Cleanroom positive pressure

[AMS]

Good Morning all, I was wondering if anyone knows how to improve the positive pressure inside of the cleanroom. HEPAS are new and setup to a medium speed and the pressure is nearly positive. Is there any recommendations? Thanks to all. AM

[waynem]

Could you elaborate. Is this a cubicle or huge passage or warehouse? What are the pressures that you are trying to attain? What have you tried and what has caused this system to go out of spec.. just a filter change??

[AMS]

waynem, Thanks so much for your interest in the question. The clean room/cubicle in shape is setup for positive pressure (I do not have the information here in front of me but I think it is reading 0.07 inches WC). What I am looking for is at least .5. I thought that by changing the HEPA filters (They were like 7 years old) that I will get to increase it but it did not work. The readings prior to the change were similar. I have calculated that the minimum HEPA filters required for the square footage is 9 HEPA filters and I have 12 in service (active). Also I calculated the returns + the exhaust system I have + the air intake and yeah I know I am not introducing as much air as I am taking out but other than increasing the HEPA filter #’s I am not sure what else to do. Maybe reducing the returns (meaning lowering the returns walls) so the air going out to the plenum is reduced… I am not sure what to do, is there anybody that can help me please.
Thanks so much in advance.

[meyert]

You could do a smoke study and figure out where your air is going. Sounds like the room is leaking somewhere. Check the door sweeps, or maybe lower the pressure in the surrounding rooms.

[AMS]

Meyert, Unfortunately the clean room is 100% functional (production is in full force) and certified for class 6. At this point I can't (you know potential contamination will be an issue). I did check all around the areas and plug some holes and disconnect some air returns connected to the office (do not ask me why that was). Unfortunately, it only increased from 0.05 to 0.07. I also, add a door skirt to prevent for any leaking… thrust me I am really lost in this one. Is there any logic to reduce the air returns volume? As you can tell I have never done this kind of job, my specialty is in something totally different…

How can I reduce the pressure in the office areas?
Thanks

[waynem]

Well AMS, it seems to me you might have to start from the bottom and work your way up again. The changing of Filters is not really going to help unless the pressure difference across the filter has passed the 600 Pa mark, as the system should automatically adjust itself to keep the correct airflow. I would say, measure the outlet of your AHU as close to the unit as possible, measure all the outlets and compare. Should be as close as dammi*. Otherwise you have leakage. The measure the return air as close to the unit as possible, measure all the return air points and compare. Should be as close as dammi* aswell. Otherwise you have leakage. With that sorted out, what contolls your system?? Can the fan handle more air?? If yes, the have a look at the control setpoints, i.e what gets the fan to increase in speed, is it a velocity or pressure transducer and what are the setpoints? Maybe that there is still room to increase the fan motor speed without taking the fan off the fan curve.

Hope this is of some help.

[AMS]

It does makes a lot of sense. Thanks so much. Now I am almost positive that the fan can't handle more air - not leaks detected otherwise.

[RobinE]

Funny you should ask this when I just had a conversation with our HVAC contractor. He noted that most of the Big Pharma in my area (Philadelphia) that he serves have their clean rooms pressurized to 0.05" (water) - too much more will start doing damage to the building, believe it or not. I'm sure you're aware of the cascading effect you should be setting up with the highest 'positivity' in the most critically clean areas, stepping down to a 'cleanish' zone, etc. This might be a critical area of ISO 2 being the most positive with respect to its ante-chamber at ISO 3, which would in turn be positive to the hall, etc. Hope this helps.

[Dennis]

Maybe you don't have a leak but an obstruction somewhere in de air handling installation. If the resistance is bigger than your AHU can gif the curve of your ventilator/motor is at his top en this result in less air. So, mesure the pressure inside your canals. Or is there air leaking through sockets?
Gr Dennis

[AMS]

To all,
It is not that I am ignoring all options -sorry it is taking me longer to check than I thought. At this point I have not found any obstructions and/or leaks. Any other ideas... I will hate to have a new expense if I do not have to.
Robin I am not sure about what you are telling me, by doing some research for class 6 cleanrooms--- it supposed to be around 0.1 - maybe I am wrong.

