【讨论】单向流风速测定讨论

zhulikou431

【说明】一下内容来自ISPE论坛转载，说明一下。
此问题有一定意思，放在这里，希望大家深入讨论。
因为涉及多个帖子，恕不一一注明原帖作者。
---------------------------------------------------------------------------------------------
Topic: Air velocity measurement for an unidirectional airflow device
主题：单向流装置的风速测定
by Mr. Hyeok Je Kwon - 10 November 2010 at 4:10 AM ET
I have to perform operational qualification for an unidirectional airflow device (formerly referred as laminar work bench). Within our organization, some controversial exist.
我必须进行一个单向流装置（以前称层流工作台）的运行确认（OQ）。在我们公司，对此有一些争议。
Some people said the airflow velocity must be measured at working position AND 6 inches from the filter face. The guidance value (0.36 ~ 0.45 m/s) must be maintained at the working position, not filter face. The data from the filter face is just for informational purpose. In my humble guess, if the guidance velocity is achieved at the working position the velocity from 6 inches from the filter face will be much higher, which will create negative impact on the airflow pattern at the working position due to high supplying air velocity. But not sure. Eventually airflow pattern test will tell me the everything for this approach. Also there is a chance that high face velocity will result in the leaks in HEPA filter media and frame.
有人说，风速必须在工作位置和离过滤器表面6英寸的位置测定。工作位置的风速必须维持在指导值（0.36～0.54 m/s），而不是过滤器表面。过滤器表面的数据仅用作参考信息。在我看来，如果工作位置达到了风速指导值，那么离过滤器表面6英寸的风速将会高很多，这将会对工作位置的流型造成不利影响，因为送风速率很高。但不是很确定。最终，流型测试将告诉我这个方法的一切。高表面风速还有可能会造成HEPA过滤器介质和边框泄漏。
But others believe that the guidance value must be maintained at 6 inches away from the filter face. This is historically used for many years in the pharmaceutical industry. But in order to comply GMP, the velocity data at the working position is also important. As you know the measurement of the velocity at the working position would be highly variable due to the equipment size and configuration within the unidirectional airflow device. Proper airflow pattern at the working position is more important than achieving the specified airflow velocity at the working position. So velocity measurement at the working position will be information purpose but can be helpful in understanding the observed airflow pattern.
但其他人相信，离过滤器表面6英寸的位置必须维持在风速指导值。这在制药行业已经运用多年了。但是为了符合GMP，工作位置的风速也很重要。正如你们所知，根据单向流装置内的设备大小和布局不同，工作位置的风速测定值可能变化很大。工作位置适当的流型比达到其特定的风速更重要。因此，工作位置测得的风速虽用作参考信息，但对于理解观察到的流型很有帮助。
I’m very frustrated which approach is right interpretation. Especially, this plant should comply EU GMP, rather than US FDA. EU GMP Annex 1 states that “Laminar air flow systems should provide a homogeneous air speed in a range of 0.36 ~ 0.54 m/s (guidance value) at the working position in open clean room applications”. For EU GMP compliance, the former interpretation will be correct. But practical and historic point of view, the latter interpretation will be correct.
我非常困惑哪种理解是正确的。尤其是，这个工厂需要符合EU GMP，而不是US FDA。EU GMP附录1写到“在开放式的洁净室的工作位置上，层流系统应该能提供风速为0.36-0.54 m/s（指导值）的均匀气流”。对于EU GMP的符合性，前面的理解是正确的。但是从实际和经验来看，所者的理解是正确的。
Does anyone have the experience on this case for complying EU GMP? Also I’d like to listen the current industry interpretation (practice) for US FDA compliance.
有人在这个问题上有符合EU GMP的经验吗？我也想听听现在符合US FDA的行业理解（做法）。

