Futaba digital servo pulse-width range

For the very beautifully made Futaba digital servos, Futaba tech support absolutely insists that the neutral-center command has to be a 1500us pulse.

This appears to be incorrect. I have run numerous tests using the SERVO command on a PICAXE 08M, and 1310us sets the neutral-center position.

Futaba tech support insists that 1310us dates back to very old analog servos, but, I have this thing right in front of me, centering itself when I send a SERVO 2,131 command.

What about the ends of the rotational arc? Futaba tech support sez, "send a pulse anywhere from 1000us to 2000us." Well, this is incorrect also. My digital servo DOES NOT RESPOND to a 2000us pulse. (How about yours?) Laborious trial and error shows that it responds within this range only:

"Low" end: 680us
Center: 1310us
"High" end: 1940us

Since the SERVO command divides microseconds by 10, you can use any number in the range 68 to 194.

If you send SERVO commands outside this range, the servo ignores them. This makes perfect sense since it is a DIGITAL device and you may notice that the range I have stated above allows 127 steps of 10us each. In other words the servo probably has a 7-bit internal controller.

What about frame rate? Some say a digital servo requires more than 50hz. Futaba tech support says 50hz is fine. I think a more accurate statement is that the servo will tolerate faster frame rates, but mine works fine with the PICAXE 08M sending pulses at 50hz.

I hope this info may help some users who may otherwise be as frustrated as I was, wrestling with this problem yesterday. As for Futaba tech support, they got tired of arguing with me.
 

BeanieBots

Moderator
Welcome to the forum.

1500uS is the 'normal' today's standard for neutral.
1310uS was in use back in the days when Freddy Flintstone made servos.
ALL digital servos are post 1310uS standard.

Servo positioning is WAY better than 7-bit. Typically 16-bit, many 32-bit internally.
(a few early ones were only 8-bit)

'Standard' frame rate is 50Hz.
SOME digital servos will accept >200Hz.
MOST servos will be happy with 100Hz.
SOME servos don't like < 50Hz but most are OK.

Many digital servos can be programmed. It is possible to program in different neutral positions and/or end stops. It is also possible to program different behaviour for 'out of spec' pulses and/or no pulses.

Maybe your servo has been subjected to such programming?
Do you have a programmer?
If not, take it to your local hobby store and ask them to look into it.

Exactly which digital Futaba servo do you have?
 
Last edited:
Programmed by vendor?

I've only just started playing with servos so, naturally, I respect your experience. But I just hooked up another digital servo, model BLS451 straight out of the box (mail ordered from Tower Hobbies) and it behaves identically to the BLS551, with center-neutral at 1310us and the extreme settings as I specified in my initial post! It ignores any value outside that range.

Here's my guess: If the servo is programmable, as you say, maybe Tower Hobbies programmed it to be compatible with the old AM transmitters which had center-neutral 1310.

I'd take it to a local hobby store for "deprogramming" except that the nearest hobby store is 50 miles away and they don't have a clue.

Thanks for your input...I guess my initial message should be retitled, "For those who mail-order digital servos from Tower Hobbies."
 
I'm looking at a Hitec HFP-20 servo programmer/tester, supposedly works with all makes of servos; wondering if I should try that. I also emailed Tower Hobbies asking if they reprogram servos before shipping them.
 

BeanieBots

Moderator
An interesting hypothesis but I don't think it's likely.
I've never used Tower Hobbies so cannot comment but I can't see them buying in from Futaba and re-programming them.

Futaba however are one of the world's leaders and I would trust their servos to "do what it says on the tin".

It might even be that your PICAXE is off frequency!
If it is using the internal oscillator it could (worst case) give a pulse as short as 1425 for 1500. Accumulated with a tollerance in the servo itself might be enough to produce what you have observed.

Check with a 'scope if you have one.
Alternatively, if you have another PICAXE, test the servo with that and see if you get the same results. You could also use another PICAXE (ideally using a resonator or Xtal) to measure the pulse width using pulsin.

