upgrading servos
#4

My Feedback: (8)
ORIGINAL: TedMo
As Ryan asked what is wrong with the Spectrum servos/ i believe they are digital so what more could you want?
As Ryan asked what is wrong with the Spectrum servos/ i believe they are digital so what more could you want?
I put the Airtronics 94772's in the ailerons. If you want to upgrade the ailerons and the elevators, specifically for the Venus II, they are amazing and around $70 right now. Very fast at .09s on 6V, and about 90oz of torque which is plenty for the surfaces on that plane.
[link=http://www.hobbypeople.net/gallery/474372.asp]Airtronics 94772[/link]
I never got around to doing the elevators on that plane before I sold it, but it would have been a major improvement. I had the plane loaded with JR DS821 to start.
#5

My Feedback: (42)
Joined: Jul 2004
Posts: 878
Likes: 0
Received 0 Likes
on
0 Posts
From: Randolph,
NJ
As Ryan asked what is wrong with the Spectrum servos/ i believe they are digital so what more could you want?
The term "digital servo" is basically a popular misnomer. None of the servos used in this hobby at this time are truly digitally controlled. Each and every one responds to the same control input, which is nothing more than a signal with a varying pulse width from about 1 to 2 ms in duration. The "digital" terminology is useful to the extent that it helps to distinquish legacy technology electronics from newer, uP (microprocessor) based offerings.
One basic, and very important performance difference between uP servos and older technology stuff is that the uP servo can apply some pre-programmed intelligence to smack the servo motor more frequently to have it attain or hold a position with more authority than is otherwise possible. Legacy servo designs can only command the servo motor at the rate at which they receive an updated control pulse from the receiver. When each control pulse arrives, they do the best that they can to assert position control by looking at the error signal that exists between where the output arm is, vs. where the input signal is telling it to be. The electronics then stretch what can be a very thin sliver of an error pulse into a pulse of voltage to the servo motor that is long enough to actually get the motor to do something useful.
By contrast, uP based servos can deal with the fact that it's going to take a while in between control updates from the receiver, and that even a stretched and amplified error signal that is applied to the motor can be greatly enhanced by doing one simple thing: apply the correction signal to the motor more frequently under progammatic control (embedded software inside the servo). The net effect of this is that you get greater efficiency out of the motor & gear train, because it is being tasked to do work over a greater duty cycle (higher percentage of time) than a legacy servo can ever do.
All that aside, it can be summed up by saying "digital" is a lot better than "non-digital" when it comes to moving or holding a control surface with authority. That's your baseline, i.e., if you care about that part of the performance equation (and you should), don't even consider anything other than a "digital" servo for precision flying these days.
Now then, the real question that you want answer is, how to choose the right "digital" servo for your application. It helps to ask yourself, what do you care about? Is it pure speed of response? Is it raw torque? What about centering and tracking precision? It's most likely a combination of these things. Since this is the Pattern forum, it is a foregone conclusion that precision is a key factor. Therefore, look at deadband specs, which directly correlate to centering accuracy. The smaller the deadband is, the better. Try to find a zero deadband servo if you can. Let me know how that works out for you...
If you care about how responsive the servo is to your inputs, which basically means how quickly its movement responds to your stick inputs, then you should also pay attention to motor technology, as well as transit time. Many legacy servos used 3 pole motors (slower start/stop response). Some used 5 pole motors (faster start/stop response). Some digital servos still use these older motor designs, because they are cheap. The best servos, "digital" or not use coreless motors. The Readers Digest version of why this is so: 5 pole motors can apply angular torque more efficiently than 3 pole motors, and especially, there is much less rotating mass in a coreless motor as compared to any conventional motor , so it is easier to start/stop than a conventional motor design.
To further refine your search for the best servo, you should also consider the gear train construction. Again, since this is the Pattern Forum, you don't need a metal gear train to have your model make it through a flight. Nylon, or at least non-metal gear trains will almost always have less slop to begin with, and have virtually no wear as compared to metal gear trains. Obviously, they are fragile by comparison. There are in between gear train solutions of various types, but just use this as general guidance to help pick apart what those variations may offer.
Hopefully this helps. It's not necessarily a Spectrum vs. other brand question, and especially, it is not a digital vs. non-digital question. Start with "digital" and then look at everything else.



