Why do you use HV and / or why don't you?

[Terryscustom]

That's the thing though, why test them at 8v when the onboard pack comes down to less than that with even a 1 amp load. What I want to see is this.

For elevator servos:
1 or 2 - Fully charged LiPo pack(s)
2 - Servos with arms that are actually what is used on giant scale planes (let's say 1.75")
2 - Scales

For aileron servos:
1 or 2 - Fully charged LiPo pack(s)
4 - Servos with 1.5" arms
4 - Scales


That's the real number that you are pulling in the plane, these video tests with 1" arms and a power supply that holds 8v under load are laughable IMHO as they present no real data that applies to real world flight applications.

One Exception would be if you run a 3S pack like @ghoffman through a BEC. If I go to HV I'd probably look seriously at that option because it seems the most effective with the very least voltage drop under load.

 

[DeathHorse]

I think the idea is making the tests comparable. Lipos add a big variable to the mix.

The arm length at 1" is also standardized, but how the torque changes if you add 50% more length to an arm should be easy to calculate.

 

[Terryscustom]

I think the idea is making the tests comparable. Lipos add a big variable to the mix.

The arm length at 1" is also standardized, but how the torque changes if you add 50% more length to an arm should be easy to calculate.

Calculate yes, but that does not account for different motors and such used in different servos and the ability to hold under those circumstances. Real world torque may be completely different from what you calculate. I guess what I'm not understanding is that they do these tests with arms that are more for the RC car and 10 degree throws of an IMAC setup than for the majority of users. I do understand standard testing, but just not the ones used for hi-perf aircraft servos.

 

[Xpress]

Well keep in mind these tests are to show what the servo can put out in a controlled environment using what is basically a standardized test. The torque/speed values are based off of a 1" arm. Obviously if you add in something like a 2" arm, the values will diminish. However, being that the HSB servos use a way overpowered NEU motor inside of them, I don't foresee them diminishing in a linear value based off of the length of the arm (so 1/2 for a 2" arm, 1/3 for a 3" arm, etc.). The graph would have a curve to it.

Obviously I'm in no position to test them, I'd be biased, however you guys are more than welcome to do some testing on the servos and report back what you find

 

[DeathHorse]

I think the torque values should almost perfectly follow arm length. Also the ratio of leverage between the servo arm and hinge point also matters. If an RC car has a .75" arm going to a hinge point with the same .75" that's no different in torque than 1" to 1" or 2" to 2".

Maybe they should do these tests where they max out the travel or close to it, since 3d setups often use most of the travel.

 

[Terryscustom]

Agree. I was going to bring up the leverage point at the hinge line too, but that's hard to test and every plane mfg. has their own arms and adjustments so that's nearly infinitely variable.

So servo guru's, lets see some real world testing numbers

 

[3D-Joy]

A poorly designed mechanical installation in a plane will require a stronger servo to achieve the same results.
IMO a good setup will actually GAIN mechanical advantage as you get to the end points so effectively this makes the surface move with more force but with less travel per degree of servo movement.

Torque is just a force times a distance. This is a fact. A NEU castle motor is not going to deliver more torque if you're using a longer arm, this is just not understanding what torque really is. If your graph is anything else than a straight line then you have poorly executed the readings and/or lots of other possibilities.

 

[jhelber08]

Since last year everything new for me has been HV. I started off using Lions as opposed to Lipos but then switched to using Lipos due to price and weight savings. For the cost of one Fromeco Lion I can get 3 pulse 2550s or 3600s. I have 6 of each that are interchangeable in most of my planes and so far have been very pleased with them. I still have Lions in one plane and in my jet where the batteries are in a difficult spot to get to. I have also lost a cell on a Life battery and on an A123, both ignition batteries, which kinda left a bad taste in my mouth. I never discharge past the storage point and always balance just to be safe and leave them stored in a lipo bag on my garage floor next to the door so that if they decide to self destruct they're not next to anything flammable. Its also nice having them interchangeable throughout airframes so that if I get to the field and change my mind as to what plane I'm going to fly I just throw the batteries in and attempt to start hucking I also like the benefit of higher torque and faster servos on HV.

 

[DeathHorse]

This may or may not be relevant, but from setting up electric motors in trucks, they seem to have brutal torque. Way way more than they need. A 1/10th scale motor could be put in a much heavier 1/8th scale truck and still tear around with more torque than any nitro engine. The issue is the 1/10th scale motor would quickly overheat and the hotter it gets the more resistance in the wires and the faster it gets hotter. Usually the only reason to put a bigger engine in to get some more copper in the motor to deal with the amperage without heating up as quick, but even an undersized motor can put out a silly amount of torque.

Xpress can comment, but maybe this explains how they could setup a brushless servo normally rated around 400oz/in to pull around 800oz/in. I believe the motor could easily pull 800oz/in, but the wiring would be a little undersized for the amperage this would require and it wouldn't be able to dissipate the extra heat it would create from the higher watt output. The case strength being another factor...

To tie this back to planes I'd be curious to know how fast brushless servos are compared to coreless under load. So if you put 375oz on a 400oz servo, would a brushless version maintain speed better than coreless. In theory the motor can produce way more torque/wattage than it can cool so maybe the brushless servos can put more power down moving the surface. There must be some pretty sophisticated programming controlling the motor output. In a RC truck it's a dumb system so if you gear it to the moon or stick the motor in a much heavier truck it will try and succeed to turn at the speed asked of it until it overheats.

Another thought would be to run brushless in the rudder for knife edge loops etc..as the brushless should produce less heat than the coreless and have a bigger margin before failure.

 

[Xpress]

Well we could make the HSB 9380 push 800oz if we wanted to, but the case, geartrain, and wiring wouldn't handle the loads so we made it more conservatively powered.

A standard or coreless digital will slow down as the servo arm nears its outer control throw limit because the loads tend to be their greatest out at those points from the airflow on the control surfaces trying to push everything back. Brushless servos don't slow down (not nearly as much) as the control throw nears the outside limits because of all of the additional power available with the motor. You can still stall a brushless servo, but it's more difficult to do. It's almost like having a servo with more torque.

This can explain why you won't get a linear scale of diminished torque with a longer arm when comparing to other standard or coreless digital servos. Like I said, you guys are more than welcome to do your own testing if you wish to see for yourself.