[Shahnawaz]

To all,
It is not that I am ignoring all options -sorry it is taking me longer to check than I thought. At this point I have not found any obstructions and/or leaks. Any other ideas... I will hate to have a new expense if I do not have to.
Robin I am not sure about what you are telling me, by doing some research for class 6 cleanrooms--- it supposed to be around 0.1 - maybe I am wrong.

Dear AMS
Please let m know the details of said room and their adjacent room data sheets.

Thanks

[AMS]

Shahnawaz,
I am sorry I am not quite getting it…what is the information you need?
Thanks AMS

[joey]

Have your problem been solved if not reply

[greymass2003@yahoo.com]

i had a similar problem at a company in PR about 28 years ago...here we had to refit every door, passageway window and electrical outlets etc. Most of the air was leaking thru the interlocks and doors. Try closing the returns completely while taking measurements; monitor the pressure while having someone take readings every 10 to 15 minutes until the pressure builds-up. Also, perform a trace test with smoke sticks (colored smoke) to see if possible leaks exist...

[wiston]

Dear:
I worked for one enginering company, as far as I know, the postive pressure is kept throught the difference between air supply and air exhausted or returned or both, besides, the air leakage in the seam of partition, ceiling window and door shall be considered, but I don't think there is a lot of relationship with the HEPA, becuase as long as the HEPA aren't be clogged, it is no need to change the HEPA.

in my opinion, solution should be: 1, increase the air supplied by increase the capacity of the blower.
2, decrease the air exhausted or returned or both, but the precondition is that the cleanliness of this room shall be ensured, because this meausre will decrease the air change rate of the air, and the leaked air can not take the particle away.

But you can try, if the positive pressure can not be maintained when you take the measure of item 2 by 2-3 times of the cubic of the room, absolutely the leakage of air will be responsible for the pressure. in this case, you have to change some critical fixtures, like the door (priority), window etc.

Also what is the situation of the adjacent rooms, what about their pressure difference?

Best Regards
Wiston

[Alfred]

waynem, Thanks so much for your interest in the question. The clean room/cubicle in shape is setup for positive pressure (I do not have the information here in front of me but I think it is reading 0.07 inches WC). What I am looking for is at least .5. I thought that by changing the HEPA filters (They were like 7 years old) that I will get to increase it but it did not work. The readings prior to the change were similar. I have calculated that the minimum HEPA filters required for the square footage is 9 HEPA filters and I have 12 in service (active). Also I calculated the returns + the exhaust system I have + the air intake and yeah I know I am not introducing as much air as I am taking out but other than increasing the HEPA filter #’s I am not sure what else to do. Maybe reducing the returns (meaning lowering the returns walls) so the air going out to the plenum is reduced… I am not sure what to do, is there anybody that can help me please.
Thanks so much in advance.

Dear AMS,

where does your specification of 0,5 inches WC come from? 0,5 inches WC is equivalent to 127 Pa, and the usual pressure difference between differently classified cleanrooms lies between 15 and 25 Pa (or 0,06 - 0,1 inWC), which is perfectly in line with the figures you found. With 0,5 inWC you will probably get your ceiling blown up. To find out the maximum pressure your room can reach, cover the returns grids with ordinary polyethylene bags (the positive pressure will make them seal perfectly), so that the entire inlet air must find its way through leaks, door slits, passthroughs of utility ducts, etc. The pressure achieved is the maximum your system can achieve in its present state, and it is easier to spot leaks by this way.

Regards

Alfred

[ValGuy01]

If you are following ISPE recommendations for a sterile fill facility you only need 10-15Pa (0.04"wc - 0.06"wc) between different room classifications (i.e. ISO 8 to ISO 7).

However, here is an even better question that I need some help with. We have three cascading rooms that have differentials pressures set at 0.05"wc between each room. Each room's differential pressure gauge is measuring between atmospheric and the room, such that the analog gauges read like so:

Room 1 (0.05"wc) <----- Room 2 (0.10"wc) <------Room 3 (0.15"wc)

We do not have a microprocessor controlled system and the rooms are under one HVAC unit. The room pressures are controlled by mechanical dampers. The arrows above indicate the direction of airflow. As I stated above the ISPE guideline states that the rooms should have between 0.04"wc - 0.06"wc between rooms. Because the gauges are reading to atmospheric, all the gauges vary slightly (perhaps +/-0.01"wc) due to atmospheric conditions, room heating and cooling, etc.

