#390
Attachments to CAT Chat Minutes - March 15, 2004
W. Wesolowski

reg wright consultancy
attach
1, Severn Drive, NEWPORT PAGNELL, Bucks. MK16 9BU
Tel: 01908-611617 Fax: 01908-618255 e-mail:   reg@wrights.f9.co.uk

REPORT ON INVESTIGATION INTO AUTO-RESTART PROBLEM

The necessary parts were purchased from RS Components and a mock-up of the Auto-Restart circuit built. An additional relay was also connected to the circuit to simulate the action of the main contactor. On testing, the circuit was found to operate reliably over more than 100 test cycles. Different sequences of mains failure were tried, with efforts concentrating on a 2-failure sequence of short duration and interval. Eventually it was found that if 2 failures, of less than 3 seconds each, occurred within 3 seconds of each other then the circuit failed to restart.

The cause of this failure was investigated and found it to be due to the hold-up time of the 24-volt power supply. The 3 second single shot timer should be triggered, each time the power is restored, by the rising edge of the of the 24-volt supply, therefore, if a mains failure is too short for the 24-volt supply to fully decay, on restoring power, the rising edge can be of too small an amplitude to trigger the timer. In these tests, the power supply used was very lightly loaded, taking several seconds to decay, thus exaggerating the problem. Accurate figures were not available for the hold-up time and loading of the 24-volt power supply used in the Cryo-Coolers or the actual power-out interval between a mains failure and the stand-by generator kicking in.   I can therefore only surmise that this is the most likely cause of the problem, other than a wiring error. I personally tested all of the coolers fitted with auto-restart and found them to be working reliably but did not test them for short duration drop-outs.

The problem is further complicated by the fact that 2 of the coolers exhibiting this problem are at SBC­CAT and UNI-CAT both of which are "B" type coolers.   "B" type coolers have a 24-volt AC contactor, rather than the 24-volt DC one used in the "C" types, and therefore do not have a 24-volt DC supply.   As the auto-restart circuit requires a 24-volt DC supply to energise it's relays, when it is fitted to a "B"type coolers a rectifier diode and capacitor is connected to the 24-volt AC supply to provide the DC for the relays. In this case, the capacitor and load determined the hold up time, but unfortunately there is no record of the capacitor value used. This difference could explain why the problem seems to affect "B" types more than the others.

Assuming that the hold-up time of the 24-volt supply is the cause of the problem then the solution is simple.   The trigger input to the timers can be either AC or DC, so by connecting the trigger input of the single shot timer to an AC source at, a point before the mains contactor, the hold-up time problem is removed. This change can be easily achieved, in the case of "B" type coolers, by removing the 10k resistor connected to pin 3 of the single-shot timer and connecting pin 3 instead to the AC side of the rectifier diode.   It may also be advantageous to change the single shot time from 3 seconds to 1 second.

In "C" type coolers the solution is more difficult as there is no suitable AC supply available in the unit containing the auto-restart circuit. The AC supplies in this unit are all after the mains contactor.   A possible solution would be to make use of one of the spare ways, available in the 7-way Harting connector, to bring in an AC source from the mains input to the 24-volt power supply.   This would need to be connected to pin 2 on the single shot timer and the 10k resistor connected to pin 3 removed, leaving pin 3 unconnected.

I suggest that this change is first tried on the "B" type coolers as 2 of them are exhibiting the fault and the modification is a lot simpler.   If this proves successful then an test should be made on a "C" type to determine if they exhibit the same problem. If so, the modification could be simply tested by using a flying lead to the AC supply, with modification kits then prepared for a more permanent solution.

In the event that hold-up time proves not to be the cause then an on-site investigation would be necessary.

REG WRIGHT

11th October 2002

"B" Type Circuit Showing Additional Rectifier and Capacitor

attach

 

"B" Type Circuit With AC Trigger Modification

attach

 

"C" Type Circuit

attach

 

"C" Type Circuit With AC Trigger Modification

attach