We have a robot similar to 8059A’s, (I am pretty sure every knows about that robot) and we have been having a problem with the battery draining very quickly. We have a power expander to lighten the load on the batteries. The flywheel normally runs at 90% speed and when we are not using the flywheel in competition we bring it to 30% power to conserve. We have 4 turbo motors powering the flywheel with the same gearing as 8059A. With a full battery we can shoot the ball from the starting tiles. After about 30 seconds we have to start moving forward a little every few seconds to make it into the goal because the battery is decreasing so much. After a couple minutes the battery seems to stops draining and we never have to get any closer than halfway to the goal to shoot into the goal. Does anyone have any suggestions on how we could extend our battery life just a couple more minutes? Thanks in advance. (There is a picture of our robot below).
Do you have all four of your turbos for the launcher on one battery? I would recommend putting each half on a separate battery. Also, from what I’ve read of team B.L.A.N.K.'s thread, there launcher doesn’t normally operate at 90%. Is may not be necessary for your flywheel to go that fast. On one of their other threads they also said they had a similar problem. If you want to stick with that design, it may be something that you have to deal with though.
Also, the reason 8059A’s robot had zipties around the wheel was to keep the rubber casing on, it didn’t add any functionality 
We had the same problem as 8059A, the wheel was literally expanding off of the plastic insert, so we had to put zip-ties on the wheel just like they did. The reason I don’t want to put the flywheel on two different batteries, is because one battery could drain quicker then the other causing one side of the flywheel to have less power flowing to it. 8059A stated that they run their flywheel somewhere around 80-90% I believe, so we decided to run our flywheel at 90% as well.
Have you checked your battery at all. Have you tried replacing it? Cause I know I’ve had batteries that go bad like a week out of the box and can’t hold a charge for too long.
First of all, since you have several batteries, check that they behave identically by logging voltage level every 1-2 sec for 2 min, while running flywheel at 90%, 50%, and not running it at all. Plot the results.
If all batteries behave about the same then it either motors or flywheel assembly. To check if motors have too much friction take them off and test.
First run each motor without any load and measure its angular velocity, preferably with IME. Then connect two motors with a shaft and drive second unconnected motor with the first one. Drop in the angular velocity will indicate amount of viscous friction in the second motor. Repeat it for all motors. If there is a problem with one of them you will easily see it.
To test for a problem with flywheel assembly see if voltage plots differ when flywheel idles vs when you launch balls every few seconds. If idle voltage holds, but launching drains battery fast then you may be compressing balls too much, too long, or you have some problem with flywheel bearing.
If I had to make one guess, I would point at the flywheel bearing. Its axle runs at the highest speed of all other geartrain parts and any imperfection there would cause significant power losses.
For example, we used to have a lot of problems with single flywheel prototype because of the improper bearing alignment. At the beginning it used to take two force motors to spin up the idle flywheel. After properly realigning left and right supports and securing them with more standoffs we got it down to 1 force motor, but single speed motor still couldn’t move it. There was just too much friction somewhere.
The trick that technic jr came up with, was to tighten screws holding flywheel bearings, while the flywheel was running! Now it could be spun up with just one speed motor. The next task is to see if we could remove even more friction to use a single turbo motor.
The task after that is to get back plate right. After that we will add two or three more motors to achieve fast recovery.
Don’t forget that the more motors you have the more friction losses they will cause. You need just enough motors to achieve recovery time so that it matches the speed of the intake. There is no need to recover in 1 second if intake delivers one ball every 5 seconds at best.
Maybe you could try using a PID (or PD or P) loop, I mean at least that will adjust your speed based on your flywheel speed, so if you have a lower battery voltage, it will not effect the ability for your flywheel to function (to an extent).
That is still very fast for your battery to be draining. Are you sure your batteries are up to standard, it sounds like you could be using dud’s (or similar).
This highlights the wider issue that it’s a really good idea to label batteries with their ‘birthdate’. It gives a good indication about when a battery should be retired. They don’t last forever, and no-one wants an expired battery while competing.
Hi Aiden,
Think you have misinterpreted our specs.
