Salih,
You obviously haven't considered your "automatic sequence" very well.
I'll take a wild guess... after starting the first pump, your "automatic sequence" simply looks to "see" if the previous motor in the sequence is ON before starting the next pump.
The result of that type of sequence is that all 10 pumps start, or rather, try to start within a second or so. That is, even if you are waiting to "see" the "AUX" contact of a previous pump come ON before you try to start the next pump, all of the pumps start in sequence, very quickly, until the Main drops out.
If this is the case, then your "so-called" automatic sequence is inadequate. The design of the sequence needs to provide the delay.
I can't remember the AB style sequencer, but in TI, there are two types of sequencers. One is Event Only and the other is Event/Timer.
The Event Only type simply looks for signals. If you are using a sequencer, this sounds like the one you are trying to use.
The Event/Timer type looks for signals and then applies programmed delays before moving to the next step in the sequence. That is, Step-2 in the sequencer looks for Pump-1 being ON and then the sequencer applies the delay, programmed for Step-2, before moving onto Step-3.
If AB does not provide this type of sequencer (I wouldn't be surprised) then you need to run your "AUX" signals through an ON-Delay Timer. Then, in your sequencer, monitor the output of the ON-Delay Timer instead of the "AUX" contact itself.
If you are not using a "sequencer" then your "automatic sequence" needs to include a timer for each of the "AUX" contacts (except for, maybe, the last one... depends).
In any case, you should not start your pumps using a simple timer/counter combination nor an elapsed time scheme.
Any time you attempt to start a pump, if the pump does not start within a particular short period of time then your program needs to provide a Fault Indication.
Simply using a timer/counter combination, or an expired time scheme is "minimalist thinking". There are too many things that can go wrong.
I can not think of one situation where those methods are reasonable or adequate (unless, of course, you have complete faith and "KNOW" that the system will respond as you require... even if that damned #3-Pump is on fire!)
BTW, Mickey,
Since your timer had to have some start-value anyway, Rube was suggesting using a timer start-value that covered the entire start-up process - without the additional complication of a counter.
In many situations (not this one) that is an excellent solution to a time-sensitive sequence. The primary benefit is that each time-out situation can be individually tuned. That is, you can vary the sequence by varying the timeout values! That is BIG JU-JU!