"Have you actually counted the teeth (or measured the diameters) on all the gears and done the math on the gear ratio? There is no way it is 200:1 as has been stated on here so many times. Think about that for a second..."

I have counted the teeth. I do not make claims I wont back up.

"Fine, you got me."

I don't mean to kick a man when he's down, but if that Futaba 9250 that "Four Stroker" mentioned is not enough for you, here is one from Hitec and one from Airtronics:

Hitec HS-7965MG
Pinion = 10
1st Gear = 61 / 10
2nd Gear = 50 / 16
3rd Gear = 29 / 13
4th Gear = 42
(That is a 178.6:1 gear ratio)

Airtronics 94774
Pinion = 11
1st Gear = 56 / 12
2nd Gear = 44 / 14
3rd Gear = 42 / 12
4th Gear = 41
(That is a 191.3:1 gear ratio)

With the Hitec HS-7965MG, If I rotated the 1st gear while having the servo powered up, I would get about 9 - 12 teeth of movement out of that gear before the motor kicked in to counteract me. That means roughly 1 motor revolution before it kicks in. That means the servo has around +/- 2.01 degrees of deadband. In all fairness, I think this servo is programmable and you can lower the deadband with the Hitec servo programmer.

With the Airtronics 94774, If I rotated the 1st gear while having the servo powered up, I would get about 3 - 4 teeth of movement out of that gear before the motor kicked in to counteract me. That means roughly 1/3 of a motor revolution before it kicks in. That means the servo has around +/- 0.63 degrees of deadband.

2.01 degrees of deadband works out to +/- 0.035" if your linkage is mounted 1" out on your servo arm.

0.63 degrees of deadband works out to +/- 0.011" if your linkage is mounted 1" out on your servo arm.

"Why hasn't anyone been measuring servo resolution?"

I have been for a while

-Farrell

EDIT: With my trim authority set to 1%, and end points set to +/- 150%, both servos will move with each click of trim. (They both have a resolution that is greater than 300.)

John;

I tried to get you to expand into servo testing about a year or so ago ! They aren't as advertised ! There are some grotesque exaggerations in the achieved resolution. MAN used to do actual servo testing but advertising revenue soon trashed that since there were very large differences between brands.

Hootowl;

Direct servos is a bad bad thing. My Kalt Mercury 50 has this setup and it eats servo pots. I am happy the Align 600N goes through bellcranks.

Have a look at this from a gliding forum:

http://www.rcgroups.com/forums/show...1057205&page=15

or servo testing:
http://www.youtube.com/watch?v=KQgl...player_embedded

Not heli servos, but a comparison between 'good' and 'bad' servos

Simon

This calls for a test jig.

Cue 'Jaws' theme music . . .

op-thwõp-thwõp-thwõp-thwõp-thwõp-thwõp-thwõp-thwõp-thwõp-thwõp-thwõp-thwõp-t

> I tried to get you to expand into servo testing
> about a year or so ago !

I know. I just don't have the time or access to servos.

> Have a look at this from a gliding forum:

Someone needs to make videos like that for our beloved heli servos.

I've always loved DS3421 servo and could tell a dramatic difference between it and other similar size servos (like the 9650) while flying the same heli with each servo. The video of the 3650 helps confirm why.

Farrell,
> I have been for a while

Where is all of your data posted?

- John

The worst servos ever made electronically used the World Engines chip. Someone had the bright idea of using the servo pot as a resister instead of a voltage divider thereby saving one wire in production. I think this guy's name was Jim Lanterman. This idea bankrupted many radio companies that had been around for decades. The graphs of servos using this chip in MAN clearly showed AN ORDER OF MAGNITUDE greater deviation from where the servo was supposed to be as compared with other servos of the day. When the pots got dirty, the servos would run over to one end instead of just buzzing like current servos.

The worst servos ever made mechanically used a "patented harmonic drive" and were sold with the Bonner 4RS and 6RS. This was a flexible band planetary gear system which broke.

Of course, I owned a set of each of these.

Hi JKos

Do you have any data for the Futaba 14mz, 12z, 12fg using the new HS receivers?

Is there any difference between the 14 ch and 8 ch HS receivers?

Info would be greatly appreciated.

Regards
Paul

Well extrapolating from John's data the 12Z/14MZ should be 38.4 - 3.5 ms avg. with 6008HS. Not that much and probably completely unnoticeable. Next gen high end may be a long wait.

Farrell,
> I have been for a while
Where is all of your data posted?
- John

I never posted the data as no one seemed to care. Those two I mentioned were servos I checked recently. In the past I recall checking the Hitec 525MG, Hitec 965MG, Airtronics 94359, Airtronics 94361, Airtronics 94758, Futaba S9551 (might have been a S9550) and a few $10 servos just for kicks.

What got me into checking servos was when I modified one for continuous rotation (no pot or board... just a motor and gears) I realized the gear ratio couldn't possibly be around 500:1 which would be needed to deliver on some manufacturer's claims.

-Farrell

The 9C is horrible for eCCPM and a good example of how bad things were when eCCPM and 3D flying kicked in. You can compare the 9C to any of the newer radios and say the same thing.

