potpotpotpot

1. How would I compute for the pulse width that I should give my servo. example I want my servo to be in 65 degrees position. How will I compute for that. My servo is HS-755HB (hitec). Its specs are here:

Control System: +Pulse Width Control 1500usec Neutral
Required Pulse: 3-5 Volt Peak to Peak Square Wave
Operating Voltage: 4.8-6.0 Volts
Operating Temperature Range: -20 to +60 Degree C
Operating Speed (4.8V): 0.28sec/60 degrees at no load
Operating Speed (6.0V): 0.23sec/60 degrees at no load
Stall Torque (4.8V): 153 oz/in. (11kg.cm)
Stall Torque (6.0V): 183 oz/in. (13.2kg.cm)
Operating Angle: 45 Deg. one side pulse traveling 400usec
360 Modifiable: Yes
Direction: Clockwise/Pulse Traveling 1500 to 1900usec
Current Drain (4.8V): 8mA/idle and 230mA no load operating
Current Drain (6.0V): 8.7mA/idle and 285mA no load operating
Dead Band Width: 8usec
Motor Type: 3 Pole Ferrite
Potentiometer Drive: Indirect Drive
Bearing Type: 1 Bearing and 1 Oilite Bushing
Gear Type: Karbonite Gears
Connector Wire Length: 11.81" (300mm)
Dimensions: 2.3" x 1.1"x 2.0" (59 x 29 x 50mm)
Weight: 3.88 oz.. (110g)

2. The datasheet states that the minimum pulse I can give the servo must be 900microseconds. I tried giving it 600 microseconds and I can get a bigger angle change and I need that much angle, can I use 600 microseconds even if the datasheet says that the minimum is 900 micro?

Nigel Goodwin

It's up to you if you want to run it like that?, most servos will perform fine outside their specified limits - but obviously the manufacturer only guarantees them to operate within their specifications.

__________________
PIC programmer software, and PIC Tutorials at:
http://www.winpicprog.co.uk

potpotpotpot

hi,. what do you mean by 1/4 scale. Anyway, about the computation of the pulse for the angle, do you know how? thanks

Nigel Goodwin

You don't compute anything, you simply adjust the pulse width until it's where you want it - it's a model servo, not a precision laboratory instrument!.

__________________
PIC programmer software, and PIC Tutorials at:
http://www.winpicprog.co.uk

potpotpotpot

sorry but what's a model servo? Can I use it in robotics???

Nigel Goodwin

It's one for use in a model, a model plane, a model car, a model boat etc.

Like the one you mentioned!, that's what servo's are designed for, a robot is just another type of model.

__________________
PIC programmer software, and PIC Tutorials at:
http://www.winpicprog.co.uk

philba

this thread is going backwards...

first, google is your friend. seek and yee shall be rewarded with a mountainous landscape of information.

by model servo, he probably meant "RC Servo" typically used in model aircraft. don't worry about it, it's still a servo. Yes, you can use it for robotics, just don't expect great precision from the unit. Also, if you want to use the servo as a wheel driver, you will need to modify it. many pages in the book of google tell you how to do that.

the specs you put up are mostly useless to answer your question. however, in general rc servos are commanded by a pulse width technique. Yours *may* be different but here is what most servos respond to:
- one pulse every 20 mSec
- the min pulse width is 1 mS, max width is 2 mS
- 1 mS pulse commands the servo to fully clockwise
- 2 mS pulse command the servo to fully counterclockwise (antiCW for you queens english types)
- 1.5 mS pulse commands the servo to neutral (i.e. centered)
- the pulse width can command to proportional positions. 1.75 mS will command it to half way between CCW and neutral

you should hook up the servo and play with it to get a better feeling for how it works.

Papabravo

Sorta like a servo with a minimum width pulse for an input(ROFL).

So, will servos that sees pulses at the right frequency, 50 Hz., but less than the minimum specified width go to the stop in the CW direction or will they just stay where thry are, not knowing what to do so to speak? For that matter what happens for pulses that exceed the maximum? Maybe (pot)^4 can tell us at some point.

Nigel Goodwin

The 50Hz frequency is VERY non-critical, only the pulse width really matters, I presume there is a point where the frequency can cause problems?, but anything from 25Hz to 100Hz (and probably more?) is fine.

As for the pulse width itself, notice that the servo only gives 45 degrees movement for the 1mS pulse change - it uses a 270 degree poteniometer to control it's position, so 'theoretically' it could move 270 degrees. However, I suspect that's not very likely

You need to consider their design criteria, it's important to remember that they are designed for multi-channel remote control of models - with aircraft being a common use.

So, as far as the frequency goes, you need to make it high enough for the servo to get the command 'instantly' - as far as the operator is concerned. If it has to wait too long then the plane might crash , but even a small noticeable delay would make it un-flyable (see another thread about this!). That's the 'too slow' side dealt with, the other side of the storey is 'too fast', this is restricted by the number of channels in use - if you have eight channels then the pulses alone require 16mS (8 x 2mS), which you can just squeeze in your 20mS time frame.

Likewise with the pulse widths, if you increase your 8 channels to 3mS then the pulses alone now take 24mS, which won't squeeze in the 20mS window.

As far as the movement is concerned, 45 degrees is all you need, and you design the model based on that - but for a robot, when NONE of the above design criteria are required, you can safely push it beyond it's normal design limits. I've no idea how far you can actually go?, and it probably varies from make to make, but I seem to recall 90 degrees is easily attainable?.

Perhaps someone might like to test one and see?, I keep meaning to - I bought a two channel radio control set a couple of years ago, with a view to a new PIC tutorial - but I haven't got around to it yet

__________________
PIC programmer software, and PIC Tutorials at:
http://www.winpicprog.co.uk

Papabravo

Thanks Nigel. I was not thinking about multi-channel mode. That certainly clears up a few things.

philba

The ones I use on my sumobots are GWS and they are very forgiving of the period. I converted them to continuous rotation but spent a few hours experimenting. A 10 mS period was fine. They will hold the position for a very long time in the absense of a command. this implies that you can keep shoveling a lot more servos in a system. The repeatability wasn't that great but those are pretty cheap servos. However, i suspect the design target is well less than 8 servos per receiver channel. I think the idea is that with more servos, you move to an all digital system.