Transconductance opamp set up as "Voltage follower"?....Boost PFC battery charge

[ocset]

Hello,
We have special batteries which are pretty much always 400V +/-5V no matter what  the state of charge.
We want to charge them with  a  non isolated  “controlled  open loop” BCM Boost PFC converter as attached. (we’ll be careful). PDF schem and LTspice sim as attached.

This involves setting up the internal transconductance error amplifier in an LT3799 (Boost BCM PFC)  IC  as a simple buffer as shown...
Then one simply adjusts the charge current by adjusting the voltage into the CTRL3 pin of the LT3799 controller as  shown. (“iterative set and check algorithm”)

Can you see any problems with this? That is, the connection of the internal transconductance opamp as a "voltage follower"  as shown.

Datasheet:   LT3799 BCM Boost PFC controller IC
https://www.analog.com/en/products/lt3799.html

[NiHaoMike]

You still want voltage feedback as a failsafe mechanism to stop charging. Actually, you'll want several of those mechanisms since such a large battery can fail in a very dramatic way.

[ocset]

Thanks, yes...we will have a FET switch in series with that battery (not shown in top post) , and if its voltage goes too high, we'll switch it out.

Do you think maybe we'll need to connect inputs and output of the OTA together, then apply the control voltage to that? (instead of the way we've done it in the top post)?
After all, if inputs of OTA are the same, then its output current is zero...and so its output impedance must then be highest, which is probably  good for our situation where we want to "Inflict" an external voltage onto the OTA output? (in order to control the power throughput of the Boost PFC stage)

[David Hess]

It should work but I suspect there are better ways albeit probably not as inexpensive.

Thanks, yes...we will have a FET switch in series with that battery (not shown in top post) , and if its voltage goes too high, we'll switch it out.

That is going to be a problem.  Protection against open circuit faults require the output to conduct 2% of the output current which is done with some type of shunt, like a zener diode.

Do you think maybe we'll need to connect inputs and output of the OTA together, then apply the control voltage to that? (instead of the way we've done it in the top post)?
After all, if inputs of OTA are the same, then its output current is zero...and so its output impedance must then be highest, which is probably  good for our situation where we want to "Inflict" an external voltage onto the OTA output? (in order to control the power throughput of the Boost PFC stage)

The application notes show the control inputs connected to the positive reference to disable them but I do not see where it matters.