zhulikou431

by Vincent Lapierre - 10 November 2010 at 3:11 PM ET
The usual method for testing HEPA filters velocity is as per IEST-RP-CC002.3. However this well accepted test method latest edition is still specifying to "measure the airflow velocity in a plane parallel to and 12in, or as otherwise specified, downstream from the HEPA or ULPA filters.".
通常测定HEPA过滤器风速的方法参考的是IETS-RP-CC002.3。但是，其最新版中对测试方法的描述仍然是“在一个平行于HEPA或ULPA过滤器表面的下游平面上测定气流速率，距离过滤器表面12英寸，或其它指定值。”
Now, EU GMP states the airflow measurements mus be made at the working position as stated in the Annex 1 or as well in the PIC/S Annex 1.
现在，EU GMP附录1，以及PIC/S附录1都表述为，风速测定必须在工作位置上进行。
for FDA, you can refer to FDA document "Guidance for Industry - Sterile Drug Products Produced by Aseptic Processing - CGMP" in section IV where its says:" The velocity parameters established for each processing line should be justified and appropriate to maintain unidirectional airflow and air quality under dynamic conditions within the critical area - note 5: a velocity of 0.45m/s +/- 20% has generally been established".
对于FDA，你可以参考FDA文件“行业指南——无菌工艺生产的无菌药品——cGMP”，第四节：每个工艺中对风速的要求都应该有理有据，并能在动态条件下适当地保持关键区域的单向流和空气质量。注5：已以建立了0.45±20% m/s的风速。
For Health Canada (where I am !) GMP, regulation C.02.029: "Unidirectional airflow systems provide an homogeneous air speed of 0.45 m/s +/- 20% (guidance value)at the working position in open clean room applications.
对于加拿大的医药行业（我所在的），GMP法规C.02.029：“在开放式的房间的工作位置，单向流系统提供风速为0.45±20% m/s的均匀气流。”
In my opinion, all GMPs are about the same requirement: adequate airflow demonstration at working position. They do not care on 6 in from the HEPA filter, now they focused their attention on the critical zone in the areas A (product exposed).
在我看来，所有的GMP差不多都是一样的要求：在工作位置有足够的气流保护证明。他们不关心离HEPA过滤器6英寸的位置，现在他们关注A级内的关键区域（产品暴露）。

zhulikou431

by Mr. Todor Draganov Massev - 11 November 2010 at 4:50 AM ET
你好Hyeok，
In the unidirectional, laminar or whatever you call the normal down-flow booths the quantity of air flowing downwards remains unchanged in the track between the supplied HEPA filtered air distribution and the working plane, which is above any confinement - equipment, personnel hands etc. As the cross-section surface of the enclosure remains unchanged within the above mentioned track, the velocities above and below should be the same. This is according to the laws of physics. (Of course, this is true at the low air velocities in the booth - fraction of m/s, at which the air can safely be considered uncompressible fluid.) If you measure the velocity above and below and the results prove to be different, you should look for abnormalities - e.g. 'unauthorized' air intake or spillage.
在单向流、层流或其它称谓的垂直气流布室，向下气流的质量在从HEPA过滤器送风口到工作台面整个区域都保持不变，并经受任何干扰——设备、人员的手等。由于在以上提到的区域内的交叉区域表面保持不变，上面和下面的风速应该是一样的。这依据的是物理原理。（当然，在布室内风速较低时是这样的，以m/s的形式，在此风速下空气可以认为是不可压缩液体。如果你测得的上面和下面的风速不一样，你就应该看看有没有什么异常，如“未授权”的进气或漏气。
Of course, in order to ensure containment or product protection you have to 'manipulate' the air flow: in the first case discharging to the outside some quantity of the air circulationg through the down-flow booth air system in order to ensure underpressure in the enclosure and thus air intake in the enclosure from outside or adding some air to the circulating air stream in the second case in order to create some overpressure in the enclosure). But all these 'manipulations' are performed before the air stream enters the unidirectional/laminar section (called 'enclosure'/'track' above). So, the text of the first paragraph applies.
当然，为了确保密闭或产品保护，你必须控制气流：第一种方式，从垂直气流布室排出一定量的空气，以确保布室内的负压，因此布室内的空气从外部吸入；或者第二种方式，给循环气流加入一定的空气，以在布室内形成一定的正压。但是所有这些控制都是在气流进入层流/单向流区域之前进行的。因此，适用于第一段文字的描述。
As for the place of air velocity measuring, comply with the recommendation of the party which you consider more powerful/important.
至于风速测定的位置，参考你认为更权威或重要的组织给出的建议。




