Hello,
I am relatively new to roboteering, so forgive my ignorance.
I have been creating a power supply/harness/thingamajig for a Lynx L6 Arm with upgraded servos. I fried one servo when it was stalled, and I wanted to avoid doing that again, so I got some Sub-C battery packs, fuses, and a voltage regulator. Now the batteries should be able to provide stall current without catching on fire, and the fuses should protect the servos from smoking.
Now here’s where my question is. Is it bad to run servos rated for up to 6V at 6.2V for hours at a time? The voltage regulator is supposed to be adjustable, but no matter where I adjust it to, it is set to 6.2V.
On the L6 arm, I have a 5990TG, two 5645MGs, two 475HBs, and two HS-81s. All of those are rated for 4.8 - 6 volts except for the 5990TG, which is 6-7.2V. So, I would like to run them at 6V or just below, but I have already tried two of these adjustable voltage regulators and both have been less-than-adjustable.
Am I safe or should I try a different regulator or make my own?
Many thanks!
6.0vdc is the preferred voltage for servos. There is no easy way to protect servos from burning up with fuses. The problem is not necessarily with the amount of current, but with heat buildup. This heat buildup will happen when the servo is doing a lot of work even when the servos are operated within their voltage limit. Probably the best way to protect the servos is to limit the amount of time the servo is doing a lot work. This is actually how the 5980 and 5990 servos protect themselves from overheating. They don’t have a way to measure the temperature, but they can easily keep track of how long the control board is providing x amount of current.
You can go with some temp sensors like Analog Device’s ADT75, the one I like to use is the ADT75ARMZ, which uses the 8-pin package slightly smaller then the SOIC-8 package. These are the fastest temp sensor I’ve found and uses the I2C bus.
Since the temp sensor can’t be directly mounted on the motor, you’ll probably have to mount it outside on the servo housing. You can calibrate out what the temp sensor is seeing versus a “touch” with your hands and place limits accordingly. Your 5990TG servo with the aluminum extrusion for heatsinking will probably transfer heat better, so this “fudge” factor will probably be different as compared to your other servos.
Thanks for the help! Sorry I haven’t responded sooner, I didn’t get the notification email until today… I probably need to check my settings.
I’ve got a cooling fan on the servos to mitigate any trouble with heating (although I know it’s not the ideal solution).
I’m more concerned about the input voltage. I’m using the Adjustable 5vdc - 6.5vdc Regulator from LM seen here
After a bit more testing, I have found that it does regulate the voltage correctly when I put 11.2V into it. So there must be some threshold. The manual for the regulator says if the input is above the desired voltage then it will regulate, otherwise there will be a 0.3V drop. My battery packs off a fresh charge are up over 7V and the voltage regulator gives out about 0.6V less than the input. The manual also says it is designed for 5-Cell NiMH packs, which is what I’m using.
I have been successfully running the servos at about 5.8V with a partially drained battery. Will this damage the 5990TG with input range from 6V-7.4V?
So to be safe I can partially drain the batteries to 6.6V where the regulator will output 6V, but then I lose a lot of the battery life. Can anyone recommend a voltage regulator that actually outputs 6V when the input is reasonably above 6V (and can handle at least 5 amps)?
The other alternative is that I just run the servos at 6.5V or whatever the peak is for this regulator for some amount of time until the batteries have drained to the safe level. Is the problem here heat dissipation? What are the problems that can happen while running servos above their spec voltage?
Thanks for your patience and help.
I just tried something on a whim, and I think it solves all of my concerns here. The image that shows connections on LM’s site is wrong. Jim or somebody, you should double check this.
I looked at MPI’s other voltage regulators, and the inputs and outputs are swapped from what is shown for the MX-07 image. I thought the connectors were pretty backwards anyway, but I believed the image.
I just tried switching my inputs/outputs, and with the battery at 6.4V I get the full range of regulation… 5-6.5V. In other words, it works as advertised.
I am still curious to know what the problems are with running servos at slightly higher than spec voltage, though… just for my education.
Thanks again!
you can introduce a couple of problems with high voltages. the most obvious is that it wears on the brushes in the motor within the servo more so it will fail sooner. less obvious are that the transistors used to drive the motor need to withstand the increased voltage, the increased current, increased power dissipation, and the increased noise and spikes casued by the additional abuse of the motor brushes. last thing I can think of is that a lot of servos don’t regulate the supply voltage to the controller IC so if you jack the voltage up too much it can burnout the controller itself. “too much” really depends on the servos being used.
Ah, that’d be my fault! Very sorry for the confusion there. I’ve fixed the image; thanks for pointing out my error. 
It’s no big problem… I’m just glad that the thing didn’t go up in smoke with so much current going through the wrong way for so long.
And EddieB, thanks for the response. The general impression I get is that a slight increase in voltage may not instantly kill a servo, but will decrease the lifetime and give jittery behavior.
yeah, kinda, just remember that slight is a pretty non-specific term. boosting digital servos scares me more than the analog ones, but that might just be my wallet talking, heh. 
Really though there are quite a few posts on the forum with people talking about what they have used for supply voltage on many different servos and projects so, while no one has compiled a list as far as I know, you may be able to search up some useful info to use as a guideline.
All good things must come to an end… eventually…
Has anyone used a servo to death? Until the motor is nearly worn out and puts out torque that are no longer useful or cannot hold position with the slightest load?
I always believed that my servos will outlast my desire to use it, that is, a tougher, badder, better, stronger servo will come out before I kill my current one (hope I’m not jinxing myself)…
Really though there are quite a few posts on the forum with people talking about what they have used for supply voltage on many different servos and projects so, while no one has compiled a list as far as I know, you may be able to search up some useful info to use as a guideline.