Battery swap on Mazak QT10N T2 control

[Nxt-1]

I wanted to share my experience changeing the backup batteries of a Mazak QT10N with T2 controller.

This process started with the "23 NMI Battery Alarm" shown on the machine. This did not come as a big surprise as the batteries have a 10-ish year lifetime in my experience (we have had the machine since '85). Instead of forking out the €1000 for a Mazak tech to come out and swap the batteries out I decided to do it myself and make some QOL changes as well.

Some preface on the batteries. In the back electrical cabinet of the machine, behind the white/grey plastic cover containing the computer area, there are 3 cards containing the machine's memory chips. These are volatile memory. Meaning that if they lose voltage, they are wiped. That is why they are backed up by a battery pack that provides power if the machine is turned off. Each of the cards has one set of batteries. Historically these were 3.6V 200mAh NiCd pack, but over the years they have been replaced NiMH cells.
While the NiMH cells are still available for purchase (type V200H-3ME), they are not very easy to access and replacing them can be troublesome. When the machine was new a replacement was trivial. First take a backup of all parameters with the Mazak Microdisk system (floppy style disk) for good measure and then simply swap the batteries. A small capacitor on each board would provide a short amount of power to bridge the gap when swapping batteries. These days, the microdisk system is obsolete for longer than I am years old and the capacitor is still the same it was in '85, with an unknown capacity.

With my engineering pants on, I decided it would be better to:

1. Physically move the batteries away from the pcb's for better access
2. Use connectors to facilitate easy swapping of batterypacks
3. Install a 2nd set of wires in parallel so you can hotswap the batteries without needing to rely on the capacitor's health
4. Use much larger cells to be able to leave the machine off for longer than 2 weeks without the batteries being dead, wiping the machine

I decided on a 3D-printed holder, containing 3 standard C-format NiMH cells. This gives me the same voltage and battery chemistry to ensure compatibilty, yet over 10 times the capacity. After very carefully hotswapping the old batteries for the new ones (I did not have a backup sadly) I am pleased with the result. I still have to do wiremanagement and tidy it all up, so forgive the tape.

So far, the machine works without issue though I still have to test what the actual autonomy of the batteries is at this point.
Hope this can help someone.

- Nxt-1

 

[Mr. Atoz]

Pretty slick, I must say.

 

[Nxt-1]

I got a PM from @tmiller2013 asking for an update.

I've had 0 issues the last month now. Though I've been using the machine semi-regularly the past few weeks.
A question was asked with regards to charge time of the battery. Details are very very scarce, but the charging circuit should be a constant current sournce. Meaning that if the batteries double in capacity, the charge time double as well. So from fully empty to complely full it would likely take multiple days with the much increased side I use now.

Happy to help
- Nxt-1

 

[MelaniePayne]

Your experience is very useful, and I appreciate the detailed explanation. I'll definitely take note of the tips and solutions you've provided for the Mazak QT10N with the T2 controller. It's great to see the creative engineering approach you've taken to address the battery issue. On a related note, I recently had a successful experience using lifepo4 cylindrical cells for my electric car. This upgrade significantly improved its performance and power, and I thought it might be useful to share with the community. It's amazing how these battery innovations can make a big difference in various applications.