by Mr. Carsten Rasmussen, Head of Department - 11 November 2010 at 9:47 AM ET
For multipurpose UDF equipment measurement of flow near the outlet filter face is quite relevant as this is basic quality parameter, giving information of the UDF ability to provide clean condition, with very little predetermined information on the obstacles and disturbances that will take place during actual work. If however the UDF is above equipment or subject to standard working procedures, I believe focus should be shifted from, air velocity measurements to analysis of flow condition around the product, that is to be protected. In my opinion that traditionally there has been too much focus on filter face velocities and too little on establishing efficient flows where it actually matters. Much too often I have observed equipment with poor aerodynamic properties being placed in UDF flow. Equipment that obstruct flow, created stagnation and other disturbances of flow, and hence is able to obstruct even the most uniform flow. Disturbance of flow is likely to reduce flushing efficiency of the flow and even by creating vortices is able to make paths for contaminant flow to the product. The return air flow pattern og UDF is in my opinion another parameter not receiving enough attention.
对于多用途单向流设备，过滤器出口表面的风速测定值是相关的，因为这是个基本的质量参数，用于评估单向流设备提供洁净环境的能力，但关于实际工作中使用的物体和干扰几乎没有预见性的信息。然而，如果单向流设备是吊顶设备或有标准操作规程，我相信应从关注风速测定转变为关注需要保护的产品周围的气流条件。在我看来，传统上太过于关注过滤器表面的风速，而在真正重要的地方建立有效的气流却没怎么在意。经常我都看到单向流中安装有空气动力学很差的设备。破坏气流、造成静滞和其它气流干扰的设备，甚至能够干扰最均匀的气流。气流扰动有可能降低气流吹洗的效率，甚至通过形成涡漩造成污染产品的气流通道。在我看来，单向流设备中气流的回流流型是另一个没有受到足够关注的参数。
Improving on present use of UDF technology requires a holistic approach during design. Personnel, equipment, heat sources should be taken into consideration. Using CFD is at present a very mature technology and I fail to understand why this is not mandatory to use this technology when designing UDF system in the Pharmaceutical Industry.
改善目前单向流技术的应用，需要在设计中采用全局式的方法。人员、设备、热源都应纳入考虑。使用计算流体力学（CFD）目前是非常成熟的技术，我没有理解到，制药行业怎么没有在设计单向流系统中强制要求使用这个技术。
Finally I dont understand why the thinking behind the concept of perforated floors cannot be used in Pharmaceutical Industry, that could be a way of really improving flow under UDF.
最后，我没有理解到多孔地面的理念背后的想法为什么不能运用到制药行业，那可以是一个强化单向流设备下方气流的方式。





by Mr. Kishor D. Datar - 20 November 2010 at 4:58 AM ET
各位，
The issue is really difficult & I can understand the confusion. The reason for such confusion is the guidance value of velocities given by all regulatory agencies & is historical for the Laminar Air Flow Pattern.
这个问题真是复杂，我也能理解大家的困惑。产生困惑的原因，就是监管机构给出的风速指导值，也是层流气流的传统做法。
This is why historically it was referred as Laminar Flow Station. However if analyzed technically it is not feasible to maintain the air velocity such that the flow will remain laminar at filter face & at work level. When this was understood the name got changed from Laminar Flow to Unidirectional Flow.
这就是为什么过去称之为层流站。然而，如果从技术上分析，维持风速使流型在过滤器表面和工作台面保持层流是不可行的。当这一点被理解之后，称谓就从层流变为了单向流。
However the historical indicative values are same as were in earlier guidelines (Only change is from FPS system to Metric System)
然而，传统指示值与早期指南中的一样（仅仅是从英尺/秒变成了公制单位）。
Coming to your issue, if you want both the things, forget it, not feasible technically (With whatever current technology available)
I suggest to use following methodology, which is widely acceptable,
1. Ensure that filter face velocities are within limits prescribed (0.45 m/sec +/- 20%)
2. Check the velocities at working level at various points (Don’t worry they are going to be different )
3. Check whether the unnecessary constraints are being created by any equipment, if so remove them
4. Check again the velocities mentioned in point 2 above.
5. Carry out air flow pattern using smoke test
6. Ensure that filming is done during such air flow pattern
7. Carry out validation (media fill) as per your norms
8. Check the results
9. If they are within acceptable limits,
10. Conclude stating that as long as air flow at filter face are what verified & air flow pattern is what is being filmed, the entire system including UDF/Equipment below are validated.
11. Any change in any component may result in change in air flow pattern entire study will need to be repeated.
至于你的问题，如果你两者都想要，还是算了吧，从技术上不可行（无论使用什么当前可用的技术）。
我建议使用以下广泛接受的方法：
1.确保过滤器表面风速在限度内（0.45±20% m/s）
2.检查工作台面不同点的风速（不要担心他们会不同）
3.检查是否有设备形成了不必要的障碍，如果有就清除。
4.再次检查上面第2点中提到的风速。
5.用发烟试验测气流流型。
6.确保在流型测试中进行了摄像。
7.按你们的标准进行验证（模拟灌装）。
8.检查结果。
9.如果在接受限度内，
10.得出结论，只要过滤器表面的风速在验证的范围内，并且流型与摄像记录一致，整个系统，包括单向流装置和其下的设备，就是经过验证的。
11.任何组件的变更可能都会造成气流流型的改变，因此需要重新进行一次完整的检查。
This approach is better till we have some new invention to resolve both the logics.
直到我们新的方法来解决这个两个矛盾前，这个方法是最好的。

