EBME Forums Forums Maintenance Technical Assistance Isolated power systems in Group 2 locations

Font

I am afraid the answer to your interesting 'mind-teaser" query about a stack or, for that matter, any piece of medical equipment that has a leakage current higher than that allowed, has to remain a definite "no it can not". Even if on the IPS, and it is almost certain to be so, it reads well below any danger level of leakage current without the isolation transformer present. Immediately though, there is one obvious very good reason why it should not be allowed to be a 'yes' anyway, and that is purely due to the possibility that the stack could be wheeled/moved into another area that is not isolated, and then it would be in violation of safety standards if the Biomeds were not informed in time to adapt it, or worse, not informed at all.

However, there is a more subtle scenario that needs to be examined. The answer lies in a scenario where there are two different pieces of equipment in an IPS area, each of which has a single-fault condition. (Safety limits for all equipment being based only upon a single fault condition occurring in that piece of equipment under test). Let us say that one piece of equipment in the IPS area has accidentally got one of it’s internal main input power lines (let us say L1), shorted to the chassis due say to insulation breakdown. The important thing to recognize here is that now all the equipment in this IPS area is no longer isolated from the secondary side of the isolation transformer i.e. the two power lines L1 & L2 (live and neutral in a non-isolated system) are now referenced to the area earth and the IPS system has reverted to a non-isolated system.

At this point, this single-fault condition would only be a minor problem as having one power line connected to the area earth is, of course, just like having a neutral connected to the earth. In this case, the only thing that would happen is that the isolation monitor should indicate, by an audible alarm, that it has detected a lower-than-normal impedance between one line and earth. This would not cause the power to be tripped but just be a warning to the operators that something was wrong in the system and it needs checking out ASAP.

(If we were to continue to follow this further, it would become obvious that there now is a potentially very dangerous situation if this insulation break-down is not discovered soon. If the other conductor (L2), either in the same, or any other, piece of equipment in the IPS area has a similar case of insulation break-down, consequently shorting to it's earthed chassis, and since all machine earths are still present (and therefore connected together) in and IPS system, there will be a major line-to-line short and a CB should trip, of course. This is one of the minus marks where IPS systems are concerned).

However, let’s now assume that the staff have heard the alarm produced by L1 in the faulty piece of equipment but decided to mute it and carry on with their procedure, and combine this single-fault scenario on the one machine above with your stack (minus transformer) that we will say has a single-fault condition whereby the main power feeder cable has it’s earth-wire detached inside the mains plug going into the wall socket. (This could be due to the staff always pulling the plug out of a socket by the cable alone). Now attach the stack system to a patient on say an operating table. Due to the excessive (over-the-limit) earth-leakage current in your transformerless stack, we now have a potentially dangerous situation waiting to be exposed. If say the patient’s arm or hand touches a grounded point through say the frame of an earthed bed, or chassis of another piece of equipment nearby to the patient, all the leakage-current from your now non-earthed stack has the potential to pass through the patient to earth via any of these points. Since the system earth is now connected indirectly (via L1 in the other piece of faulty equipment) to the secondary side of the IPS transformer the likely results are now no different to those that would have occurred if this had been a scenario on a non-isolated system, i.e. if it is a cardiac patient on the table then your stack might be mainly responsible for an RIP situation?
Hope this clarifies things.

As to the second part of your query, could you please clarify your question as it is not clear as to why you should wish to prevent the use of stacks in either situation and which type of supply in particular you wish to avoid using the stack ie non-isolated or ISP.

I have worked for over ten years with camera systems and I am glad someone has highlighted this problem .I have found that all makes of medical grade CRT monitors and all sizes (21 & 14 ) causec this problem on CP systems and the problem gets worse as the monitors get older but they all pass EST I have got both Sony and Panisonic involved and they confirmed that there was nothing wrong with their monitors .But nobody could tell me what the leakage sensors should be set at and should they be set with all the equipment in the CP room or not .The awnser I always got was put an isolating transformer on the camera system but to me this is only hidding the problem and not finding the reason also it seems silly to have a VERY expensive CP system and then use isolating traffos on the camera systems .Some people suggested that there could be and resistor beteew the monitors mains traffo and earth and this increase in value as it ages!!!!!!!! anohe reply is CRT monitors are old hat and flat sreens are the new thing but tis dose not awnser the ? what is the value settings for the leakage sensors

Hi Nick,

Appreciate your long answer! My advice before and after introduction of these IPS supplies has been to fit an ISO TX. I was wondering whether anyone had fitted dedicated style plugs/sockets to stack systems/IPS supplies so they could only be plugged into the IPS supply so there would be no need to fit an extra ISO TX. We all know staff like to move these systems around so it would not be practical.