The million dollar question is: At what setpoints should I set each individual room's differential pressure Hi and Lo alarm set-point? QA requires a minimum of 0.04"wc between rooms at all times, engineering/testing has shown anything within 0.01"wc of the setpoint will give nuisance alarms (we have a 2 minute delay on the alarm), we do not have continuous monitoring of values, and we need to know when any room is below specification (0.04"wc between rooms).

I can provide more information / clarity, but any immediate best practice out there is definitely appreciated.

[alexkennedy]

First a word of warning - playing around with a live active HVAC system can be dangerous – high pressure will blow out windows and sometimes doors – low pressure can cause the whole structure to collapse inwards with disastrous results for occupiers. This has happened at several pharma plants.

It is almost standard practice to measure all the room pressures relative to outside ambient. Your room differential should all be 15pa plus or minus 3. You will go into alarm when doors are opened, however it is pretty standard to fit delays so that these transients are ignored.

Three rooms in cascade, I hope you don’t mean the same air enter one room and leaves that room to enter another. That would raise quite a problem about the quality of the air.

You don’t know whether the pressure drop over the filters is low because of low flow in the system, or flow leaving the system.

Measure the flow, there are excellent instruments for the job – remove one of the plugs (they are designed into the system for carrying out these tests) in the trunking and insert a sensor tube for a flow-meter. Measure the flow as it leaves the fan, compare this to the drawing specification (since it is pharmaceutical it must be validated so there will be drawings available.
If the flow is Low;-

Check where the AHU draws its air from, it is common to recirculate a percentage of the air. If recirculation is used check the recirculation air damper settings.
No luck there then:-
Go to the air duct inlet filter or screen, there has to be a filter there to stop rodents and birds entering the system.
It will not be a HEPA filter it might be a fine mesh screen, but they can get blocked with debris especially foliage and or ice.
Check the bag inlet filters actually on the AHU are clear.

Satisfied the AHU is not getting starved of air, check the bag filter inlet filters actually on the AHU are clear – progress to the AHU fan speed controller.
There are several ways a design engineer will control the air flow in a system such as yours.

Fit an attenuator (silencer) into the system. The air flow through the attenuator causes a small pressure drop across the attenuator. A sensor is used to measure this pressure drop. A certain flow will have a certain pressure drop.

The main AHU fan motor is powered from a solid state inverter. The sensor through a controller can automatically raise or reduce the frequency from the inverter - which (since motor speed is a function of supply voltage frequency) increases or reduces the fan speed.

Now as the system ages and the HEPA filters progressively clog up – the flow in the system is reduced – the reduced flow is sensed by the attenuator sensor, which causes a signal to be sent to the fan motor speed controlled, which increase the fan speed to reinstate the flow.

Another way is to have the sensor and the attenuator, but have a motorised airflow control valve. So in this case the when the sensor senses the HEPA filter are reducing the flow - it opens the air flow controller valve slightly and allows more pressurized fan air into the system.

Obviously any of a series of faults in this system can cause low fan speed. So you must test this system to ensure it is not the cause of your problem.

If you still have trouble gives us an email and we will take it further.

Alex Kennedy
www.validation-online.net

[Sambhu]

Hellow everybody,
I am very much excited with the conversation on this specific topic. Let me take participate with the group.
The process of achieving required pressure within cleanroom associated with adjucent areas are called pressure balancing. Its a pure engineering work.
Theoretically the pressure differencial is caused due to bleeding of excess air from a specific room to its adjucent rooms through clearances.
In an equilibrium condition, the incomming air (X) entering in the cleanroom goes out through raisers (Y) & the additional amount of air (X-Y) is subjected get bleeded out through clearances. This X-Y amount of air is resposible for pressure diffrencial.
So, if we want to alter the pressure differencial, all we need to increase/decrease the bleed air quantity. This can be done by following ways:

(1) Increase the cfm of incomming air (enhancement of blower capacity).
(2) Reduce the return air damper opening.
(3) Reduce the clearances (e.g. door clearances, plugging the leakages, etc.)
(4) Reduce the pressure diffrencials of the adjucent rooms by the one of the above methods.
But one thing we should keep in mind that in 1st case, changing the inlet air cfm means significant change in room air changes which is not desirable due to design considerations.
So, the later two options are more handy.
Hope I am able to make some sense, plz reply.