The power setting is between 80 to 100 (when the battery level is lowered) out of 127.
And that means it is running at about 63% to 79%.
And we do not keep the motor spinning when it is not shooting any balls.
All these will drain your batteries.
We do experience batteries being drained faster than any other seasons’ games.
But definitely not as bad as what 8931 is experiencing.
So no - sticking to this design need not necessary means you will have to live with this issue. And there are also many other teams that made use of 8059A design (or variation of it) and they do not have as big an issue (of battery draining) as 8931.
It is definitely not a design issue.
PS… you guys are doing lifting?
Thank you guys for all of the help. I have purchased 4 new batteries for this season ad have only used them several times, so they shouldn’t be bad, and we use all of them back and forth and they all act relatively the same way, so I don’t think the batteries are bad. We decided to have the flywheel run at 30% when not shooting, we did not take that from 8059A. The reason we leave it constantly running is because it takes a lot of power to initially startup the flywheel, so if we keep it constantly running but at a low speed, we don’t loose momentum but we save power; by stopping it and starting it over and over, it causes lots of damage to the motors. I had remembered 8059A saying something about running somewhere between 80-90% but I guess I was a little off on that…sorry. 90% seems to work for us, so we will roll with that for now. Yes we are attempting lifting with a rack and pinion. We can lift ourselves 15" off the ground, and a robot can drive underneath us, but the catch is, is that the robot that drives under us has to be 13" wide or less. (We are still working on it).
How would I go about programming a PID loop?
I definitely know that 30% is not from 8059A. Didn’t recall my guys doing anything like that :rolleyes:
2 tips that I can offer you:
-
there are ways to startup the flywheel from rest every time and it will still not damage the motors. Eg. Speed gradient. Let the motor spin up gradually and the stress won’t be that high for the motors, and at the same time you can save some juice from the batteries.
-
if you have time, check through all the friction, etc. In the long run, all these unnecessary friction will cause you even more problems.
Running at 90% for prolong period is definitely not ideal.
If yours is a similar setup to ours, then it shouldn’t be too far off in terms of the power required.
Hope all these help.
Well you have a couple of options here, firstly you could use the tried and tested ‘jpearman velocity controller’ which uses TBH (you can find the simplified version).
In terms of PID controllers I reccomend you have a look at 1727B’s video on PID loops on flywheels. I recently wrote a PD loop (which is honestly all you really need) on a flywheel velocity error thread. (My Post)
In case your wondering, simple PID is: (k denotes a constant)
Proportional = error * Kp
totalError += error
Intergral = totalError * Ki
Derivative = (error - lastError) * Kd
Good luck, feel free to ask if you have any questions (if I can’t answer it someone else will :p)
Thank you all for the help, I really appreciate it!
I don’t know if you misunderstood what I said, or were agreeing with what I said, but I said:
We have gone through and eliminated as much friction as possible. (There is the possibility though that we missed something).
I will try the speed gradient, but the problem I see with that, is the efficiency going down because it takes a while first of all to slow down the motor and then second to start it backup to full speed,(loose a lot of time when you can shoot the ball). I’ll go ahead and program it anyway as well as the PID loop and test to see if it helps our efficiency, got nothing to loose!
Thanks again for the suggestions! You all are a great help!
No… I didn’t misunderstood what you said or missed out on what you said.
I was agreeing with you that my team didn’t do that 30% thingy.
But I was also at the same time trying to tell you that keeping your motors running at 30% even when you don’t need to do it will definitely eats up your batteries.
And I was also trying to hint to you that you don’t need to do that to improve on your efficiency.
But do what you think is best for your robot
It might be benificial to find how many volts your battery has. This thread (https://vexforum.com/t/battery-voltage/22402/1) explains battery methods that you could send to a speaker or lcd.
I have an LCD set up to show us our battery voltage already.
Oh, ok, I understand now. So, basically it doesn’t matter if it is running at 100% or 30%, it will drain the battery pretty much the same. I will try to program the speed gradient and PID loop. Thank you so much for the help. Sorry for misunderstanding a lot of things. You all are a great help.