I bought 10 Hitec 5990TG for my biped scout, and i seem to have a big problem with them. all the servos jitter if they are under a very small force ( barely touching the bracket controlled by the servo, and the bracket is not attached to the robot right now) ; if i apply more force(not enough to leave any mark on my finger ) the jitter stops, and the servos does it’s “normal” high pitch sound. Can anyone help me please. When the Scout is fully assembled this becomes a big problem as it looks like an epileptic robot… i am quite dissapointed by the servos if this is the way they work.
I am using a 2cell lipo and tried to control the servos either via radio control, or from a dedicated board. I checked with an oscilloscope and the pulse is steady as a rock, but i still get this problem.
Was thinking to try and get the dead band higher, but i don’t find any refrence of how to do this, and i don’t want to spend another 150$ on a programmer as i am not even shure this would fix the problem.
I’ve 12 of those myself. So far I’ve just tested them using a servo-tester. I can confirm you findings. Some of my 5990’s do jitter under now load and I’ve to agree I was a little disappointed myself. I bought used ones so I was thinking that might be the reason but your findings confirm that it wasn’t.
It appear like the servos need some sort of load to stay calm…
I can’t help you more before I’ve done some more progress on my MorpHex.
If you are using a 2S LiPo directly its definitely not a power issue, its a 5990 issue.
EDIT: Maybe some sort of dampening/friction spacers in the bracket could solve it. I might do some experiments with that.
Yes, i am using a 2.2Ah 7.4V Lipo battery. Tried to supply from a 5V output source and the same thing. It happens no matter if there is 1 servo connected or 2 or 10. it’s not a power problem. I checked with the oscilloscope and there are no reall drops on the line.
I am using a self made board with a cortexM3 from ST (stm32f103rb). This board gives the control PWM. i tried with a lot of diffrent periods but nothing helped ( 4ms, 8ms, 10ms, 40ms, 80ms -> thought it may be a control issue). Tried to change to diffrent command signals from 800 - 2400ms and the result is the same. My conclusion is that tere is no command or pwer problem. ( tested with various other normal servos on the same commands and they worked perfect)
What i think happens with the 5990tg is that the gear on the potentiometer can move ever so slightly, but the rest of the gears don’t fell this motion because the gear train lacks precision and so the motion isn’t stoped. The servo orders the corection, and the motor moves a bit, but as the gears are not precise enough, it does a small overshoot. and we go back to square 1. For 120$ this is a pitty as in the uC program they could have easilly solved this problem by software. All you have to do is to enshure that your regulator can’t go into a undampted oscillation state.
As far as i am concerned, apart from the torque this servo isn’t worth the 120$ everyone asks for it. I have some cheap china made ones ( but only 5kg/cm) that don’t jitter at all… Any ideea if 7950TH has the same problems? i am thinking of selling this bad investment and get some other ones.
If you find any fix please tell me. Do you have a programmer? maybe try lower the gains? or make the dead band larger?
increase it how?. i am not ready to invest 150$ in a programer, when i am not even shure it will matter at all… Do you know how can i self program the servo ? i found some details over that HMI protocol but no command for the dead band and for the gains.
I’ve an HFP-10 programmer. The DB width programming function don’t work on the 5990’s. I’ve tested it on the 5645’s and that work fine. So I’m not sure if you need a newer programmer or if the 5990’s doesn’t support DB changes.
On hitec site, i see that the hfp-25 is for the 5xxx and 7xxx series. I’d have really liked if we could get the full hmi code; i’d just program them myself
I saw now that you can use the HMI Servo Programmer Kit to program this servo. this is quite cheap (24$ on lynx) pitty the transport would cost me 120$ ( why ups only ? )
“All Hitec servos require 3-5V peak to peak square wave pulse. Pulse duration is from
0.9mS to 2.1mS with 1.5mS as center. The pulse refreshes at 50Hz (20mS).”
I bought a HPP-21 Plus+ programmer, and I’ve been able to program deadband on the HS5485hb servos I also just received. Nice little gadget! Works connected to PC or in the field. Needs a 5V receiver battery pack to operate, 'tho (even w/USB). 'Spose they assume R/C modelers would have an extra battery around. (I’m not an R/C modeler, but I bought an R/C for a 'bot control a few years ago.)
So you get deadband, endpoints of servo range, some safety checks, and exercise of you servos. Really handy along with an R/C Servo Protractor…
i picked up 12 of these servos to go into my scout, but have the same problem with them jitter and the hipitch sound
and i had the hole put together and it looked like the scout was freaking out
i do have the Lyxn programmer but have never used it yet
I had forgotten about this. The high pitched sound is normal. Annoying, but normal. Increasing the deadband should improve this though. Let us know what you find out.
i used the hmi servo programmer on the 12 X 5990tg’s and changed all settings on it using the hmi programmer and still no go they set there and jitter all the time.
so i called hitec to see what they say about it. and they asked me a few question on what hardware i was using and the robot that i am using in it and the battery that i am using to power the servos with so i told them
told them this i am using the scout from lynxmotion
battery lipo 2s 5000 mh 25c and the ssc-32
and the verion of the software for the hmi programer i was using ver 1.0.2
hitec sent me a newer version of the software which is supposed to take away a lot of the noise away and the jitter out of the 5990tg
this ver is 1.0.4 and along with directions to upgade the software and the firmware on the 5990tg
and for the most part took a lot of the jitter out of the servos, but didnt take away the noise level of the servos so now i can at least continue to set up my scout, i have one side of the legs done and no jitter in them just the high pitch noise, witch is better than the jitter and the nosie at the same time 8)
? )
The high pitched sound is normal. Annoying, but normal. Increasing the deadband should improve this though. Let us know what you find out.