Do you have any 'regular' RC gear to test the servo with?

There are still some outbacks in China which make servos to the older standard but I'd be very surprised if they ever ended up with a Futaba label on them. (though not impossible:mad:)

All I can say is that I've not seen a servo with 1300uS neutral for decades now but it has been mentioned on here before and fairly recently. Maybe they also got their servos from Tower Hobbies!

EDIT:
Just picked on the first google instane of BLS451
http://www.servocity.com/html/bls-451_brushless_servo.html
Spec shows 400uS - 1520 - 1900uS
 
Last edited:
Tower Hobbies says they don't reprogram servos; they just pass them along as they receive them from Futaba. Here's the very simple BASIC listing that I'm using to cycle the digital servos to extreme clockwise, extreme counter-clockwise, and center, via a 330 ohm resistor on pin 1, using a tactile switch on pin 3 (with pulldown resistor) to trigger the cycle.

main:
servo 1,131

checkinput:
if pin3 = 1 then sendtomotor
pause 100
goto checkinput

sendtomotor:
servopos 1,68
wait 2
servopos 1,194
wait 2
servopos 1,131
goto checkinput
 
Even if the PICAXE oscillator is off, I don't see how the lower and upper limits could just happen to be +63 and -63 from the 131 center point. (Note, I checked values outside that range and the motor simply ignores them.) Seems more likely to me that a batch of motors was programmed to be backwardly compatible with old transmitter/receivers. Incidentally the motors are made in Taiwan.

I don't have any RC equipment. I only discovered servos a few days ago. I'm using them to rotate a cam that squeezes a tube in some prototype medical equipment. A low-cost high-speed pinch valve.
 

Technical

Technical Support
Staff member
The PICAXE commands on the lower range parts (08M etc) are not fully accurate, it is quite possible that a servopos value of 150 does not give out exactly 1500us. This is due to internal tolerances with the timers etc. However the value it does give out will always be consistant.

However in the real world this makes very little difference and doesn't matter, as on almost all servo setups you need to manually select the value which moves the servo to the desired position by experimentation. So if the end position is 194 or 200 doesn't really matter - just select the value that works for you!
 

andyshrimpton

New Member
Textstar module

Hello

Have look at this nice Module it has a servo test mode that will tell you what singal is being sent to the servo. I used it to find a problem on my RC helicopter. You can also use it with your Picaxe. Have a look around the site and the videos.


http://cats-whisker.com/web/node/101



TextStar 2 - The World's smartest serial LCD module!
The Cat's Whisker 'TextStar' module is a multi-function 16 x 2 LCD display that can be used as:

a simple serial LCD display module for connection to a UART/serial interface
a self-contained serial terminal with four programmable buttons
a servo signal analyzer function for radio control applications
'TextStar' has been designed to be a small, easy to use and adaptable LCD display for connection to a UART or serial-enabled device. It will work at speeds between 2400 and 115200 baud and with both TTL and RS232 level signals - no MAX232 required! Built using ultra-thin 'chip-on-glass' (CoG) technology it is supplied in a transparent heat-shrink covering and can be used in a number of different ways, e.g. mounted in an enclosure, held in the hand or even stuck onto surfaces with double-sided/Velcro® tape.

Set-up is extremely simple; there are no DIP switches or jumpers to configure and connecting the module the wrong way will not damage it which means it is ideal for educational and experimental applications. It will also operate with power supplies varying between 3.5 and 20 Volts making it extremely adaptable.

Great product

Andy
 
Last edited:

andyshrimpton

New Member
sorry about link

I have corrected the link. Sorry are we not allowed to post links to products that we have purchased and find useful? Please advise. I was only trying to help.
 

Wrenow

Senior Member
Lookss to be a nifty gadget for servo signal analysis. Sadly, it doe not apopear to hav a servo signal generator like most servo testers, and is a bit more dear. Howqever, very6 tiny and a nice addition to the RC toolbo9x, methinks. And the Picaxe toolbox, too.

Cheers,

Wreno
 
Top