The controller is working as intended for you. On default settings the containment field stabilizes around 20%. I WILL make a youtube video because many ppl have trouble setting it up correctly it seems. Anyways - your setup seems to work fine.
Do NOT worry about the temperature as long as it is not above 9000. The default settings are going to throttle the reactors output between 8000 -> 9000 C (= it's adjusting the target saturation from 50% to 90%) between 9000 and 9900 the output rate will be reduced from 100% to 0%. Anything above 9900 will trigger an immediate shutdown (although it should do so before that point is reached).
Do NOT worry about the containment field unless it either starts dropping rapidly when it's well below 20% on default settings or it gets lower than 8% (which should NEVER EVER happen on default). Anything above 0 is perfectly fine.
Why is the controller keeping the containment field low?
To keep the containment field at higher levels, you're required to pump in more RF per tick. But since anything above 0% is fine, we can save a lot of energy by having the containment field low. How much energy you ask? Keeping the containment field at 50% cost 50% more energy than keeping it at 25%. 25% costs 16% more energy than 12.5%. You might've guessed it: The formular is: <somevalue> / (1 - field_percent).
If you go to the preset page and use my special settings, the containment will go to roughly 9.09%, dropping somewhere to 8.xx% if you cut the load and suddenly put load on it again. I'm running these settings on a server on...like...11 (?) reactors now.
Do not worry!
If your reactor is setup correctly, it is extremely hard to make it explode by accident. I blew up 4 reactors on my server:
3 blew up simultaineously because of dimension loading issues (do not put your reactor in a different dimension than your energy storage. Do NOT do it!)
1 blew up because I accidentally put a flux point directly on a stabilizer - with unlimited throughput. INSTA FIREWORKS!
I've really tried everything (= could occur during normal operation) to blow them up - including cutting the load and reapplying it when the reactors were nearing their shutdown sequence (which is the most dangerous stage). Everything is still running.
Last but not least some explaination:
The jump in the containment field you see is because the controller did not yet run again and reset the feedback gate yet. It's still running at 1M rf/t for a few ticks, thus the surge in containment field strength you saw.
Actually "running" the reactor at 2000 is propably impossible. I acidentally cut the load to one of the reactors entirely so it went to 100%ish saturation with 0rf/t output - and temps did not go below 2150 C (at ~11% fuel conversion).