zhulikou431

by Mr. Mohmmad Isak Mohyuddin Shaikh, Head Engineering - 20 November 2010 at 12:03 PM ET
各位，
Kishore is absolutely right. I agree with him not only in principle but tried & tested that you cannot have it at both the filter face & working height. Be careful in many instances if you increase the blower speed with a target working height velocity (as desired )than most of the time the flow turns very turbulent. Infact there is guaranteed assurance that it wont be a UDF anymore. Thanks Kishore for a more pragmatic understanding of the guidlines.
Kishore说得完全正确。我同意他的观点，不仅在原则上，也试验和测试过，你不可能同时在过滤器表面和工作高度都达到要求。在许多情况下都要小心，如果你按照工作高度的目标风速（需要的）来增加风机的速度，超出了平常的范围，气流就会变得非常扰动。实际上，可以肯定这时已经不再是单向流了。感谢Kishore对指南做出了一个更为实际的理解。





by Mr. Scott E. Mackler - 21 November 2010 at 8:07 AM ET
The 6" criteria is provided in order that when testing for filter face uniformity (and secondarily filter to filter uniformity) the correct standoff distance is used. 6" below the filter face is the correct distance for the instrument in order to minimize any localized effects of velocity striations (often times we see users measuring too close to the filter face and thereby generating incorrect readings). The comments in the thread are correct however, in that once you have established that the filter face (from 6" below) is performing at the specified velocity (whatever value that may be based on requirements) and with the appropriate / correct uniformity, then it is the velocity and the unidirectional flow pattern performance at the work surfcae (point of interest) that truly matters. Filter face measurements are simply a means to that end. I have found that different filters, from different manufacturers, do not perform the same at 6" downstream of the filter face - some provide excellent uniformity, others do not. Those that do not will obviously not support the unidirectional flow requirement at the work surface. Same goes for laminar flow benches, some designs (push / pull) work far better than others!
给出6英寸的标准是为了在测试过滤器表面均一性（以及过滤器与过滤器之间的均一性）的时候，采用了了正确的距离。过滤器表面下方6英寸是仪器的正确距离，以减少风速的任何局部效应（通常我们看到用户在离过滤器表面过近的距离测定，因此得出了错误的读数）。这些评论在思路上是正确的，然而，一旦你确定了在过滤器表面（下方6英寸）按指定风速（根据你的要求不论是什么值）进行测定，并有适当/正确的均一性，那么真正重要的就是工作台面（关键点）的风速和单向流流型效果了。过滤器表面测定仅仅是达到这个目的的一个方法。我发现不同的过滤器，来自不同的生产商，不过滤器表面下游6英寸处的性能是不一样的——有的均一性很好，有的不是。那些不好的，显然不能在工作台面上达到单向流的要求。同样的道理也适用于层流工作台，有些设计（推/拉）比其它的效果好很多！







