heh.. that sounds like a right old rant!!

Either way.. Chris can solve his problem for free with a bit of care and patience with the machine and transmitter.. no need to spend money on new tools when you already have everything to hand!

Cheers

Mark

[ChrisB]

I would rather solve the problem without an extra electronic device strapped to my heli..Also i would rather my Tx did the mixing than rely on software on an onboard device..

I am sure the Cyclock would be good but all i want is for the heli to track straight and climb straight,if the Tx cures the interaction then thats fine by me...

[Nutz]

I am with adders here, if you have to adjust the ATV to solve interactions you will in effect adjust the resolution of that servo and and could introduce other interaction, it is the same reason why you don't run sub trim on the cyclic.

There is many benefits of running a Cyclock but you problem could lye with that servo maybe it ain't 100% perfect?

As I don't think you always had this problem, did you Chris?

From what I have understand form what your have said is that one of the servos seems to travel to far one way and not enough, it is sounding like the centre is out, 2 ways of doing this. 1st adjust the link form the swash to the bell crank and re centre servo or Cyclock it as it will also solve ele, ail & col interactions as well as phasing, you already have servo sync with the DX-7 so you will not benefit form that on the Cyclock but the Cyclock with give you finer resolution on the servo traveling.

I agree a Cyclock will not make the servo move faster but the processing of the CCPM signal could well be faster on the cyclock than the TX in CCPM mode (depending on what TX you are using) hence faster servo travel.

I could be wrong but I like to keep an open mind on things

[ChrisB]

I will fly them both this afternoon and see where i am at,the standard 600N is only very slightly out,the Pro was more noticable but i cannot be sure if the issue with the servo has recently developed or whether it was already there and i was sometimes just flying around the issue and not really noticing it..

Its not until i have started deliberately looking for these issues with just pure pitch input and not attempting to correct the climb that i have actually noticed the deviation if this makes sense..

Either way,i appreciate all the input,it certainly gives me a better understanding for the causes and fixes...

[mchristyuk]

Nutz,

Agreed Chris will alter the resolution of the servo by changing the ATV's.. and this will add some interaction in.. BUT.. lets quantify what that would be..

1024 resolution.. at 150% travel adjust. Lets presume he's running 100% travel.

That equals only 682 bits of resolution.

For arguments sake lets say the servo moves through 90 degrees for full movement on the swashplate. That gives us 0.13 degrees of movement per bit of data.

If he adjusts his travel on one servo to 110% that will give him 0.12 degrees of movement per bit of data..

I challenge any human to notice that!

You're going to get greater differences through wear in the servos, flight load deflection, slop, etc..

Just as a little eye opener.. the Japanese F3C guys have probably the most meticulously set up machines you could ever set eyes on. They quite simple are works of art.. and the precision and accuracy of them is breath taking to look at and watch fly..

Governors? nope.
Regulators? nope.
Cyclocks? nope.

I nearly fell over when I saw they were running 1400MAH nicad batteries to run their equipment... 1400MAH??!!

However.. crashes from overly complex equipment failure? 0.

I'm a great believer in KISS.. I have seen far to many machines crash from failed regulators or some other unnecessary link in the chain failing.

If you can't get your machine to fly as you wish without these devices then there is something fundamentally wrong on your machine which needs sorting before you mask it with one of these.

They are great devices don't get me wrong.. however there is no substitute to getting the basics right first.. Only then will they be able to give you that extra percent of performance that your standard setup can't quite give.

Cheers

Mark

[mchristyuk]

Nutz,

You're kind of right about the Cyclock helping with CCPM processing..

However it won't make the servos move faster.. thats a physical impossibility..

What it will do is remove interaction from transmitter signal channel latency on older sets.

For example.. An older PCM transmitter will not send all the channels in one frame of data. It will probably do something like the first 5 channels then the next 3 to 5. If you are running a CCPM setup and 2 of your servos are in the first frame, and the 3rd in the next frame you are going to get a bit of a "wobble" on the swashplate with fast stick movements.

The DX7 and the PCM10x send the main CCPM channels together in the same frame.

The Cyclock helps the old systems do the same thing by allowing you to drive the 3 servos from a 1 channel command. That way your servos signals arrive in the same frame and your latency issue is removed.

Cheers

Mark

[ChrisB]

Mark

My ATV`s were all 100% out of 150%

The offending servo needed a 5% correction to cure,i have just rechecked it with digital calipers and measuring the full neg,zero,and full positive servo top ball on horn distances to a fixed airframe point and they are both identical where as before they were not

I cannot see any ill effects on the bench to other controls,proof will be in the pudding later and i will post tonight what happens