[vijaypedineni]

Dear AMS
Please let m know the details of said room and their adjacent room data sheets.

Thanks

Mr. Shanwaz,

I have a doubt reg. differential pressures,
Pl correct my statement, clean room differential pressures should be individual and should be compared with atmospheric pressure and not with the adjacent lower class.

[Al-Pharao Air Data]

To All Valid


Can you tell us more about Air Flow Parallelism Test



Regards

A.Aziz
Al Pharao Air Data

[usazambia]

This problem may either be resolved by now, or else accepted "as is".
Some points to consider:
1) the 0.5 Pa value is absolutely too high; refer to the lower values suggested in the string.
2) "Positive Pressure" refers to the pressure in the room being higher "relative" to the pressure in the adjacent room(s). This has been correctly suggested by some in the string, where the descriptive phrase being used was "cascade". This means that your Target room should have a higher pressure relative to the room(s) next to it. Several methods have been mentioned on how to do this, while other methods mentioned pertain more to maintenance applied troubleshooting; not performance validation.
3) "Positive Pressure" does not refer to the INCOMING air being higher than the OUTGOING air. This refers to "System Pressure Drop". The general way Positive Pressure is achieved is for the Target room to have a higher air changeover per hour (ACH) than its adjacent rooms. So its air handling unit's flow rate (X cfm) divided by the room's volume (Ft3) gives the number of times the air changes over per hour.
ACH = X (ft3/min) / V (Ft3) * 60 (min/hour)

So cascade would look like this:

ACHlab > ACHgowning > ACHhallway

NOTE: Incoming = Exhaust - Leaks;
if using Incoming > Exhaust & Leaks, then the room would continue to build up pressure to intolerable and dangerous levels; this is what one commentor meant when he stated "the system will balance itself".

4) If you have leaks, then that is normal, and should be expected. The important thing is this: is the leakage at those areas in the Target room egressing, or ingressing? You want it to be egressing (exiting). In other words, if it were ingressing, then particulates/contaminants would be flowing from the adjacent room (like the gowning room) into the lab. This is precisely what "Positive Pressure" prevents (or minimizes).

[DURGA PRASAD]

Cleanroom pressures are actually measured by taking incosideration with adjescent room pressures.

If you look the EU guidance:

Pressure differential between rooms of different classes should be 10–15 Pa, and 12.5 Pa between a classified and nonclassified room.

[Shahnawaz]

The most widely accepted pressure differential for achieving containment between two adjacent zones is 15 Pa, but pressure differentials of between 5 Pa and 20 Pa may be acceptable. Where the design pressure differential is too low and tolerances are at opposite extremities, a flow reversal can take place. For example, where a control tolerance of ± 3 Pa is specifi ed, the implications of the upper and lower tolerances on containment
should be evaluated.
The pressure differential between adjacent rooms could be considered a critical parameter, depending on the outcome of risk analysis. The limits for the pressure differential between adjacent areas should be such that there is no risk of overlap, e.g. 5 Pa to 15 Pa in one room and 15 Pa to 30 Pa in an adjacent room, resulting in no pressure cascade, if the fi rst room is at the maximum tolerance and the second room is at the minimum tolerance.

Cleanroom pressures are actually measured by taking incosideration with adjescent room pressures.

If you look the EU guidance:

Pressure differential between rooms of different classes should be 10–15 Pa, and 12.5 Pa between a classified and nonclassified room.

[siddhipharma]

Mr. Shahnawaz,

Thanks for sharing & I do agree that the system should not be maintained at Lower & upper cascade range of pressure in adjescent rooms.

In addition to this few facts:
1. Minimum pressure must be maintained in adjuscent rooms to keep the cleanliness level maintained.
2. This minimum pressure either with respect to room or with respect to ATM.
3. Air pressure & door directions should be taken in consideration while designing the pressure philo..

Opinions.....

Happy Reading !

[Shahnawaz]

Mr. Shanwaz,

I have a doubt reg. differential pressures,
Pl correct my statement, clean room differential pressures should be individual and should be compared with atmospheric pressure and not with the adjacent lower class.

Dear Mr Vijay,
No, differentail pressure should be consider with adjacent area and also consider the pressure cascading from process core (clean area) to transition zone, then to Ancillary area and in the end outdoor environment.
I hope it will be cleared your query and doubts.
Regards,
Dr. Shahnawaz Sajid