We are going to look at the ISO TX's supplied by Monmouth Surgical that have inrush limiting and integral IEC sockets to see if it improves the situation.

Font, your suggestion about the use of a dedicated plug is a good one.
However, if you sort out the issue of the in-rush current either with a transformer designed to reduce this or, alternatively, replace the CB in the panel (at the source of the IPS power–circuits) relating to the wall-socket where your stack will be connected, with a dedicated less sensitive CB (a much cheaper option if technically agreed to by your hospital electrical engineers), then you will not need to fit a dedicated plug and so allowing the equipment to still leave the IPS area if need be.

John, the contributions from this forum seem to show categorically that in the latest IPS system requirements, both in-rush currents and unexpected DC currents are candidates for mischief-making with the detector circuits. If your displays are giving a dc current into the mains/earth line for whatever technical reason (Meg resistors to ground etc!), then If your IPS system does not use say a (high-frequency?) pulsed-dc to monitor the insulation-to-earth of the mains wires, you will have to isolate this with a transformer to block the dc component. Blocking the external pathway for the display’s dc component will in no way affect the well-being of the display itself but if not blocked is obviously a no-go for safety monitoring in your area. As to in-rush currents, see what Font comes up with through his investigations and that should reduce/eliminate completely this nuisance-factor to the detector circuits and, of course, still allow the equipment to be used outside the IPS area. If you choose a low in-rush type of transformer this will equally prevent CBs potentially tripping in non-isolated areas too in which you might be using your display. Could be a case of “killing two birds (these two issues) with the one stone (ie the transformer)”

Now to what value the “leakage-setting” control (for your sensors) should be set to (as I am assuming it is this setting that you are concerned about and not a setting for the in-rush detection circuits – not that I would be expecting to hear that there was one for this!). Well if the manufacturer has provided you with no information for guidance then the conventional IPS leakage trip is generally set to trip around 2 to 5 milliamps. In lieu of any other value, you can quite happily use this. If your setting-control has an ohms scale then this current-equivalent value would equate (for the low current-trip point of 2mA) to a reading of 115 Kilo-ohms with your mains voltage of 230, and an ohm’s value of 46K for the higher 5mA trip-point For anyone with a mains voltage of 115 volts, the corresponding current readings are halved, of course

With regards to how to set it up, “no worries mate”. Just set the trip to a value within the above range (normally a starting point of 2mA /115K will be fine). As to equipment being connected to wall-sockets in the area when you set this up, you do not need to take this into account at all. However, just remember that the way an IPS works is that the more equipment that is added to the system (ie plugged into the wall sockets in the area at any one time, the lower the overall impedance of the system will be (reduction in impedance being provided by the aggregate of all the equipments resistances and reactances to earth). This means that if, in the rare event, you have so much equipment in the room that the lower trip-point (2mA/115K) is almost reached, when you add another piece of equipment the alarm may sound. Temporarily turn up the control setting to compensate. Obviously if turning it up does not cancel the alarm then you know you have a real over-load from a piece of equipment or just too much equipment connected in that area. Note too that this detection circuit is not protection against our “earth leakage currents”. An IPS is not capable of providing such protection.

This is a very interesting thread to have just read as we are experiencing very similar problems with a famous manufacturer of camera stacks.
On this particular stack it has a new flat screen monitor and the usual items of equipment that the surgeons always complain about. In our hospital this stack is being moved from one theatre to another and is the common link in causing the 32A current trip to trip! It seems to happen predominatley first thing in the morning, but has been known to fail at other times. The stack has a isolating transformer with an in-rush PCB and when we have tested the stack it only pulls 2.5Amps on swith on when the fault occurs. Our test kit is not good enough to see the raising edge of the current spike we think is the problem at switch on. What we are attempting to do is by-pass the manufacurers isolation on the base of the stack and introduce our own RS isolation trans while we await the manufacturer to fix this fault.
Is the problem caused when the system is cold first thing in the morning and the isolation trans acts like a short cicuit, hence the current spike or is it the in-rush PCB?

Double D.

Yes, this is indeed an interesting thread, and worthy of resurrection, surely? Especially as it ties in with recent discussions about theatre isolated power systems, isolation transformers on theatre stack carts, and other such intriguing matters (including, unfortunately, cases of people still using mains extension cables in operating theatres).

Hopefully, DD has things squared away by this time, and so perhaps we shall hear about the final resolution of the problem.

For timid souls who may shy away from Nick's "home-made" approach to LIM testers ... ready made versions are available!

New link.

Other LIM testers may be (and indeed are) available!

If you are serious about testing medical isolated supplies and their associated LIM then take a look at the Metrel MI 3110 http://www.metrel.si/products/electrical...est-im-new.html

There's a lot of nice test kit about these days.

Today's young biomeds don't know how lucky they are!