Casio CZ-1 battery drain

Hi all,
I am working on a Casio CZ-1 that is draining the backup battery too quickly.
A fresh battery will drain in a matter of hours, forcing constant reinitializing, loss of all settings and loss of customers mind.(and mine :cry:)
The keyboard powers up and is otherwise in working condition.
What would cause the excessive drainage of this battery?
I removed IC3 and it slowed down, but still loosing .001VDC every couple of seconds, also replaced 100uf/6.3V cap (D1, 1.5K bottom line), removed VCC/VOO header from PSU.
no shorts and diodes/resistors all test good in circuit.
Could it be any of the transistors?
I have never experienced this issue before! anyone else?
Thanks
Mateo

All that energy is going somewhere, are any components getting unusually warm/hot?

No hot chips or components.
I have been monitoring drain since this post.
In this circuit, I have changed out the TLO82, all electrolytics, and passive test again on all resistors/diodes/transistors, no shorts or opens.
Seems to me the circuit is not being told to charge a cap from the battery to keep the 3V for the RAM chip. If that is how a backup battery circuit like this works, i don’t know.
When powered on 5V is sent to VBR and when turned off it reads 2.8V without battery, it will slowly drain in about an hour or so.
When battery installed, power turned on, again 5V at VBR and when turned off it reads 2.8V. It will take a couple of days to drain in this scenario but it will drain to nothing.
Remove battery and it will read 3V!
Don’t believe its a battery drain issue anymore.
I don’t know how these types of circuits work so i’m puzzled. Im thinking more and more it’s a cold solder but would like to verify if my ideas on how this works are correct so I can go searching for the cold solder or something else!
The thing works fine otherwise.
Thanks
MAteo

Update
I let the voltage drain overnight to nothing, no battery connected. I installed battery, 3.1V, and measure at VBR, which feeds the RAM, Voltage reading of average of 1.775V. voltage swinging between 1.785 to 1.745 constantly. Would this be considered right?
Disconnect battery and voltage goes to 0.
Is it the RAM chip? It holds memory when power is on.
Negative side of battery is not connected to ground. Reading of 100 ohms between. Saw a cz-1 video on you tube where the person beeped to ground? If I connect to ground, display is non-repsonsive.
Thanks

D1 blocks the battery from draining when the main power is to work back from that point. that leaves the capacitor, the resistor and the switch.

Check the switch or connector for contamination or faulty connections.
If that didn’t work, remove the capacitor (16V1c something something and D1) and see which part of the circuit is drawing most power. Transistors are easy enough to test with a multi meter, but may need to be removed form circuit.

Its rare for a ram chip to act as a current sink, generally they just fail to work. So that just leaves another hidden component such as a capacitor somewhere on the ram chip circuit that could be leaking. Perhaps its on another part of the schematic, or perhaps its an addition not on the original schematic, but an afterthought modification.