by Mr. Jozef A. Corver - 3 January 2011 at 2:49 AM ET
你好Kishor，
I sympathize with the evidence based verification method. I have two additional remarks:
- I would include a step before your first point: do a CFD analysis/calculation of the air flow pattern. In that way you identify the weak spots/areas where specific attention should be put.
- I should specify the acceptance criteria based upon the purpose of the UDF with the aid of the flow analysis results, the major goal being to effectively expell unspecified air and avoidance of particle-carrying air reaching the sterile region.
我赞同基于证据的验证方法。我还有两点评论：
-我会在你的第一点前面加入一个步骤：进行气流流型的CFD分析/计算。这样你就可以找到需要特别关注的较差的点/区域。
-我会根据单向流系统的用途和气流分析结果明确接受标准，主要目的是有效排出非洁净空气并避免携带微粒的空气到达无菌区域。
Once you have performed the series of activities: CFD analysis, flow measurements, smoke test, media fills, you also have a basis for (cost) effective redisign and optimization afterwards.
一旦你进行了这一系列活动： CFD分析、气流测定、发烟试验、模拟灌装，你就有了依据进行（经济）有效的再设计和优化。







这个话题很好，我们的上一个无菌制剂项目请专家来做过预检查，当时灌装机灌装加塞段安装了2个在线风速仪，一个在离过滤器散流板20cm左右处，一个在离工作台面10cm处。上面的风速仪显示速度为0.4几，下面只有0.2几，被他们提了一条层流罩风速未达标准，我们也找不出有依据的解释，只好把下面的风速仪拆掉。显然这不是正确的做法。
我刚刚看了一下ISPE那篇文章"Working Height Velocity Measurement in Conventional Cleanrooms"，提炼一下其中的主要内容，供大家参考，有指导意义。
首先，FDA的指南“sterile drug products produced by aseptic processing”中是这样要求的：
HEPA filter leak testing alone is insufficient to monitor filter performance. It is important to conduct periodic monitoring of filter attributes such as uniformity of velocity across the filter (and relative to adjacent filters). Variations in velocity can cause turbulence that increases the possibility of contamination. Velocities of unidirectional air should be measured 6 inches from the filter face and at a defined distance proximal to the work surface for HEPA filters in the critical area. Velocity monitoring at suitable intervals can provide useful data on the critical area in which aseptic processing is performed. The measurements should correlate to the velocity range established at the time of in situ air pattern analysis studies.
单独的HEPA过滤器泄漏测试不足以监测过滤器的性能。周期性地进行过滤器属性监测例如过滤器（及与相邻过滤器）的风速均一性是很重要的。风速的变化可以导致湍流，会增加污染的可能。单向流的风速应在据过滤器表面6英寸处测试，对于关键区域的HEPA过滤器还应在靠近工作表面的距离测试。以适当的间隔监测风速可以为无菌生产的关键区域提供有用的数据。这些测试应与现场气流流形分析研究时建立的风速范围相关联。
欧盟附录1中是这样要求的：
Grade A: The local zone for high risk operations, e.g., filling zone, stopper bowls, open ampoules and vials, making aseptic connections. Normally, such conditions are provided by a laminar air flow work station. Laminar air flow systems should provide a homogeneous air speed in a range of 0.36 to 0.54 m/s (guidance value) at the working
position in open cleanroom applications.
A级：高风险操作的区域，如，灌装区域、胶塞斗、开放的安瓿和西林瓶、无菌组装区。通常，这样的条件由层流工作站提供。层流系统应在开放式洁净室应用的工作位置提供均一的气流速度，范围0.36-0.54M/S。
制药工厂的检查缺陷中关于工作位置风速（WHV）有这样的条目：
1. “In parenteral products manufacturing, the air velocity testing of HEPA filters in critical, (Class 100) areas, is done 4 to 6 inches from the filter face and not at the critical working level where sterile product is open to the environment.”
在注射剂产品生产中，关键区域（100级）的HEPA过滤器的风速测试是在距过滤器表面4-6英寸的地方进行的，没有在关键的工作平面上进行，而此处有无菌产品暴露于环境。
2. “The firm has not performed any studies under dynamic conditions to show there is a correlation between air velocity at the filter face and velocity at the critical working area.”
该公司没有在动态条件下进行任何研究以表明过滤器表面的风速和关键工作区域的风速有相关性。
行业情况：
从2005年6月至2006年2月，进行了一个行业的基线研究，包含了8个制药公司以及3个洁净室。这些基线结果揭示：
1.大部分制药公司没有测试WHV。某些公司在首次确认时测试了WHV，但几乎没有进行日常的再测试。
2.当测试WHV数据时，没有建立可接受标准。
3.经验表明“WHV不是可以可靠地重现的东西”
4.“金标准”仍然是与过滤器表面风速相关联的气流模型测试。过滤器风速进行日常测试，至少6个月一次。当发生重大变更，重复气流模型测试，或者按照日常计划，从每年到每3年不等。
作者对此进行了试验研究，在层流罩下进行测试。确定日常的WHV测试位置和接受标准需要考虑以下因素：
1.微生物的分散形式（如，人员传播、物料传播、气流）
2.实际生产中的微生物污染的可能性
3.影响产品质量的可能性
4.监测能够可重现地取样的位置
对每一个关键点位置进行技术研究以确定WHV限度的上限和下限。进行研究时必须确保：
1.测试使用热球式风速仪
2.测试在层流罩设置为低、中、高风速时进行。风速应基于由气流模型测试建立的接受限度。
3.测试之前，HEPA过滤器表面风速应在限度范围内
4.服务于该区域的HVAC系统及压差应正常工作
5.测试包括在关键点取10个风速读数，以获得合理的样本以解释测量变化
6.取样，包括每个测试运行之间的设备设置，应重复3次以解释设备设置和布置的任何变化
7.测试点位于工作表面12英寸处（30.5cm）
8.所有用于确定接受标准的数据值都反应了代表可接受的气流模型的条件。
最后，直接上作者的结论：
1.风速在相对很小的位置变化时可以发生显著变化。特别是使用热球式风速仪。
2.在HEPA过滤器散流板下的风速测量值与工作高度的风速测量值相比，是更好的层流罩性能指标，因为工作高度的风速标准偏差和变化更大。与工作高度风速与气流模型之间的相关性相比，HEPA过滤器表面风速与气流模型测试之间的相关性更可靠、更可重复。
3.热球式风速仪可以用于在复杂气流环境中获取详细的多方向的风速变化。也可以提供能与CFD模型进行比较的数据。

