So each player has 90 minutes to complete 40 moves then a sudden death for 30 minutes. Question: is the 30 minutes after 40th move added onto any remaining time a player has or do both clocks get set to 30 minutes for the completion of the game?
Thanks
The former.
Alex Relyea
US Chess rated tournaments are also supposed to specify whether there is an increment or delay component to the time control for the event.
For example: 40/90,SD/30;d5 would mean that there are five seconds of delay at each move before a player’s clock starts counting down.
40/90,SD/30;+5 would mean that five seconds are added to a player’s clock at the beginning of each move, and then the clock starts counting down.
The difference between delay and increment may not seem like much, but the increment can add up if one plays a series of quick moves. In serious chess tournaments a 30 second increment is often used, which means that after 10 moves each player will have received an additional five minutes of time.
If neither increment nor delay is to be used, the time control should make that clear.
For example: 40/90,SD/30;d0 would mean neither delay nor increment was in effect.
The increment, delay or lack thereof in effect can change what rules apply in time pressure situations.
The 30 minutes gets added to the player’s remaining time.
The details depend on the clock brand and the setting used.
With some clocks, and in some modes on other clocks, the 30 minutes is added as soon as the clock thinks the player has completed his 40th move. I say “the clock thinks” because it’s possible the clock’s move counter may be incorrect. If, for example, one player forgot to press his clock at some point in the game, and the other player went ahead and made his next move without waiting for the first player to press his clock, then the move count will be off by 1 move.
On some other clocks, or in some other modes, the 30 minutes is added only when the initial 90 minutes is used up, regardless of the move count. In the long run there is no difference which way the clock does it (the total time will end up the same either way) as long as the player did not overstep the time control at move 40.
The time control should have been specified with an increment or delay, for example 40/90 SD/30; inc/30 or 40/90 SD/30; d/5. If there is to be neither increment nor delay, this should have been spelled out, for example 40/90 SD/30; d/0. If neither increment nor delay is specified, and “d/0” is omitted, the organizer might get a spanking 
from U.S. Chess, and the players might have the right to demand that a 5-second delay (d/5) be used, even though no delay was mentioned in pre-event publicity.
Bill Smythe
The time control actually must make that clear. The failure to specify a delay, increment, or the lack thereof means that a default delay is to be imposed, by rule.
Correct. But 40/90,SD/30, without more specified, is 40/90,SD/30;d5 if the rule is correctly applied.
Almost correct. Removing the second instance of the word “might” makes it fully correct.
Personally, I abhor the continued existence of ‘default’ increment or delay settings. It’s the organizer’s responsibility to decide what the time control will be, unambiguously and without any implicit information.
An organizer who can’t or won’t do that shouldn’t be running tournaments.
US Chess failed to do this for a handful of national events since the delegates blew up the longstanding default so that the organizers of the soon-after defunct Four Rated Games Tonight could let players choose between G/30 d0 or G/25 d5.
Inertia is real, and US Chess’s own failures to comply with this rule demonstrate that. When players or organizers who’ve not been active in a while return to the game, others ought not be forced into a d0 situation because the delegates unwisely made a fundamental change for a to acommodate a tournament that was soon dead anyway.
OK, but what happens when the organizer doesn’t do that? That’s when the “default” becomes necessary.
And currently, the default is the minimum recommended delay – 5 seconds regular, 3 seconds quick, 2 seconds blitz. That’s as unobtrusive as it can possibly be.
An understandable attitude, but one that (by itself) doesn’t solve the problem of how to square things with the players when it happens.
Strongly agreed, but –
I disagree with “unwisely”, and with the characterization that this change was made for a single organizer. The idea had a lot of support.
Bill Smythe
I looked up the FIDE standard time control. It’s a split time control: 90 minutes for 40 moves, then 30 min +30 seconds SD.
Been years since I looked up the FIDE standard time control. I was curious, since the OP was asking about a similar time control.