I am by no means a 3D flier (yet), but I learned to fly on a 9C super. I started out with planks, and started flying helis. When I got my first Trex, I thought the servos acted slow then I started doing some research. I ended up selling my 9C (which I loved to a sailplane flier), and bought a 10c (after everything, it cost me $30), and I felt like I needed to learn how to fly again. I copied over the models, and everything felt so much more responsive, almost twitchy. I am still using PCM (I bought a 10chp so I didn't have to buy all new receivers) and I can notice a difference.

Four Stroker

ROFLMAO

Man, I am dating myself admitting I knew all those
servos and the MAN articles where I too noticed the politics of
editorial corruption sold out to the transnational corporations...
They were good before the fit hit the shan...

just like is always the case today...

Did you ever fly reeds?

I would love it if John had time to
slay some servo dragons...
and I must extend deep gratitude for the magnificent
work he has done to date.
And, there is still a great deal of room for improvement in
servo motors, quality, and longevity...and specs that
are actually true...my big helis demand all the
8717 HV units can give, and I am waiting to see what the brushless
ones to come will do.
I do not mind at all spending a lot more for a servo that
actually does measure up to UAV/industrial specs, and want that
type of performance.

best

michael

Not everything that can be counted counts, and not everything that counts can be counted.

I have seen reed systems and tuned one but I was poor and was doing the Adams actuator, Rand galloping ghost and fast pulse thing about then.

Everything that I have ever owned is here: http://www.rchalloffame.org/

I think a Rand actuator with a brushless motor would sell today as a novelty.

I would also like to see a servo that can deliver 1024 steps. I don't know of any. Manufacturers can claim higher, but if the output shaft of the servo can't deliver the claimed resolution it means nothing to the user.

That JR 3421 and it's cousins (3650 and 3517) deliver over 1024 steps in just 110 degrees of movement. It takes between 11-12 trim clicks on a 14MZ set to 1 trim step to move the servo 1 degree.

You can see each movement and it's less than .1 degree. I wonder if the trim step were to be able to set lower, if we can see more steps to move 1 degree.

BTW, Thanks so much for the effort JKOS! Even if I am a few years late.
TL

We need an accurate pulse width generator and a 14 bit absolute shaft encoder. Then we could make some graphs and compare servos and brands! Then you could put a load on them and try again.

For those of you that seem to get caught up in latency numbers maybe you should take a look at the following description of human reaction time.

There are four basic means of measuring RT given different operational conditions during which a subject is to provide a desired response:

Simple reaction time is the time required for an observer to respond to the presence of a stimulus. For example, a subject might be asked to press a button as soon as a light or sound appears. Mean RT for college-age individuals is about 160 milliseconds to detect an auditory stimulus, and approximately 190 milliseconds to detect visual stimulus.[7]

Go/No-Go reaction time tasks require that the subject press a button when one stimulus type appears and withhold a response when another stimulus type appears. For example, the subject may have to press the button when a green light appears and not respond when a blue light appears.

Choice reaction time tasks require distinct responses for each possible class of stimulus. For example, the subject might be asked to press one button if a red light appears and a different button if a yellow light appears. The Jensen Box is an example of an instrument designed to measure choice reaction time.

Discrimination reaction time involves around Natwain comparing pairs of simultaneously presented visual displays and then pressing one of two buttons according to which display appears brighter, longer, heavier, or greater in magnitude on some dimension of interest.

Due to momentary attentional lapses, there is a considerable amount of random variability in an individual's reaction time. To control for this, researchers typically require a subject to perform multiple trials, which are then averaged to provide a more reliable measure.

This doesn't really apply to this conversation. You are comparing a reaction time that requires someone to see something, process what they are seeing, remember what the reaction is supposed to be, and then respond to it. With R/C, you are usually performing a maneuver that was thought out and planned ahead of time.

I work at a Porsche/Audi dealer. In the past couple of years, Audi has started to use LED brake lights. Now, the human eye/brain cannot see the process of a filament based bulb heating up and giving off light (don't know in miliseconds what that is). We just see it as on or off. A LED lights up much quicker than a bulb. With that being said, I can tell you in a second it a car has LEDs or bulbs. Even though you brain cannot supposedly distinguish the difference, in "real world" conditions, we can.

> For those of you that seem to get caught up in latency numbers...

It's not just about reaction time which would imply a response to something the helicopter is doing, but more importantly how quickly the helicopter responds to user inputs once that input is decided and made. That response time directly impacts the "connection" felt with the heli.

The human mind is incredibly capable of predictive actions for which pure "reaction" cannot explain. An easy example is hitting a fast ball. Science says the batter has to decide they are going to swing before the ball even leaves the pitchers hand. Yet, batters are obviously capable of hitting a fast ball. Another example is stopping a watch exactly on X.00 seconds. With practice, your mind knows when to initiate a muscle command to press the button to stop at X.00.

If the watch responded differently to each button press (variable latency) or had a latency greater than some finite amount, it becomes much harder to stop it at X.00.

The radio is part of a control loop. Reducing delays in a control loop are almost universally a good thing. Each piece of the control loop has an impact. There are parts of the control loop we can control such as the radio latency and servo speed. There are parts we cannot have much impact on.

- John

Very well stated John.

Steve

So Steve, when can we expect the next gen high end ?

For those of you that seem to get caught up in latency numbers maybe you should take a look at the following description of human reaction time.

Speed matters. Anyone want a good PC XT? Nobody wants to go back to slow.

Without getting into a lot of specifics the next generation is going to be a lot different. I'm seeing the beginning of it now.

TM

Proud Member of the FFC, Nitro Engines Forever