Mastering

1 yes they are probably off
2 no there is no way to restore mastering if you lose battery power to the pulse coder.

Mastering is based on an offset applied to a known position. That is what your pulse coder counts are for. They were created at a certain position and then recorded. When you lose battery power the pulse coders reference to that position is gone. It is stored in a chip and it is backed up by a battery. The pulse coder data that is in the CPU is only for reference, the pulse coder is the master.
If you set up quick master reference you can restore the mastering but you may have to touch up points anyways.

the 2 most common service calls I had were for batteries gone dead and loss of programming and the customer had no backups.

Replace your batteries annually, and don't shut the robot down. The power consumption is minimal if the robot is idle.
Back up your robot. once a month or anytime something is changed.

Hi,

A helpful hint is when you replace the batteries write the date you changed them on at least one of the batteries in each battery pack.
Also don't forget the batteries for any auxiliary items they may have them.

Cheers,
Geoff.

Press Menu, 5, (I/O), F1(TYPE), SOP, Scroll down to Battery output and cycle it on, If in a few moments if the output is turned off, the batteries are still good. if not replace them!

Fanuc robot only gives you a warning when the battery is low. No one hardly noticed it.
A few years ago, I started to check the battery low bit at the begining of every production cycle. Halt the program and remind the maintenance about the battery.
For those arms, I have never got a singel dead battery call.

I know this is an old thread but the content is most relevant to my post which is still an active topic.

Restore factory mastering after BZAL

I have seen a lot of questions and answers about this topic but none of the ones i found actually explain how it works. I've run into lost mastering many times on Fanuc robots over the years but never took the time to figure out if it was possible to restore back to factory master counts until recently i had the opportunity with a new robot. Normally remastering on a commissioned robot is no big deal and all the programs will work as expected with the new mastering, however the counts probably won't be equal to what is on the master count sheet supplied with the robot, and you should scratch out these numbers replacing them with the new master counts for the next guy. This may be fine for most, but if you're anal about it like me, you want the factory numbers. It turns out there IS a way to recover the factory master counts and zero positions. This procedure only works if there have been no mechanical changes to the motors such as decoupling from the reducer input, basically losing battery power while the controller is off. The key is establishing the new pulse count in the same location as it were when first established at the factory (after being coupled to the reducer and powered on for the first time). The difficulty with the pulse counts is that after power loss, you have to turn the shaft a certain amount before finding a sync pulse that wakes up the counter and starts counting from zero. This sync pulse occurs once per motor revolution and many times over the full traversal of the mechanical unit, but it is always occuring at the same positions. Understanding these actions allowed me to develop a procedure to find the correct sync pulse positions and get my robot back to factory master count zero. So without further delay, heres how to do it.

Assumed starting point: Robot was powered off when battery power was lost resulting in a BZAL alarm after powering back on. Reset pulse code alarm from Master/Cal screen then cycle power from function menu. Upon reboot you should have pulse not established alarm from axis 6 through axis 1 respectively.

Step 1 - Move all axes to mechanical zero:
You are probably already familiar with this step, just jog the axes to the alignment position as close as possible. Precision here is not really required unless a sync pulse location happens to be right in the way, otherwise if you are within a half a degree (less than 1 motor revolution), it should be ok. If your robot was already at or close to zero alignment when the pulse count was lost then u will need to jog each axis to establish a pulse count first, then back to zero alignment.

Step 2 - Zero master counts:
Set dmr_grp[1].master_done to false
Set all dmr_grp[1].master_count to zero
Set dmr_grp[1].master_done to true
Calibrate robot
The main reason for this step is to get the joint angles to zero so u dont hit an axis limit while looking for the sync pulse.

Step 3 - Forget the pulse count:
In order to find a new pulse, we need to forget the current ones. Thats right, the big no no - power off the controller and remove battery power by lifting the battery cover briefly. A quick second of battery loss is enough to do it. Replace the cover but just leave it slightly loose because you may have to do this several more times. Power on the controller, reset pulse code alarm (BZAL) and cycle power from the function menu.

Step 4 - Find the location of a sync pulse for each axis:
At this point, the pulse coders are what i call sleeping, waiting for a wakeup signal - the sync pulse. We want to find a system variable that shows pulse counts, dmr_grp[1].spc_count. Find the master count for axis 6 on the master sheet and note the polarity of the number. If its positive then the master sync pulse resides on the negative side of the alignment marks and vice versa. My robot had master count -41417414 for axis 6 which means the sync pulse is on the positive side. Using my robot as an example, begin jogging positive at 1% while watching spc_count. As you jog axis 6 positive, spc_count remains unchanged only showing the last count value from before the memory was lost. Until at some random moment, the number pops to life and starts counting quickly from zero. It will probably reach 10 or 20 thousand before you can react - thats precision! Take note that jogging some axes positive may move the pulse counts negative, not all robots are the same. Also, since axis 6 is continuous but not direct driven, the difference in the reducer will offset the pulse counts more and more for each rotation, so hopefully it hasn't been jogged way out of position or else the correct sync pulse may be next to impossible to find.

So now you have found a sync pulse, continue jogging until spc_count is equal to the master count from the sheet, then go beyond another 100-200 counts. The servo will oscillate to maintain position so the number will continue to bounce around a little bit. This is just one sync pulse location for axis 6, now that you know what to look for, do the same with all axes. Put a piece of tape at the zero alignment mark in this position for each axis so you have a reference if you need to retry.

Step 5 - Forget the pulse count again:

The robot is now in the approximate position as it were when first powered on at the factory. Power off the controller and remove battery power by lifting the battery cover briefly then replace it. Power on the controller, reset pulse code alarm (BZAL) and cycle power from the function menu.

Step 6 - Locate the correct sync pulse:
Starting at axis 6, begin slowly jogging towards zero alignment while watching spc_count. Your pulse count should wake up quite quickly, this should be the real master pulse. Continue jogging to the alignment and when u reach it, spc_count should be very close to your master count from the sheet. Do this for all axes in descending order 6-1.

Step 7 - Set mastering:
Set dmr_grp[1].master_done to false
Set all dmr_grp[1].master_count to master sheet data
Set dmr_grp[1].master_done to true
Calibrate robot
Create a zero joint position and run the robot to it, you should be at perfect zero alignment.

Step 8 – Wash, Rinse, Repeat:
You will likely have to repeat this process for each axis a few times because the master sync pulse might be just on the other side of where you started jogging from so next time jog in the opposite direction to establish the pulse, then jog to the master count to see if the zero marks align. In my case, I found axis 1,4,5 master pulse the very first pulse away from aligned at zero. Axis 2 & 3 were tricky at first because they are coordinated, find axis 3 master pulse before axis 2. Start at axis 6 and work in order down to axis 1.

I am writing this from memory so let me know if this doesn’t work or if there is a missing step, I will revise. Good luck!

Hi
On Step 3, to have it clear Do you want us to do the big no-no ? Do you want us to turn the robot power off and then remove the batteries ? I just want to be sure i did bot misinterpret.

Anyway, I don't have any robots with those problems right now, but if it happens, I would follow your instructions. Seems to me that you suffered enough that you learnt how to solve the problem

Correct. At step 3 you will power off the controller and then remove battery power.

I tried the whole process and worked fine. Just a tip to make the process much easier:

Instead of removing the battery cover, you just have to disconnect the encoder at the motor side, and this way you are not losing the achieved correct sync pulses