zhulikou431

http://blog.ispe.org/forums/showthread.php?t=389

冰城

好乱啊，呵呵，这个风速在国内肯定还是按照指导值去做，单向流用气雾发生器去做会更有说服力。

ZL221

学习了，谢谢。

Heavenlo

是啊，这个帖子在原ISPE论坛上看上去排版要规整一些，跟论坛的版面设置有关吧。
Henry

snow200289

学习了     

xqliu

学习一下啦，谢谢提供分享，很不错的。

jacky-zhao

感谢分享
学习一下

bencao89

学习了，归纳一下自己的理解，不知有无偏差，肯请LZ或者其他专家点评一下吧：
1.3.6~5.4这个指导标准是针对出风口下6英寸的检测值；
2.只检测出风口的风速还不够，还必须要监测操作面的风速，但操作面的风速标准不一定是3.6~5.4，它需要各企业自己根据自己的操作空间布局、设备安装情况，人员操作方式等相关因素通过实验数据在确定；
3.确定操作面风速标准的原则，至少要满足以下条件：出风口下6英寸处风速在3.6~5.4；操作空间的气流型态符合单向流的要求。

lumang

谢谢分享！   

vincentchampion

FDA指南里面也有说道风速为 0.36~0.54 距离工作点 1英尺

scslk0045

认真学习

leelsy

认真学习

litao8702

好帖，必须赞一个，GMP需要这样的争论和钻研，。，

爱生活1

认真学习，正在查阅相关内容，很好。

zhaosc8326

很好的帖子。赞

十八有点呆

zhulikou431 发表于 2014-5-5 12:14
by Vincent Lapierre - 10 November 2010 at 3:11 PM ET
The usual method for testing HEPA filters velo ...

谢谢